From 877838f79111f91f11e681438840cfcce6c649fa Mon Sep 17 00:00:00 2001 From: Andrey Date: Thu, 10 Oct 2024 15:35:35 +0300 Subject: [PATCH] lec6 --- README.md | 2 + assets/eda/eda.ipynb | 317 +- assets/mlflow/.gitignore | 5 +- assets/mlflow/comment.txt | 4 +- assets/mlflow/requirements.txt | 5 +- assets/mlflow/research.ipynb | 7504 +++++++++++++++++++++++- lectures/lec6-feature_engineering.odp | Bin 0 -> 127631 bytes lectures/lec6-feature_engineering.pptx | Bin 0 -> 124084 bytes 8 files changed, 7562 insertions(+), 275 deletions(-) create mode 100644 lectures/lec6-feature_engineering.odp create mode 100644 lectures/lec6-feature_engineering.pptx diff --git a/README.md b/README.md index 16e2968..dcbadfc 100644 --- a/README.md +++ b/README.md @@ -9,6 +9,8 @@ | 19.09.2024 | [Разведочный анализ данных](./lectures/lec3-eda.odp) - [в формате pptx](./lectures/lec3-eda.pptx) | | 26.09.2024 | [MLFlow](./lectures/lec3-eda.odp) - [в формате pptx](./lectures/lec4-mlflow.pptx) | | 26.09.2024 | [MLFlow - практика](./assets/mlflow/research.ipynb) | +| 26.09.2024 | [Feature engineering](./lectures/lec6-feature_engineering.odp) - [в формате pptx](./lectures/lec6-feature_engineering.odp) - код в [ноутбуке к mlflow](./assets/mlflow/research.ipynb) | + ## Лабораторные работы diff --git a/assets/eda/eda.ipynb b/assets/eda/eda.ipynb index 752009a..47b3b5c 100644 --- a/assets/eda/eda.ipynb +++ b/assets/eda/eda.ipynb @@ -1281,7 +1281,320 @@ }, { "data": { - "application/javascript": "'use strict';\n(function(root) {\n function now() {\n return new Date();\n }\n\n const force = true;\n\n if (typeof root._bokeh_onload_callbacks === \"undefined\" || force === true) {\n root._bokeh_onload_callbacks = [];\n root._bokeh_is_loading = undefined;\n }\n\nconst JS_MIME_TYPE = 'application/javascript';\n const HTML_MIME_TYPE = 'text/html';\n const EXEC_MIME_TYPE = 'application/vnd.bokehjs_exec.v0+json';\n const CLASS_NAME = 'output_bokeh rendered_html';\n\n /**\n * Render data to the DOM node\n */\n function render(props, node) {\n const script = document.createElement(\"script\");\n node.appendChild(script);\n }\n\n /**\n * Handle when an output is cleared or removed\n */\n function handleClearOutput(event, handle) {\n function drop(id) {\n const view = Bokeh.index.get_by_id(id)\n if (view != null) {\n view.model.document.clear()\n Bokeh.index.delete(view)\n }\n }\n\n const cell = handle.cell;\n\n const id = cell.output_area._bokeh_element_id;\n const server_id = cell.output_area._bokeh_server_id;\n\n // Clean up Bokeh references\n if (id != null) {\n drop(id)\n }\n\n if (server_id !== undefined) {\n // Clean up Bokeh references\n const cmd_clean = \"from bokeh.io.state import curstate; print(curstate().uuid_to_server['\" + server_id + \"'].get_sessions()[0].document.roots[0]._id)\";\n cell.notebook.kernel.execute(cmd_clean, {\n iopub: {\n output: function(msg) {\n const id = msg.content.text.trim()\n drop(id)\n }\n }\n });\n // Destroy server and session\n const cmd_destroy = \"import bokeh.io.notebook as ion; ion.destroy_server('\" + server_id + \"')\";\n cell.notebook.kernel.execute(cmd_destroy);\n }\n }\n\n /**\n * Handle when a new output is added\n */\n function handleAddOutput(event, handle) {\n const output_area = handle.output_area;\n const output = handle.output;\n\n // limit handleAddOutput to display_data with EXEC_MIME_TYPE content only\n if ((output.output_type != \"display_data\") || (!Object.prototype.hasOwnProperty.call(output.data, EXEC_MIME_TYPE))) {\n return\n }\n\n const toinsert = output_area.element.find(\".\" + CLASS_NAME.split(' ')[0]);\n\n if (output.metadata[EXEC_MIME_TYPE][\"id\"] !== undefined) {\n toinsert[toinsert.length - 1].firstChild.textContent = output.data[JS_MIME_TYPE];\n // store reference to embed id on output_area\n output_area._bokeh_element_id = output.metadata[EXEC_MIME_TYPE][\"id\"];\n }\n if (output.metadata[EXEC_MIME_TYPE][\"server_id\"] !== undefined) {\n const bk_div = document.createElement(\"div\");\n bk_div.innerHTML = output.data[HTML_MIME_TYPE];\n const script_attrs = bk_div.children[0].attributes;\n for (let i = 0; i < script_attrs.length; i++) {\n toinsert[toinsert.length - 1].firstChild.setAttribute(script_attrs[i].name, script_attrs[i].value);\n toinsert[toinsert.length - 1].firstChild.textContent = bk_div.children[0].textContent\n }\n // store reference to server id on output_area\n output_area._bokeh_server_id = output.metadata[EXEC_MIME_TYPE][\"server_id\"];\n }\n }\n\n function register_renderer(events, OutputArea) {\n\n function append_mime(data, metadata, element) {\n // create a DOM node to render to\n const toinsert = this.create_output_subarea(\n metadata,\n CLASS_NAME,\n EXEC_MIME_TYPE\n );\n this.keyboard_manager.register_events(toinsert);\n // Render to node\n const props = {data: data, metadata: metadata[EXEC_MIME_TYPE]};\n render(props, toinsert[toinsert.length - 1]);\n element.append(toinsert);\n return toinsert\n }\n\n /* Handle when an output is cleared or removed */\n events.on('clear_output.CodeCell', handleClearOutput);\n events.on('delete.Cell', handleClearOutput);\n\n /* Handle when a new output is added */\n events.on('output_added.OutputArea', handleAddOutput);\n\n /**\n * Register the mime type and append_mime function with output_area\n */\n OutputArea.prototype.register_mime_type(EXEC_MIME_TYPE, append_mime, {\n /* Is output safe? */\n safe: true,\n /* Index of renderer in `output_area.display_order` */\n index: 0\n });\n }\n\n // register the mime type if in Jupyter Notebook environment and previously unregistered\n if (root.Jupyter !== undefined) {\n const events = require('base/js/events');\n const OutputArea = require('notebook/js/outputarea').OutputArea;\n\n if (OutputArea.prototype.mime_types().indexOf(EXEC_MIME_TYPE) == -1) {\n register_renderer(events, OutputArea);\n }\n }\n if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n root._bokeh_timeout = Date.now() + 5000;\n root._bokeh_failed_load = false;\n }\n\n const NB_LOAD_WARNING = {'data': {'text/html':\n \"
\\n\"+\n \"

\\n\"+\n \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n \"

\\n\"+\n \"\\n\"+\n \"\\n\"+\n \"from bokeh.resources import INLINE\\n\"+\n \"output_notebook(resources=INLINE)\\n\"+\n \"\\n\"+\n \"
\"}};\n\n function display_loaded(error = null) {\n const el = document.getElementById(\"dec2ab1e-cda5-49ce-8f7d-85a26b4e3d5b\");\n if (el != null) {\n const html = (() => {\n if (typeof root.Bokeh === \"undefined\") {\n if (error == null) {\n return \"BokehJS is loading ...\";\n } else {\n return \"BokehJS failed to load.\";\n }\n } else {\n const prefix = `BokehJS ${root.Bokeh.version}`;\n if (error == null) {\n return `${prefix} successfully loaded.`;\n } else {\n return `${prefix} encountered errors while loading and may not function as expected.`;\n }\n }\n })();\n el.innerHTML = html;\n\n if (error != null) {\n const wrapper = document.createElement(\"div\");\n wrapper.style.overflow = \"auto\";\n wrapper.style.height = \"5em\";\n wrapper.style.resize = \"vertical\";\n const content = document.createElement(\"div\");\n content.style.fontFamily = \"monospace\";\n content.style.whiteSpace = \"pre-wrap\";\n content.style.backgroundColor = \"rgb(255, 221, 221)\";\n content.textContent = error.stack ?? error.toString();\n wrapper.append(content);\n el.append(wrapper);\n }\n } else if (Date.now() < root._bokeh_timeout) {\n setTimeout(() => display_loaded(error), 100);\n }\n }\n\n function run_callbacks() {\n try {\n root._bokeh_onload_callbacks.forEach(function(callback) {\n if (callback != null)\n callback();\n });\n } finally {\n delete root._bokeh_onload_callbacks\n }\n console.debug(\"Bokeh: all callbacks have finished\");\n }\n\n function load_libs(css_urls, js_urls, callback) {\n if (css_urls == null) css_urls = [];\n if (js_urls == null) js_urls = [];\n\n root._bokeh_onload_callbacks.push(callback);\n if (root._bokeh_is_loading > 0) {\n console.debug(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n return null;\n }\n if (js_urls == null || js_urls.length === 0) {\n run_callbacks();\n return null;\n }\n console.debug(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n root._bokeh_is_loading = css_urls.length + js_urls.length;\n\n function on_load() {\n root._bokeh_is_loading--;\n if (root._bokeh_is_loading === 0) {\n console.debug(\"Bokeh: all BokehJS libraries/stylesheets loaded\");\n run_callbacks()\n }\n }\n\n function on_error(url) {\n console.error(\"failed to load \" + url);\n }\n\n for (let i = 0; i < css_urls.length; i++) {\n const url = css_urls[i];\n const element = document.createElement(\"link\");\n element.onload = on_load;\n element.onerror = on_error.bind(null, url);\n element.rel = \"stylesheet\";\n element.type = \"text/css\";\n element.href = url;\n console.debug(\"Bokeh: injecting link tag for BokehJS stylesheet: \", url);\n document.body.appendChild(element);\n }\n\n for (let i = 0; i < js_urls.length; i++) {\n const url = js_urls[i];\n const element = document.createElement('script');\n element.onload = on_load;\n element.onerror = on_error.bind(null, url);\n element.async = false;\n element.src = url;\n console.debug(\"Bokeh: injecting script tag for BokehJS library: \", url);\n document.head.appendChild(element);\n }\n };\n\n function inject_raw_css(css) {\n const element = document.createElement(\"style\");\n element.appendChild(document.createTextNode(css));\n document.body.appendChild(element);\n }\n\n const js_urls = [\"https://cdn.bokeh.org/bokeh/release/bokeh-3.5.2.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-gl-3.5.2.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-widgets-3.5.2.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-tables-3.5.2.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-mathjax-3.5.2.min.js\"];\n const css_urls = [];\n\n const inline_js = [ function(Bokeh) {\n Bokeh.set_log_level(\"info\");\n },\nfunction(Bokeh) {\n }\n ];\n\n function run_inline_js() {\n if (root.Bokeh !== undefined || force === true) {\n try {\n for (let i = 0; i < inline_js.length; i++) {\n inline_js[i].call(root, root.Bokeh);\n }\n\n } catch (error) {display_loaded(error);throw error;\n }if (force === true) {\n display_loaded();\n }} else if (Date.now() < root._bokeh_timeout) {\n setTimeout(run_inline_js, 100);\n } else if (!root._bokeh_failed_load) {\n console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n root._bokeh_failed_load = true;\n } else if (force !== true) {\n const cell = $(document.getElementById(\"dec2ab1e-cda5-49ce-8f7d-85a26b4e3d5b\")).parents('.cell').data().cell;\n cell.output_area.append_execute_result(NB_LOAD_WARNING)\n }\n }\n\n if (root._bokeh_is_loading === 0) {\n console.debug(\"Bokeh: BokehJS loaded, going straight to plotting\");\n run_inline_js();\n } else {\n load_libs(css_urls, js_urls, function() {\n console.debug(\"Bokeh: BokehJS plotting callback run at\", now());\n run_inline_js();\n });\n }\n}(window));", + "application/javascript": [ + "'use strict';\n", + "(function(root) {\n", + " function now() {\n", + " return new Date();\n", + " }\n", + "\n", + " const force = true;\n", + "\n", + " if (typeof root._bokeh_onload_callbacks === \"undefined\" || force === true) {\n", + " root._bokeh_onload_callbacks = [];\n", + " root._bokeh_is_loading = undefined;\n", + " }\n", + "\n", + "const JS_MIME_TYPE = 'application/javascript';\n", + " const HTML_MIME_TYPE = 'text/html';\n", + " const EXEC_MIME_TYPE = 'application/vnd.bokehjs_exec.v0+json';\n", + " const CLASS_NAME = 'output_bokeh rendered_html';\n", + "\n", + " /**\n", + " * Render data to the DOM node\n", + " */\n", + " function render(props, node) {\n", + " const script = document.createElement(\"script\");\n", + " node.appendChild(script);\n", + " }\n", + "\n", + " /**\n", + " * Handle when an output is cleared or removed\n", + " */\n", + " function handleClearOutput(event, handle) {\n", + " function drop(id) {\n", + " const view = Bokeh.index.get_by_id(id)\n", + " if (view != null) {\n", + " view.model.document.clear()\n", + " Bokeh.index.delete(view)\n", + " }\n", + " }\n", + "\n", + " const cell = handle.cell;\n", + "\n", + " const id = cell.output_area._bokeh_element_id;\n", + " const server_id = cell.output_area._bokeh_server_id;\n", + "\n", + " // Clean up Bokeh references\n", + " if (id != null) {\n", + " drop(id)\n", + " }\n", + "\n", + " if (server_id !== undefined) {\n", + " // Clean up Bokeh references\n", + " const cmd_clean = \"from bokeh.io.state import curstate; print(curstate().uuid_to_server['\" + server_id + \"'].get_sessions()[0].document.roots[0]._id)\";\n", + " cell.notebook.kernel.execute(cmd_clean, {\n", + " iopub: {\n", + " output: function(msg) {\n", + " const id = msg.content.text.trim()\n", + " drop(id)\n", + " }\n", + " }\n", + " });\n", + " // Destroy server and session\n", + " const cmd_destroy = \"import bokeh.io.notebook as ion; ion.destroy_server('\" + server_id + \"')\";\n", + " cell.notebook.kernel.execute(cmd_destroy);\n", + " }\n", + " }\n", + "\n", + " /**\n", + " * Handle when a new output is added\n", + " */\n", + " function handleAddOutput(event, handle) {\n", + " const output_area = handle.output_area;\n", + " const output = handle.output;\n", + "\n", + " // limit handleAddOutput to display_data with EXEC_MIME_TYPE content only\n", + " if ((output.output_type != \"display_data\") || (!Object.prototype.hasOwnProperty.call(output.data, EXEC_MIME_TYPE))) {\n", + " return\n", + " }\n", + "\n", + " const toinsert = output_area.element.find(\".\" + CLASS_NAME.split(' ')[0]);\n", + "\n", + " if (output.metadata[EXEC_MIME_TYPE][\"id\"] !== undefined) {\n", + " toinsert[toinsert.length - 1].firstChild.textContent = output.data[JS_MIME_TYPE];\n", + " // store reference to embed id on output_area\n", + " output_area._bokeh_element_id = output.metadata[EXEC_MIME_TYPE][\"id\"];\n", + " }\n", + " if (output.metadata[EXEC_MIME_TYPE][\"server_id\"] !== undefined) {\n", + " const bk_div = document.createElement(\"div\");\n", + " bk_div.innerHTML = output.data[HTML_MIME_TYPE];\n", + " const script_attrs = bk_div.children[0].attributes;\n", + " for (let i = 0; i < script_attrs.length; i++) {\n", + " toinsert[toinsert.length - 1].firstChild.setAttribute(script_attrs[i].name, script_attrs[i].value);\n", + " toinsert[toinsert.length - 1].firstChild.textContent = bk_div.children[0].textContent\n", + " }\n", + " // store reference to server id on output_area\n", + " output_area._bokeh_server_id = output.metadata[EXEC_MIME_TYPE][\"server_id\"];\n", + " }\n", + " }\n", + "\n", + " function register_renderer(events, OutputArea) {\n", + "\n", + " function append_mime(data, metadata, element) {\n", + " // create a DOM node to render to\n", + " const toinsert = this.create_output_subarea(\n", + " metadata,\n", + " CLASS_NAME,\n", + " EXEC_MIME_TYPE\n", + " );\n", + " this.keyboard_manager.register_events(toinsert);\n", + " // Render to node\n", + " const props = {data: data, metadata: metadata[EXEC_MIME_TYPE]};\n", + " render(props, toinsert[toinsert.length - 1]);\n", + " element.append(toinsert);\n", + " return toinsert\n", + " }\n", + "\n", + " /* Handle when an output is cleared or removed */\n", + " events.on('clear_output.CodeCell', handleClearOutput);\n", + " events.on('delete.Cell', handleClearOutput);\n", + "\n", + " /* Handle when a new output is added */\n", + " events.on('output_added.OutputArea', handleAddOutput);\n", + "\n", + " /**\n", + " * Register the mime type and append_mime function with output_area\n", + " */\n", + " OutputArea.prototype.register_mime_type(EXEC_MIME_TYPE, append_mime, {\n", + " /* Is output safe? */\n", + " safe: true,\n", + " /* Index of renderer in `output_area.display_order` */\n", + " index: 0\n", + " });\n", + " }\n", + "\n", + " // register the mime type if in Jupyter Notebook environment and previously unregistered\n", + " if (root.Jupyter !== undefined) {\n", + " const events = require('base/js/events');\n", + " const OutputArea = require('notebook/js/outputarea').OutputArea;\n", + "\n", + " if (OutputArea.prototype.mime_types().indexOf(EXEC_MIME_TYPE) == -1) {\n", + " register_renderer(events, OutputArea);\n", + " }\n", + " }\n", + " if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n", + " root._bokeh_timeout = Date.now() + 5000;\n", + " root._bokeh_failed_load = false;\n", + " }\n", + "\n", + " const NB_LOAD_WARNING = {'data': {'text/html':\n", + " \"
\\n\"+\n", + " \"

\\n\"+\n", + " \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n", + " \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n", + " \"

\\n\"+\n", + " \"\\n\"+\n", + " \"\\n\"+\n", + " \"from bokeh.resources import INLINE\\n\"+\n", + " \"output_notebook(resources=INLINE)\\n\"+\n", + " \"\\n\"+\n", + " \"
\"}};\n", + "\n", + " function display_loaded(error = null) {\n", + " const el = document.getElementById(\"dec2ab1e-cda5-49ce-8f7d-85a26b4e3d5b\");\n", + " if (el != null) {\n", + " const html = (() => {\n", + " if (typeof root.Bokeh === \"undefined\") {\n", + " if (error == null) {\n", + " return \"BokehJS is loading ...\";\n", + " } else {\n", + " return \"BokehJS failed to load.\";\n", + " }\n", + " } else {\n", + " const prefix = `BokehJS ${root.Bokeh.version}`;\n", + " if (error == null) {\n", + " return `${prefix} successfully loaded.`;\n", + " } else {\n", + " return `${prefix} encountered errors while loading and may not function as expected.`;\n", + " }\n", + " }\n", + " })();\n", + " el.innerHTML = html;\n", + "\n", + " if (error != null) {\n", + " const wrapper = document.createElement(\"div\");\n", + " wrapper.style.overflow = \"auto\";\n", + " wrapper.style.height = \"5em\";\n", + " wrapper.style.resize = \"vertical\";\n", + " const content = document.createElement(\"div\");\n", + " content.style.fontFamily = \"monospace\";\n", + " content.style.whiteSpace = \"pre-wrap\";\n", + " content.style.backgroundColor = \"rgb(255, 221, 221)\";\n", + " content.textContent = error.stack ?? error.toString();\n", + " wrapper.append(content);\n", + " el.append(wrapper);\n", + " }\n", + " } else if (Date.now() < root._bokeh_timeout) {\n", + " setTimeout(() => display_loaded(error), 100);\n", + " }\n", + " }\n", + "\n", + " function run_callbacks() {\n", + " try {\n", + " root._bokeh_onload_callbacks.forEach(function(callback) {\n", + " if (callback != null)\n", + " callback();\n", + " });\n", + " } finally {\n", + " delete root._bokeh_onload_callbacks\n", + " }\n", + " console.debug(\"Bokeh: all callbacks have finished\");\n", + " }\n", + "\n", + " function load_libs(css_urls, js_urls, callback) {\n", + " if (css_urls == null) css_urls = [];\n", + " if (js_urls == null) js_urls = [];\n", + "\n", + " root._bokeh_onload_callbacks.push(callback);\n", + " if (root._bokeh_is_loading > 0) {\n", + " console.debug(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n", + " return null;\n", + " }\n", + " if (js_urls == null || js_urls.length === 0) {\n", + " run_callbacks();\n", + " return null;\n", + " }\n", + " console.debug(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n", + " root._bokeh_is_loading = css_urls.length + js_urls.length;\n", + "\n", + " function on_load() {\n", + " root._bokeh_is_loading--;\n", + " if (root._bokeh_is_loading === 0) {\n", + " console.debug(\"Bokeh: all BokehJS libraries/stylesheets loaded\");\n", + " run_callbacks()\n", + " }\n", + " }\n", + "\n", + " function on_error(url) {\n", + " console.error(\"failed to load \" + url);\n", + " }\n", + "\n", + " for (let i = 0; i < css_urls.length; i++) {\n", + " const url = css_urls[i];\n", + " const element = document.createElement(\"link\");\n", + " element.onload = on_load;\n", + " element.onerror = on_error.bind(null, url);\n", + " element.rel = \"stylesheet\";\n", + " element.type = \"text/css\";\n", + " element.href = url;\n", + " console.debug(\"Bokeh: injecting link tag for BokehJS stylesheet: \", url);\n", + " document.body.appendChild(element);\n", + " }\n", + "\n", + " for (let i = 0; i < js_urls.length; i++) {\n", + " const url = js_urls[i];\n", + " const element = document.createElement('script');\n", + " element.onload = on_load;\n", + " element.onerror = on_error.bind(null, url);\n", + " element.async = false;\n", + " element.src = url;\n", + " console.debug(\"Bokeh: injecting script tag for BokehJS library: \", url);\n", + " document.head.appendChild(element);\n", + " }\n", + " };\n", + "\n", + " function inject_raw_css(css) {\n", + " const element = document.createElement(\"style\");\n", + " element.appendChild(document.createTextNode(css));\n", + " document.body.appendChild(element);\n", + " }\n", + "\n", + " const js_urls = [\"https://cdn.bokeh.org/bokeh/release/bokeh-3.5.2.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-gl-3.5.2.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-widgets-3.5.2.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-tables-3.5.2.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-mathjax-3.5.2.min.js\"];\n", + " const css_urls = [];\n", + "\n", + " const inline_js = [ function(Bokeh) {\n", + " Bokeh.set_log_level(\"info\");\n", + " },\n", + "function(Bokeh) {\n", + " }\n", + " ];\n", + "\n", + " function run_inline_js() {\n", + " if (root.Bokeh !== undefined || force === true) {\n", + " try {\n", + " for (let i = 0; i < inline_js.length; i++) {\n", + " inline_js[i].call(root, root.Bokeh);\n", + " }\n", + "\n", + " } catch (error) {display_loaded(error);throw error;\n", + " }if (force === true) {\n", + " display_loaded();\n", + " }} else if (Date.now() < root._bokeh_timeout) {\n", + " setTimeout(run_inline_js, 100);\n", + " } else if (!root._bokeh_failed_load) {\n", + " console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n", + " root._bokeh_failed_load = true;\n", + " } else if (force !== true) {\n", + " const cell = $(document.getElementById(\"dec2ab1e-cda5-49ce-8f7d-85a26b4e3d5b\")).parents('.cell').data().cell;\n", + " cell.output_area.append_execute_result(NB_LOAD_WARNING)\n", + " }\n", + " }\n", + "\n", + " if (root._bokeh_is_loading === 0) {\n", + " console.debug(\"Bokeh: BokehJS loaded, going straight to plotting\");\n", + " run_inline_js();\n", + " } else {\n", + " load_libs(css_urls, js_urls, function() {\n", + " console.debug(\"Bokeh: BokehJS plotting callback run at\", now());\n", + " run_inline_js();\n", + " });\n", + " }\n", + "}(window));" + ], "application/vnd.bokehjs_load.v0+json": "" }, "metadata": {}, @@ -1414,7 +1727,7 @@ ], "metadata": { "kernelspec": { - "display_name": ".venv_sprint02", + "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, diff --git a/assets/mlflow/.gitignore b/assets/mlflow/.gitignore index 3f86222..5f2328d 100644 --- a/assets/mlflow/.gitignore +++ b/assets/mlflow/.gitignore @@ -1,2 +1,5 @@ mlartifacts* -mlruns* \ No newline at end of file +mlruns* +catboost_info +*.zip +*.pkl \ No newline at end of file diff --git a/assets/mlflow/comment.txt b/assets/mlflow/comment.txt index f0a4963..c064a60 100644 --- a/assets/mlflow/comment.txt +++ b/assets/mlflow/comment.txt @@ -1 +1,3 @@ -Model for estate \ No newline at end of file +python3 -m venv .venv_ml2 +source .venv_ml2/bin/activate +pip install -r requirements.txt \ No newline at end of file diff --git a/assets/mlflow/requirements.txt b/assets/mlflow/requirements.txt index 971bb15..24d3659 100644 --- a/assets/mlflow/requirements.txt +++ b/assets/mlflow/requirements.txt @@ -1,3 +1,4 @@ -numpy==2.1.1 mlflow==2.16 -scikit-learn \ No newline at end of file +scikit-learn +catboost +numpy \ No newline at end of file diff --git a/assets/mlflow/research.ipynb b/assets/mlflow/research.ipynb index c1ca6cf..dc0ad55 100644 --- a/assets/mlflow/research.ipynb +++ b/assets/mlflow/research.ipynb @@ -19,6 +19,7 @@ "from sklearn.compose import ColumnTransformer\n", "from sklearn.ensemble import RandomForestRegressor\n", "from sklearn.linear_model import LinearRegression\n", + "from catboost import CatBoostRegressor\n", "\n", "from sklearn.metrics import mean_absolute_percentage_error, mean_absolute_error, mean_squared_error" ] @@ -33,24 +34,25 @@ "output_type": "stream", "text": [ "\n", - "Index: 539355 entries, 1979096 to 5189500\n", - "Data columns (total 12 columns):\n", + "Index: 547701 entries, 313199 to 690900\n", + "Data columns (total 13 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", - " 0 price 539355 non-null int64 \n", - " 1 geo_lat 539355 non-null float32 \n", - " 2 geo_lon 539355 non-null float32 \n", - " 3 region 539355 non-null category\n", - " 4 building_type 539355 non-null category\n", - " 5 level 539355 non-null int8 \n", - " 6 levels 539355 non-null int8 \n", - " 7 rooms 539355 non-null int8 \n", - " 8 area 539355 non-null float16 \n", - " 9 kitchen_area 539355 non-null float16 \n", - " 10 object_type 539355 non-null category\n", - " 11 floor_level 539355 non-null object \n", - "dtypes: category(3), float16(2), float32(2), int64(1), int8(3), object(1)\n", - "memory usage: 21.6+ MB\n" + " 0 price 547701 non-null int64 \n", + " 1 date 547701 non-null object \n", + " 2 time 547701 non-null object \n", + " 3 geo_lat 547701 non-null float32 \n", + " 4 geo_lon 547701 non-null float32 \n", + " 5 region 547701 non-null category\n", + " 6 building_type 547701 non-null category\n", + " 7 level 547701 non-null int8 \n", + " 8 levels 547701 non-null int8 \n", + " 9 rooms 547701 non-null int8 \n", + " 10 area 547701 non-null float16 \n", + " 11 kitchen_area 547701 non-null float16 \n", + " 12 object_type 547701 non-null category\n", + "dtypes: category(3), float16(2), float32(2), int64(1), int8(3), object(2)\n", + "memory usage: 26.1+ MB\n" ] } ], @@ -65,7 +67,8 @@ "metadata": {}, "outputs": [], "source": [ - "df = df.rename(columns={'price': 'target'})" + "df = df.rename(columns={'price': 'target'})\n", + "df = df.drop(columns=['date', 'time'])" ] }, { @@ -73,14 +76,6 @@ "execution_count": 4, "metadata": {}, "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/home/andrey/work/institute/MLE/assets/mlflow/.venv_lec_mlflow/lib/python3.10/site-packages/pandas/io/formats/format.py:1458: RuntimeWarning: overflow encountered in cast\n", - " has_large_values = (abs_vals > 1e6).any()\n" - ] - }, { "data": { "text/html": [ @@ -113,84 +108,78 @@ " area\n", " kitchen_area\n", " object_type\n", - " floor_level\n", " \n", " \n", " \n", " \n", - " 1979096\n", - " 1300000\n", - " 52.821098\n", - " 83.113037\n", - " 6817\n", - " 1\n", - " 1\n", - " 1\n", + " 313199\n", + " 4999999\n", + " 59.958458\n", + " 30.215530\n", + " 2661\n", " 3\n", - " 66.50000\n", - " 10.000000\n", + " 8\n", + " 13\n", + " 1\n", + " 36.00000\n", + " 7.199219\n", " 1\n", - " first\n", " \n", " \n", - " 1833303\n", - " 8800000\n", - " 55.707539\n", - " 37.467068\n", + " 2437764\n", + " 2150000\n", + " 45.072674\n", + " 41.936996\n", + " 2900\n", " 3\n", + " 5\n", + " 5\n", " 1\n", - " 15\n", - " 16\n", - " 2\n", - " 46.00000\n", - " 7.000000\n", + " 52.00000\n", + " 15.000000\n", " 1\n", - " hi\n", " \n", " \n", - " 1494335\n", - " 1958000\n", - " 54.988400\n", - " 82.783691\n", - " 9654\n", + " 4949072\n", + " 8600000\n", + " 59.939358\n", + " 30.437069\n", + " 2661\n", " 2\n", - " 13\n", - " 17\n", - " 1\n", - " 36.50000\n", - " 11.960938\n", " 11\n", - " hi\n", + " 22\n", + " 1\n", + " 37.09375\n", + " 9.796875\n", + " 1\n", " \n", " \n", - " 2747476\n", - " 1461600\n", - " 53.298553\n", - " 50.326382\n", - " 3106\n", + " 4109465\n", + " 5100000\n", + " 59.740479\n", + " 30.569540\n", + " 2661\n", + " 1\n", + " 2\n", + " 9\n", " 3\n", - " 5\n", - " 5\n", + " 74.50000\n", + " 9.500000\n", " 1\n", - " 32.59375\n", - " 9.601562\n", - " 11\n", - " last\n", " \n", " \n", - " 5027275\n", - " 3000000\n", - " 42.897934\n", - " 47.624825\n", - " 4007\n", - " 3\n", - " 4\n", - " 10\n", + " 2187702\n", + " 3470000\n", + " 56.324062\n", + " 44.005390\n", + " 2871\n", " 2\n", - " 70.00000\n", - " 12.000000\n", " 11\n", - " mid\n", + " 26\n", + " 2\n", + " 54.00000\n", + " 8.000000\n", + " 11\n", " \n", " \n", " ...\n", @@ -205,116 +194,110 @@ " ...\n", " ...\n", " ...\n", - " ...\n", " \n", " \n", - " 2476626\n", - " 1490000\n", - " 54.943806\n", - " 82.957870\n", - " 9654\n", - " 1\n", + " 5188085\n", + " 2300000\n", + " 57.750603\n", + " 40.866467\n", + " 4189\n", + " 3\n", " 2\n", - " 10\n", + " 3\n", + " 1\n", + " 38.00000\n", + " 11.000000\n", " 1\n", - " 48.06250\n", - " 14.000000\n", - " 11\n", - " low\n", " \n", " \n", - " 1487454\n", - " 19000000\n", - " 55.772240\n", - " 37.731136\n", - " 3\n", - " 3\n", - " 4\n", - " 12\n", + " 4542014\n", + " 6700000\n", + " 55.911720\n", + " 37.737419\n", + " 81\n", " 3\n", - " 100.00000\n", - " 13.000000\n", + " 2\n", + " 5\n", + " 2\n", + " 66.37500\n", + " 8.000000\n", " 1\n", - " mid\n", " \n", " \n", - " 2772844\n", - " 1200000\n", - " 54.474590\n", - " 53.531807\n", - " 2722\n", - " 1\n", - " 5\n", - " 9\n", - " 1\n", - " 32.09375\n", - " 7.000000\n", + " 3306731\n", + " 3850000\n", + " 51.704510\n", + " 39.273037\n", + " 2072\n", + " 2\n", + " 10\n", + " 18\n", + " 3\n", + " 89.50000\n", + " 14.203125\n", " 1\n", - " mid\n", " \n", " \n", - " 3982304\n", - " 2300000\n", - " 55.378265\n", - " 39.053310\n", - " 81\n", - " 1\n", + " 520293\n", + " 1878885\n", + " 54.943577\n", + " 82.958862\n", + " 9654\n", " 1\n", - " 5\n", - " 2\n", - " 49.00000\n", - " 9.000000\n", " 1\n", - " first\n", + " 10\n", + " 3\n", + " 87.75000\n", + " 12.921875\n", + " 11\n", " \n", " \n", - " 5189500\n", - " 9157730\n", - " 55.542957\n", - " 37.479919\n", - " 3\n", - " 1\n", - " 8\n", - " 17\n", + " 690900\n", + " 4097350\n", + " 59.882702\n", + " 30.451246\n", + " 2661\n", " 2\n", - " 52.31250\n", - " 17.593750\n", + " 6\n", + " 23\n", + " 1\n", + " 36.09375\n", + " 16.203125\n", " 11\n", - " mid\n", " \n", " \n", "\n", - "

539355 rows × 12 columns

\n", + "

547701 rows × 11 columns

\n", "" ], "text/plain": [ - " target geo_lat geo_lon region building_type level levels \\\n", - "1979096 1300000 52.821098 83.113037 6817 1 1 1 \n", - "1833303 8800000 55.707539 37.467068 3 1 15 16 \n", - "1494335 1958000 54.988400 82.783691 9654 2 13 17 \n", - "2747476 1461600 53.298553 50.326382 3106 3 5 5 \n", - "5027275 3000000 42.897934 47.624825 4007 3 4 10 \n", - "... ... ... ... ... ... ... ... \n", - "2476626 1490000 54.943806 82.957870 9654 1 2 10 \n", - "1487454 19000000 55.772240 37.731136 3 3 4 12 \n", - "2772844 1200000 54.474590 53.531807 2722 1 5 9 \n", - "3982304 2300000 55.378265 39.053310 81 1 1 5 \n", - "5189500 9157730 55.542957 37.479919 3 1 8 17 \n", - "\n", - " rooms area kitchen_area object_type floor_level \n", - "1979096 3 66.50000 10.000000 1 first \n", - "1833303 2 46.00000 7.000000 1 hi \n", - "1494335 1 36.50000 11.960938 11 hi \n", - "2747476 1 32.59375 9.601562 11 last \n", - "5027275 2 70.00000 12.000000 11 mid \n", - "... ... ... ... ... ... \n", - "2476626 1 48.06250 14.000000 11 low \n", - "1487454 3 100.00000 13.000000 1 mid \n", - "2772844 1 32.09375 7.000000 1 mid \n", - "3982304 2 49.00000 9.000000 1 first \n", - "5189500 2 52.31250 17.593750 11 mid \n", - "\n", - "[539355 rows x 12 columns]" + " target geo_lat geo_lon region building_type level levels \\\n", + "313199 4999999 59.958458 30.215530 2661 3 8 13 \n", + "2437764 2150000 45.072674 41.936996 2900 3 5 5 \n", + "4949072 8600000 59.939358 30.437069 2661 2 11 22 \n", + "4109465 5100000 59.740479 30.569540 2661 1 2 9 \n", + "2187702 3470000 56.324062 44.005390 2871 2 11 26 \n", + "... ... ... ... ... ... ... ... \n", + "5188085 2300000 57.750603 40.866467 4189 3 2 3 \n", + "4542014 6700000 55.911720 37.737419 81 3 2 5 \n", + "3306731 3850000 51.704510 39.273037 2072 2 10 18 \n", + "520293 1878885 54.943577 82.958862 9654 1 1 10 \n", + "690900 4097350 59.882702 30.451246 2661 2 6 23 \n", + "\n", + " rooms area kitchen_area object_type \n", + "313199 1 36.00000 7.199219 1 \n", + "2437764 1 52.00000 15.000000 1 \n", + "4949072 1 37.09375 9.796875 1 \n", + "4109465 3 74.50000 9.500000 1 \n", + "2187702 2 54.00000 8.000000 11 \n", + "... ... ... ... ... \n", + "5188085 1 38.00000 11.000000 1 \n", + "4542014 2 66.37500 8.000000 1 \n", + "3306731 3 89.50000 14.203125 1 \n", + "520293 3 87.75000 12.921875 11 \n", + "690900 1 36.09375 16.203125 11 \n", + "\n", + "[547701 rows x 11 columns]" ] }, "execution_count": 4, @@ -343,7 +326,7 @@ { "data": { "text/plain": [ - "['region', 'building_type', 'object_type', 'floor_level']" + "['region', 'building_type', 'object_type']" ] }, "execution_count": 6, @@ -391,8 +374,8 @@ "outputs": [], "source": [ "s_scaler = StandardScaler()\n", - "l_encoder = OrdinalEncoder(handle_unknown='use_encoded_value', unknown_value=999) # unknown_value нужно выбирать с умом\n", - "regressor = RandomForestRegressor(n_estimators=20, max_depth=10)" + "l_encoder = OrdinalEncoder(handle_unknown='use_encoded_value', unknown_value=99999999) # unknown_value нужно выбирать с умом\n", + "regressor = CatBoostRegressor()" ] }, { @@ -422,6 +405,1013 @@ "execution_count": 10, "metadata": {}, "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Learning rate set to 0.105957\n", + "0:\tlearn: 22102085.4544239\ttotal: 61.3ms\tremaining: 1m 1s\n", + "1:\tlearn: 21994630.3403412\ttotal: 74.7ms\tremaining: 37.3s\n", + "2:\tlearn: 21906687.8196027\ttotal: 88.5ms\tremaining: 29.4s\n", + "3:\tlearn: 21834890.5050552\ttotal: 102ms\tremaining: 25.5s\n", + "4:\tlearn: 21770820.6751194\ttotal: 115ms\tremaining: 22.8s\n", + "5:\tlearn: 21719543.9330108\ttotal: 130ms\tremaining: 21.6s\n", + "6:\tlearn: 21676510.1666598\ttotal: 145ms\tremaining: 20.6s\n", + "7:\tlearn: 21641355.8079016\ttotal: 159ms\tremaining: 19.8s\n", + "8:\tlearn: 21612289.0494648\ttotal: 174ms\tremaining: 19.1s\n", + "9:\tlearn: 21583808.7061085\ttotal: 188ms\tremaining: 18.6s\n", + "10:\tlearn: 21559288.9618040\ttotal: 201ms\tremaining: 18.1s\n", + "11:\tlearn: 21537048.9920531\ttotal: 215ms\tremaining: 17.7s\n", + "12:\tlearn: 21444526.1629239\ttotal: 229ms\tremaining: 17.4s\n", + "13:\tlearn: 21426349.3370315\ttotal: 244ms\tremaining: 17.2s\n", + "14:\tlearn: 21411901.2338278\ttotal: 259ms\tremaining: 17s\n", + "15:\tlearn: 21399279.8023459\ttotal: 272ms\tremaining: 16.7s\n", + "16:\tlearn: 21299421.1434822\ttotal: 285ms\tremaining: 16.5s\n", + "17:\tlearn: 21288560.2595435\ttotal: 298ms\tremaining: 16.3s\n", + "18:\tlearn: 21277368.8876877\ttotal: 314ms\tremaining: 16.2s\n", + "19:\tlearn: 21229205.2938305\ttotal: 328ms\tremaining: 16.1s\n", + "20:\tlearn: 21220238.4828158\ttotal: 343ms\tremaining: 16s\n", + "21:\tlearn: 21212849.7885410\ttotal: 357ms\tremaining: 15.9s\n", + "22:\tlearn: 21205304.4132821\ttotal: 374ms\tremaining: 15.9s\n", + "23:\tlearn: 21198813.8508479\ttotal: 390ms\tremaining: 15.9s\n", + "24:\tlearn: 21184627.2326983\ttotal: 402ms\tremaining: 15.7s\n", + "25:\tlearn: 21172748.3410688\ttotal: 419ms\tremaining: 15.7s\n", + "26:\tlearn: 21103305.4766520\ttotal: 437ms\tremaining: 15.8s\n", + "27:\tlearn: 21096636.4037750\ttotal: 450ms\tremaining: 15.6s\n", + "28:\tlearn: 21082202.2892557\ttotal: 465ms\tremaining: 15.6s\n", + "29:\tlearn: 21077185.5274954\ttotal: 478ms\tremaining: 15.4s\n", + "30:\tlearn: 21071613.1691098\ttotal: 491ms\tremaining: 15.3s\n", + "31:\tlearn: 21067654.8502386\ttotal: 504ms\tremaining: 15.2s\n", + "32:\tlearn: 21053425.8947843\ttotal: 517ms\tremaining: 15.2s\n", + "33:\tlearn: 21038024.0563140\ttotal: 531ms\tremaining: 15.1s\n", + "34:\tlearn: 20961357.9814339\ttotal: 545ms\tremaining: 15s\n", + "35:\tlearn: 20946027.4479676\ttotal: 560ms\tremaining: 15s\n", + "36:\tlearn: 20866676.4104322\ttotal: 574ms\tremaining: 14.9s\n", + "37:\tlearn: 20863078.3182449\ttotal: 587ms\tremaining: 14.9s\n", + "38:\tlearn: 20859910.3609500\ttotal: 603ms\tremaining: 14.9s\n", + "39:\tlearn: 20853462.2703730\ttotal: 615ms\tremaining: 14.8s\n", + "40:\tlearn: 20851610.3209036\ttotal: 627ms\tremaining: 14.7s\n", + "41:\tlearn: 20847674.0809285\ttotal: 641ms\tremaining: 14.6s\n", + "42:\tlearn: 20845384.9263391\ttotal: 655ms\tremaining: 14.6s\n", + "43:\tlearn: 20843256.7428906\ttotal: 670ms\tremaining: 14.6s\n", + "44:\tlearn: 20841580.8594834\ttotal: 683ms\tremaining: 14.5s\n", + "45:\tlearn: 20819301.2718345\ttotal: 699ms\tremaining: 14.5s\n", + "46:\tlearn: 20812094.5913582\ttotal: 715ms\tremaining: 14.5s\n", + "47:\tlearn: 20808932.0866915\ttotal: 727ms\tremaining: 14.4s\n", + "48:\tlearn: 20763172.9200413\ttotal: 742ms\tremaining: 14.4s\n", + "49:\tlearn: 20729084.6574594\ttotal: 757ms\tremaining: 14.4s\n", + "50:\tlearn: 20721820.5403996\ttotal: 773ms\tremaining: 14.4s\n", + "51:\tlearn: 20715664.3732084\ttotal: 787ms\tremaining: 14.3s\n", + "52:\tlearn: 20712658.7025295\ttotal: 801ms\tremaining: 14.3s\n", + "53:\tlearn: 20704254.1704930\ttotal: 819ms\tremaining: 14.4s\n", + "54:\tlearn: 20690967.9220470\ttotal: 835ms\tremaining: 14.3s\n", + "55:\tlearn: 20686546.8978473\ttotal: 849ms\tremaining: 14.3s\n", + "56:\tlearn: 20682362.4255777\ttotal: 868ms\tremaining: 14.4s\n", + "57:\tlearn: 20680744.8113421\ttotal: 883ms\tremaining: 14.3s\n", + "58:\tlearn: 20677926.0871267\ttotal: 897ms\tremaining: 14.3s\n", + "59:\tlearn: 20658478.3098789\ttotal: 915ms\tremaining: 14.3s\n", + "60:\tlearn: 20641964.4472246\ttotal: 937ms\tremaining: 14.4s\n", + "61:\tlearn: 20639551.4216654\ttotal: 954ms\tremaining: 14.4s\n", + "62:\tlearn: 20638344.8919341\ttotal: 967ms\tremaining: 14.4s\n", + "63:\tlearn: 20635991.3894815\ttotal: 984ms\tremaining: 14.4s\n", + "64:\tlearn: 20595846.8116432\ttotal: 1s\tremaining: 14.4s\n", + "65:\tlearn: 20592198.9483046\ttotal: 1.01s\tremaining: 14.4s\n", + "66:\tlearn: 20565316.0060422\ttotal: 1.03s\tremaining: 14.4s\n", + "67:\tlearn: 20563073.6783517\ttotal: 1.05s\tremaining: 14.4s\n", + "68:\tlearn: 20553650.4649650\ttotal: 1.07s\tremaining: 14.5s\n", + "69:\tlearn: 20545510.8230653\ttotal: 1.09s\tremaining: 14.5s\n", + "70:\tlearn: 20544114.9272186\ttotal: 1.11s\tremaining: 14.5s\n", + "71:\tlearn: 20541689.8802451\ttotal: 1.13s\tremaining: 14.6s\n", + "72:\tlearn: 20538792.7074671\ttotal: 1.16s\tremaining: 14.7s\n", + "73:\tlearn: 20517134.0713648\ttotal: 1.18s\tremaining: 14.7s\n", + "74:\tlearn: 20510477.9089445\ttotal: 1.19s\tremaining: 14.7s\n", + "75:\tlearn: 20494649.9067257\ttotal: 1.21s\tremaining: 14.8s\n", + "76:\tlearn: 20490851.9879851\ttotal: 1.24s\tremaining: 14.8s\n", + "77:\tlearn: 20488939.9621874\ttotal: 1.25s\tremaining: 14.8s\n", + "78:\tlearn: 20432532.8171644\ttotal: 1.26s\tremaining: 14.8s\n", + "79:\tlearn: 20428397.7107150\ttotal: 1.28s\tremaining: 14.7s\n", + "80:\tlearn: 20421638.7734419\ttotal: 1.3s\tremaining: 14.7s\n", + "81:\tlearn: 20421021.7388457\ttotal: 1.31s\tremaining: 14.7s\n", + "82:\tlearn: 20406404.2376730\ttotal: 1.33s\tremaining: 14.7s\n", + "83:\tlearn: 20021682.5008511\ttotal: 1.34s\tremaining: 14.6s\n", + "84:\tlearn: 20018322.6048631\ttotal: 1.36s\tremaining: 14.6s\n", + "85:\tlearn: 20004841.3476490\ttotal: 1.37s\tremaining: 14.6s\n", + "86:\tlearn: 19985666.0092745\ttotal: 1.39s\tremaining: 14.5s\n", + "87:\tlearn: 19983778.1947243\ttotal: 1.4s\tremaining: 14.5s\n", + "88:\tlearn: 19982460.1107908\ttotal: 1.41s\tremaining: 14.5s\n", + "89:\tlearn: 19979128.5494690\ttotal: 1.42s\tremaining: 14.4s\n", + "90:\tlearn: 19974094.9707357\ttotal: 1.44s\tremaining: 14.3s\n", + "91:\tlearn: 19972006.9431031\ttotal: 1.45s\tremaining: 14.3s\n", + "92:\tlearn: 19970846.2845466\ttotal: 1.46s\tremaining: 14.2s\n", + "93:\tlearn: 19968858.0073042\ttotal: 1.47s\tremaining: 14.2s\n", + "94:\tlearn: 19921720.6252972\ttotal: 1.49s\tremaining: 14.2s\n", + "95:\tlearn: 19916568.5707839\ttotal: 1.5s\tremaining: 14.1s\n", + "96:\tlearn: 19913228.5247508\ttotal: 1.51s\tremaining: 14.1s\n", + "97:\tlearn: 19901982.4625895\ttotal: 1.52s\tremaining: 14s\n", + "98:\tlearn: 19836107.7247888\ttotal: 1.54s\tremaining: 14.1s\n", + "99:\tlearn: 19834724.7455166\ttotal: 1.56s\tremaining: 14.1s\n", + "100:\tlearn: 19832811.9745741\ttotal: 1.59s\tremaining: 14.1s\n", + "101:\tlearn: 19818491.2851567\ttotal: 1.61s\tremaining: 14.1s\n", + "102:\tlearn: 19815779.3719026\ttotal: 1.63s\tremaining: 14.2s\n", + "103:\tlearn: 19814215.0962787\ttotal: 1.65s\tremaining: 14.2s\n", + "104:\tlearn: 19782274.6892663\ttotal: 1.67s\tremaining: 14.2s\n", + "105:\tlearn: 19777945.6507456\ttotal: 1.69s\tremaining: 14.2s\n", + "106:\tlearn: 19770488.9772154\ttotal: 1.7s\tremaining: 14.2s\n", + "107:\tlearn: 19769758.0023174\ttotal: 1.72s\tremaining: 14.2s\n", + "108:\tlearn: 19767541.9303017\ttotal: 1.74s\tremaining: 14.2s\n", + "109:\tlearn: 19766992.0126300\ttotal: 1.75s\tremaining: 14.2s\n", + "110:\tlearn: 19765032.8837298\ttotal: 1.77s\tremaining: 14.2s\n", + "111:\tlearn: 19705204.6771073\ttotal: 1.78s\tremaining: 14.2s\n", + "112:\tlearn: 19703649.0394020\ttotal: 1.8s\tremaining: 14.1s\n", + "113:\tlearn: 19693038.0415419\ttotal: 1.82s\tremaining: 14.1s\n", + "114:\tlearn: 19690294.4304072\ttotal: 1.84s\tremaining: 14.2s\n", + "115:\tlearn: 19686529.4709294\ttotal: 1.86s\tremaining: 14.2s\n", + "116:\tlearn: 19684887.8267152\ttotal: 1.88s\tremaining: 14.2s\n", + "117:\tlearn: 19369465.6970761\ttotal: 1.9s\tremaining: 14.2s\n", + "118:\tlearn: 19368868.0416380\ttotal: 1.92s\tremaining: 14.2s\n", + "119:\tlearn: 19334590.5868513\ttotal: 1.94s\tremaining: 14.2s\n", + "120:\tlearn: 19332200.0832597\ttotal: 1.95s\tremaining: 14.2s\n", + "121:\tlearn: 19320130.9244745\ttotal: 1.97s\tremaining: 14.2s\n", + "122:\tlearn: 19318220.9448337\ttotal: 2s\tremaining: 14.2s\n", + "123:\tlearn: 18941546.2095714\ttotal: 2.02s\tremaining: 14.3s\n", + "124:\tlearn: 18941056.2836883\ttotal: 2.05s\tremaining: 14.3s\n", + "125:\tlearn: 18939637.9662976\ttotal: 2.07s\tremaining: 14.3s\n", + "126:\tlearn: 18938172.4621610\ttotal: 2.1s\tremaining: 14.4s\n", + "127:\tlearn: 18935889.3619752\ttotal: 2.13s\tremaining: 14.5s\n", + "128:\tlearn: 18928784.7025346\ttotal: 2.15s\tremaining: 14.6s\n", + "129:\tlearn: 18926981.6933453\ttotal: 2.17s\tremaining: 14.5s\n", + "130:\tlearn: 18830178.3173696\ttotal: 2.19s\tremaining: 14.5s\n", + "131:\tlearn: 18828102.3918672\ttotal: 2.22s\tremaining: 14.6s\n", + "132:\tlearn: 18825755.9987015\ttotal: 2.24s\tremaining: 14.6s\n", + "133:\tlearn: 18793049.5462155\ttotal: 2.26s\tremaining: 14.6s\n", + "134:\tlearn: 18791452.8400128\ttotal: 2.27s\tremaining: 14.6s\n", + "135:\tlearn: 18484591.4924421\ttotal: 2.29s\tremaining: 14.6s\n", + "136:\tlearn: 18482373.1605741\ttotal: 2.31s\tremaining: 14.5s\n", + "137:\tlearn: 18414571.2543321\ttotal: 2.32s\tremaining: 14.5s\n", + "138:\tlearn: 18412913.4160574\ttotal: 2.35s\tremaining: 14.5s\n", + "139:\tlearn: 18409214.1141794\ttotal: 2.36s\tremaining: 14.5s\n", + "140:\tlearn: 18395140.1008086\ttotal: 2.38s\tremaining: 14.5s\n", + "141:\tlearn: 18390939.2248151\ttotal: 2.4s\tremaining: 14.5s\n", + "142:\tlearn: 18377925.8298573\ttotal: 2.42s\tremaining: 14.5s\n", + "143:\tlearn: 18371775.1291009\ttotal: 2.43s\tremaining: 14.5s\n", + "144:\tlearn: 18370251.1042623\ttotal: 2.45s\tremaining: 14.4s\n", + "145:\tlearn: 18332707.1499911\ttotal: 2.46s\tremaining: 14.4s\n", + "146:\tlearn: 18330693.2665230\ttotal: 2.48s\tremaining: 14.4s\n", + "147:\tlearn: 18329408.2952767\ttotal: 2.49s\tremaining: 14.3s\n", + "148:\tlearn: 18321783.9892793\ttotal: 2.5s\tremaining: 14.3s\n", + "149:\tlearn: 18321270.4958267\ttotal: 2.52s\tremaining: 14.3s\n", + "150:\tlearn: 18310325.1681801\ttotal: 2.53s\tremaining: 14.2s\n", + "151:\tlearn: 18299986.9413893\ttotal: 2.55s\tremaining: 14.2s\n", + "152:\tlearn: 18290217.7479708\ttotal: 2.56s\tremaining: 14.2s\n", + "153:\tlearn: 18280975.8537910\ttotal: 2.58s\tremaining: 14.2s\n", + "154:\tlearn: 18272215.6509019\ttotal: 2.6s\tremaining: 14.1s\n", + "155:\tlearn: 18263878.2178516\ttotal: 2.61s\tremaining: 14.1s\n", + "156:\tlearn: 18256009.4859248\ttotal: 2.63s\tremaining: 14.1s\n", + "157:\tlearn: 18248529.7799856\ttotal: 2.64s\tremaining: 14.1s\n", + "158:\tlearn: 18241388.0845094\ttotal: 2.66s\tremaining: 14.1s\n", + "159:\tlearn: 18234700.5127085\ttotal: 2.67s\tremaining: 14s\n", + "160:\tlearn: 18228095.5839778\ttotal: 2.69s\tremaining: 14s\n", + "161:\tlearn: 18222087.5153066\ttotal: 2.7s\tremaining: 14s\n", + "162:\tlearn: 18215963.2971261\ttotal: 2.72s\tremaining: 14s\n", + "163:\tlearn: 18210272.5545163\ttotal: 2.73s\tremaining: 13.9s\n", + "164:\tlearn: 18208920.7703569\ttotal: 2.74s\tremaining: 13.9s\n", + "165:\tlearn: 18204704.7145239\ttotal: 2.75s\tremaining: 13.8s\n", + "166:\tlearn: 18187135.8260335\ttotal: 2.77s\tremaining: 13.8s\n", + "167:\tlearn: 18183064.7135734\ttotal: 2.78s\tremaining: 13.8s\n", + "168:\tlearn: 18177887.1670860\ttotal: 2.8s\tremaining: 13.8s\n", + "169:\tlearn: 18173022.2110313\ttotal: 2.81s\tremaining: 13.7s\n", + "170:\tlearn: 18168573.4167384\ttotal: 2.83s\tremaining: 13.7s\n", + "171:\tlearn: 18165036.1971623\ttotal: 2.85s\tremaining: 13.7s\n", + "172:\tlearn: 18161841.9822954\ttotal: 2.87s\tremaining: 13.7s\n", + "173:\tlearn: 18129860.2061383\ttotal: 2.88s\tremaining: 13.7s\n", + "174:\tlearn: 18127931.5161091\ttotal: 2.89s\tremaining: 13.6s\n", + "175:\tlearn: 18124997.7778403\ttotal: 2.91s\tremaining: 13.6s\n", + "176:\tlearn: 18122975.2084322\ttotal: 2.92s\tremaining: 13.6s\n", + "177:\tlearn: 18120855.5325733\ttotal: 2.93s\tremaining: 13.5s\n", + "178:\tlearn: 18117907.6019994\ttotal: 2.95s\tremaining: 13.5s\n", + "179:\tlearn: 18116674.0864027\ttotal: 2.96s\tremaining: 13.5s\n", + "180:\tlearn: 18114086.9287957\ttotal: 2.97s\tremaining: 13.4s\n", + "181:\tlearn: 18087100.0827926\ttotal: 2.98s\tremaining: 13.4s\n", + "182:\tlearn: 18071944.2213105\ttotal: 3s\tremaining: 13.4s\n", + "183:\tlearn: 17952691.4261792\ttotal: 3.01s\tremaining: 13.4s\n", + "184:\tlearn: 17950298.6715866\ttotal: 3.02s\tremaining: 13.3s\n", + "185:\tlearn: 17949031.8169417\ttotal: 3.04s\tremaining: 13.3s\n", + "186:\tlearn: 17937943.5186847\ttotal: 3.05s\tremaining: 13.3s\n", + "187:\tlearn: 17937014.8027177\ttotal: 3.06s\tremaining: 13.2s\n", + "188:\tlearn: 17936493.5945773\ttotal: 3.07s\tremaining: 13.2s\n", + "189:\tlearn: 17935386.0093649\ttotal: 3.09s\tremaining: 13.2s\n", + "190:\tlearn: 17934203.8644718\ttotal: 3.1s\tremaining: 13.1s\n", + "191:\tlearn: 17928336.5184065\ttotal: 3.11s\tremaining: 13.1s\n", + "192:\tlearn: 17925443.1940046\ttotal: 3.13s\tremaining: 13.1s\n", + "193:\tlearn: 17924535.5533845\ttotal: 3.14s\tremaining: 13s\n", + "194:\tlearn: 17917225.8802206\ttotal: 3.16s\tremaining: 13s\n", + "195:\tlearn: 17904437.4148190\ttotal: 3.17s\tremaining: 13s\n", + "196:\tlearn: 17902915.3467923\ttotal: 3.19s\tremaining: 13s\n", + "197:\tlearn: 17900924.7512305\ttotal: 3.2s\tremaining: 13s\n", + "198:\tlearn: 17899976.2262471\ttotal: 3.22s\tremaining: 13s\n", + "199:\tlearn: 17896573.5977064\ttotal: 3.24s\tremaining: 13s\n", + "200:\tlearn: 17894480.1301072\ttotal: 3.26s\tremaining: 13s\n", + "201:\tlearn: 17891369.5414483\ttotal: 3.28s\tremaining: 13s\n", + "202:\tlearn: 17853776.3679239\ttotal: 3.31s\tremaining: 13s\n", + "203:\tlearn: 17851457.0828592\ttotal: 3.32s\tremaining: 13s\n", + "204:\tlearn: 17849621.6767992\ttotal: 3.33s\tremaining: 12.9s\n", + "205:\tlearn: 17848392.5509482\ttotal: 3.35s\tremaining: 12.9s\n", + "206:\tlearn: 17845597.2428619\ttotal: 3.36s\tremaining: 12.9s\n", + "207:\tlearn: 17841951.2763157\ttotal: 3.38s\tremaining: 12.9s\n", + "208:\tlearn: 17829332.8912371\ttotal: 3.4s\tremaining: 12.9s\n", + "209:\tlearn: 17825984.1152963\ttotal: 3.41s\tremaining: 12.8s\n", + "210:\tlearn: 17821360.2498463\ttotal: 3.43s\tremaining: 12.8s\n", + "211:\tlearn: 17816041.9633158\ttotal: 3.44s\tremaining: 12.8s\n", + "212:\tlearn: 17815089.0154101\ttotal: 3.46s\tremaining: 12.8s\n", + "213:\tlearn: 17812260.4222221\ttotal: 3.47s\tremaining: 12.8s\n", + "214:\tlearn: 17811642.1796060\ttotal: 3.49s\tremaining: 12.7s\n", + "215:\tlearn: 17811104.8656724\ttotal: 3.5s\tremaining: 12.7s\n", + "216:\tlearn: 17810456.2984828\ttotal: 3.51s\tremaining: 12.7s\n", + "217:\tlearn: 17809982.4909707\ttotal: 3.52s\tremaining: 12.6s\n", + "218:\tlearn: 17809543.7803178\ttotal: 3.54s\tremaining: 12.6s\n", + "219:\tlearn: 17809136.8325569\ttotal: 3.55s\tremaining: 12.6s\n", + "220:\tlearn: 17808758.7315278\ttotal: 3.56s\tremaining: 12.6s\n", + "221:\tlearn: 17808406.9145618\ttotal: 3.58s\tremaining: 12.6s\n", + "222:\tlearn: 17806754.0179687\ttotal: 3.6s\tremaining: 12.6s\n", + "223:\tlearn: 17806262.4885592\ttotal: 3.62s\tremaining: 12.5s\n", + "224:\tlearn: 17805319.3776209\ttotal: 3.63s\tremaining: 12.5s\n", + "225:\tlearn: 17805011.6013482\ttotal: 3.65s\tremaining: 12.5s\n", + "226:\tlearn: 17804724.0362310\ttotal: 3.66s\tremaining: 12.5s\n", + "227:\tlearn: 17793961.7547867\ttotal: 3.68s\tremaining: 12.5s\n", + "228:\tlearn: 17793044.3976904\ttotal: 3.7s\tremaining: 12.5s\n", + "229:\tlearn: 17791876.3449986\ttotal: 3.72s\tremaining: 12.5s\n", + "230:\tlearn: 17770039.2877531\ttotal: 3.74s\tremaining: 12.4s\n", + "231:\tlearn: 17769759.3423197\ttotal: 3.75s\tremaining: 12.4s\n", + "232:\tlearn: 17769498.1846872\ttotal: 3.77s\tremaining: 12.4s\n", + "233:\tlearn: 17769106.6516586\ttotal: 3.78s\tremaining: 12.4s\n", + "234:\tlearn: 17765866.7512613\ttotal: 3.8s\tremaining: 12.4s\n", + "235:\tlearn: 17763818.0836765\ttotal: 3.81s\tremaining: 12.3s\n", + "236:\tlearn: 17761637.5687877\ttotal: 3.83s\tremaining: 12.3s\n", + "237:\tlearn: 17755293.6166299\ttotal: 3.85s\tremaining: 12.3s\n", + "238:\tlearn: 17749597.6285121\ttotal: 3.87s\tremaining: 12.3s\n", + "239:\tlearn: 17731193.4780969\ttotal: 3.89s\tremaining: 12.3s\n", + "240:\tlearn: 17730941.1840209\ttotal: 3.9s\tremaining: 12.3s\n", + "241:\tlearn: 17730651.4109866\ttotal: 3.91s\tremaining: 12.3s\n", + "242:\tlearn: 17729951.1772204\ttotal: 3.93s\tremaining: 12.2s\n", + "243:\tlearn: 17725674.6169533\ttotal: 3.94s\tremaining: 12.2s\n", + "244:\tlearn: 17724397.3837970\ttotal: 3.96s\tremaining: 12.2s\n", + "245:\tlearn: 17723085.9667878\ttotal: 3.97s\tremaining: 12.2s\n", + "246:\tlearn: 17716068.0643361\ttotal: 3.99s\tremaining: 12.2s\n", + "247:\tlearn: 17685621.7941613\ttotal: 4s\tremaining: 12.1s\n", + "248:\tlearn: 17684272.6716694\ttotal: 4.02s\tremaining: 12.1s\n", + "249:\tlearn: 17683390.0888279\ttotal: 4.03s\tremaining: 12.1s\n", + "250:\tlearn: 17683052.4845925\ttotal: 4.04s\tremaining: 12.1s\n", + "251:\tlearn: 17678624.0868252\ttotal: 4.06s\tremaining: 12.1s\n", + "252:\tlearn: 17665657.9640584\ttotal: 4.08s\tremaining: 12s\n", + "253:\tlearn: 17664624.5487132\ttotal: 4.09s\tremaining: 12s\n", + "254:\tlearn: 17663925.0646167\ttotal: 4.1s\tremaining: 12s\n", + "255:\tlearn: 17653813.6196925\ttotal: 4.12s\tremaining: 12s\n", + "256:\tlearn: 17636698.5157040\ttotal: 4.13s\tremaining: 11.9s\n", + "257:\tlearn: 17634671.9750893\ttotal: 4.15s\tremaining: 11.9s\n", + "258:\tlearn: 17633930.6422340\ttotal: 4.16s\tremaining: 11.9s\n", + "259:\tlearn: 17633026.0861171\ttotal: 4.17s\tremaining: 11.9s\n", + "260:\tlearn: 17632489.1254856\ttotal: 4.19s\tremaining: 11.9s\n", + "261:\tlearn: 17628474.9187765\ttotal: 4.2s\tremaining: 11.8s\n", + "262:\tlearn: 17627320.9817928\ttotal: 4.22s\tremaining: 11.8s\n", + "263:\tlearn: 17626116.4772868\ttotal: 4.23s\tremaining: 11.8s\n", + "264:\tlearn: 17623329.0754817\ttotal: 4.25s\tremaining: 11.8s\n", + "265:\tlearn: 17622243.1901613\ttotal: 4.26s\tremaining: 11.8s\n", + "266:\tlearn: 17550321.8250878\ttotal: 4.28s\tremaining: 11.8s\n", + "267:\tlearn: 17549755.3651767\ttotal: 4.33s\tremaining: 11.8s\n", + "268:\tlearn: 17545607.1212430\ttotal: 4.37s\tremaining: 11.9s\n", + "269:\tlearn: 17541242.2629221\ttotal: 4.38s\tremaining: 11.9s\n", + "270:\tlearn: 17499407.7313592\ttotal: 4.41s\tremaining: 11.9s\n", + "271:\tlearn: 17499145.8282321\ttotal: 4.43s\tremaining: 11.9s\n", + "272:\tlearn: 17498934.5535116\ttotal: 4.44s\tremaining: 11.8s\n", + "273:\tlearn: 17498347.2546318\ttotal: 4.46s\tremaining: 11.8s\n", + "274:\tlearn: 17498149.7061684\ttotal: 4.47s\tremaining: 11.8s\n", + "275:\tlearn: 17497860.3337909\ttotal: 4.48s\tremaining: 11.8s\n", + "276:\tlearn: 17497134.2565818\ttotal: 4.5s\tremaining: 11.7s\n", + "277:\tlearn: 17496943.1446578\ttotal: 4.51s\tremaining: 11.7s\n", + "278:\tlearn: 17495461.7397646\ttotal: 4.53s\tremaining: 11.7s\n", + "279:\tlearn: 17492860.8467310\ttotal: 4.54s\tremaining: 11.7s\n", + "280:\tlearn: 17492256.7750564\ttotal: 4.56s\tremaining: 11.7s\n", + "281:\tlearn: 17491315.8920024\ttotal: 4.58s\tremaining: 11.7s\n", + "282:\tlearn: 17488802.8492737\ttotal: 4.59s\tremaining: 11.6s\n", + "283:\tlearn: 17479802.6541152\ttotal: 4.6s\tremaining: 11.6s\n", + "284:\tlearn: 17477169.5331720\ttotal: 4.62s\tremaining: 11.6s\n", + "285:\tlearn: 17474743.6190942\ttotal: 4.63s\tremaining: 11.6s\n", + "286:\tlearn: 17468342.7955232\ttotal: 4.65s\tremaining: 11.6s\n", + "287:\tlearn: 17467579.9985437\ttotal: 4.67s\tremaining: 11.5s\n", + "288:\tlearn: 17467009.9684055\ttotal: 4.68s\tremaining: 11.5s\n", + "289:\tlearn: 17464125.0260113\ttotal: 4.7s\tremaining: 11.5s\n", + "290:\tlearn: 17463508.0564477\ttotal: 4.71s\tremaining: 11.5s\n", + "291:\tlearn: 17453183.2620432\ttotal: 4.73s\tremaining: 11.5s\n", + "292:\tlearn: 17452971.0671546\ttotal: 4.74s\tremaining: 11.4s\n", + "293:\tlearn: 17452198.5884342\ttotal: 4.76s\tremaining: 11.4s\n", + "294:\tlearn: 17450925.6159031\ttotal: 4.78s\tremaining: 11.4s\n", + "295:\tlearn: 17450685.1155343\ttotal: 4.79s\tremaining: 11.4s\n", + "296:\tlearn: 17447975.7379237\ttotal: 4.8s\tremaining: 11.4s\n", + "297:\tlearn: 17446417.7251561\ttotal: 4.82s\tremaining: 11.3s\n", + "298:\tlearn: 17446166.7629704\ttotal: 4.83s\tremaining: 11.3s\n", + "299:\tlearn: 17445963.1442260\ttotal: 4.84s\tremaining: 11.3s\n", + "300:\tlearn: 17445745.7958927\ttotal: 4.86s\tremaining: 11.3s\n", + "301:\tlearn: 17444963.9290154\ttotal: 4.87s\tremaining: 11.3s\n", + "302:\tlearn: 17432650.1591210\ttotal: 4.9s\tremaining: 11.3s\n", + "303:\tlearn: 17430525.1210288\ttotal: 4.92s\tremaining: 11.3s\n", + "304:\tlearn: 17418414.4601453\ttotal: 4.93s\tremaining: 11.2s\n", + "305:\tlearn: 17417977.4735651\ttotal: 4.95s\tremaining: 11.2s\n", + "306:\tlearn: 17335624.2943914\ttotal: 4.97s\tremaining: 11.2s\n", + "307:\tlearn: 17323558.9233681\ttotal: 4.99s\tremaining: 11.2s\n", + "308:\tlearn: 17323047.3527617\ttotal: 5s\tremaining: 11.2s\n", + "309:\tlearn: 17322403.3488620\ttotal: 5.02s\tremaining: 11.2s\n", + "310:\tlearn: 17322187.6973801\ttotal: 5.03s\tremaining: 11.2s\n", + "311:\tlearn: 17320898.8497406\ttotal: 5.05s\tremaining: 11.1s\n", + "312:\tlearn: 17312668.7000429\ttotal: 5.07s\tremaining: 11.1s\n", + "313:\tlearn: 17299277.5985403\ttotal: 5.09s\tremaining: 11.1s\n", + "314:\tlearn: 17298175.9786240\ttotal: 5.11s\tremaining: 11.1s\n", + "315:\tlearn: 17296005.0430765\ttotal: 5.12s\tremaining: 11.1s\n", + "316:\tlearn: 17295834.3986842\ttotal: 5.13s\tremaining: 11.1s\n", + "317:\tlearn: 17295646.8271436\ttotal: 5.14s\tremaining: 11s\n", + "318:\tlearn: 17295412.2240763\ttotal: 5.16s\tremaining: 11s\n", + "319:\tlearn: 17295269.3891063\ttotal: 5.17s\tremaining: 11s\n", + "320:\tlearn: 17294720.1427139\ttotal: 5.19s\tremaining: 11s\n", + "321:\tlearn: 17280405.8179874\ttotal: 5.21s\tremaining: 11s\n", + "322:\tlearn: 17279788.6705542\ttotal: 5.23s\tremaining: 11s\n", + "323:\tlearn: 17259578.2219214\ttotal: 5.25s\tremaining: 11s\n", + "324:\tlearn: 17258995.8851109\ttotal: 5.27s\tremaining: 10.9s\n", + "325:\tlearn: 17256802.0040208\ttotal: 5.29s\tremaining: 10.9s\n", + "326:\tlearn: 17245667.9352932\ttotal: 5.31s\tremaining: 10.9s\n", + "327:\tlearn: 17245157.2383849\ttotal: 5.34s\tremaining: 10.9s\n", + "328:\tlearn: 17244420.0505767\ttotal: 5.36s\tremaining: 10.9s\n", + "329:\tlearn: 17240620.9311856\ttotal: 5.38s\tremaining: 10.9s\n", + "330:\tlearn: 17240126.6382259\ttotal: 5.4s\tremaining: 10.9s\n", + "331:\tlearn: 17239554.3263042\ttotal: 5.42s\tremaining: 10.9s\n", + "332:\tlearn: 17239249.4122676\ttotal: 5.43s\tremaining: 10.9s\n", + "333:\tlearn: 17237315.5959603\ttotal: 5.45s\tremaining: 10.9s\n", + "334:\tlearn: 17237170.4183008\ttotal: 5.46s\tremaining: 10.8s\n", + "335:\tlearn: 17235498.1709182\ttotal: 5.48s\tremaining: 10.8s\n", + "336:\tlearn: 17154286.9322136\ttotal: 5.49s\tremaining: 10.8s\n", + "337:\tlearn: 17152860.5403583\ttotal: 5.51s\tremaining: 10.8s\n", + "338:\tlearn: 17139897.5803445\ttotal: 5.52s\tremaining: 10.8s\n", + "339:\tlearn: 17139685.6194353\ttotal: 5.53s\tremaining: 10.7s\n", + "340:\tlearn: 17129406.8909698\ttotal: 5.54s\tremaining: 10.7s\n", + "341:\tlearn: 17126386.5318429\ttotal: 5.56s\tremaining: 10.7s\n", + "342:\tlearn: 17125338.5826429\ttotal: 5.57s\tremaining: 10.7s\n", + "343:\tlearn: 17124937.1764028\ttotal: 5.58s\tremaining: 10.7s\n", + "344:\tlearn: 17124773.5128614\ttotal: 5.59s\tremaining: 10.6s\n", + "345:\tlearn: 17123822.0085471\ttotal: 5.61s\tremaining: 10.6s\n", + "346:\tlearn: 17122604.8415169\ttotal: 5.62s\tremaining: 10.6s\n", + "347:\tlearn: 17121767.5370013\ttotal: 5.63s\tremaining: 10.6s\n", + "348:\tlearn: 17109471.1428348\ttotal: 5.65s\tremaining: 10.5s\n", + "349:\tlearn: 17092688.7777393\ttotal: 5.66s\tremaining: 10.5s\n", + "350:\tlearn: 17081854.5539987\ttotal: 5.68s\tremaining: 10.5s\n", + "351:\tlearn: 17081117.2220910\ttotal: 5.69s\tremaining: 10.5s\n", + "352:\tlearn: 17079431.1991192\ttotal: 5.7s\tremaining: 10.5s\n", + "353:\tlearn: 17065749.4676464\ttotal: 5.72s\tremaining: 10.4s\n", + "354:\tlearn: 17050839.2238400\ttotal: 5.73s\tremaining: 10.4s\n", + "355:\tlearn: 17050106.8831270\ttotal: 5.75s\tremaining: 10.4s\n", + "356:\tlearn: 17046033.2332065\ttotal: 5.76s\tremaining: 10.4s\n", + "357:\tlearn: 17043704.2415802\ttotal: 5.78s\tremaining: 10.4s\n", + "358:\tlearn: 17034226.2631681\ttotal: 5.79s\tremaining: 10.3s\n", + "359:\tlearn: 17019515.6806659\ttotal: 5.8s\tremaining: 10.3s\n", + "360:\tlearn: 17018472.9763746\ttotal: 5.82s\tremaining: 10.3s\n", + "361:\tlearn: 17017909.7121151\ttotal: 5.83s\tremaining: 10.3s\n", + "362:\tlearn: 17017463.3942640\ttotal: 5.84s\tremaining: 10.3s\n", + "363:\tlearn: 17016467.4317116\ttotal: 5.86s\tremaining: 10.2s\n", + "364:\tlearn: 17016320.3746025\ttotal: 5.87s\tremaining: 10.2s\n", + "365:\tlearn: 17014043.0108512\ttotal: 5.88s\tremaining: 10.2s\n", + "366:\tlearn: 17013536.3710672\ttotal: 5.89s\tremaining: 10.2s\n", + "367:\tlearn: 17011993.2014165\ttotal: 5.91s\tremaining: 10.1s\n", + "368:\tlearn: 17011849.5641841\ttotal: 5.92s\tremaining: 10.1s\n", + "369:\tlearn: 17011403.7126883\ttotal: 5.93s\tremaining: 10.1s\n", + "370:\tlearn: 17009763.5741945\ttotal: 5.94s\tremaining: 10.1s\n", + "371:\tlearn: 17009382.7519630\ttotal: 5.95s\tremaining: 10.1s\n", + "372:\tlearn: 17008464.7915054\ttotal: 5.97s\tremaining: 10s\n", + "373:\tlearn: 17008143.8161261\ttotal: 5.98s\tremaining: 10s\n", + "374:\tlearn: 16996814.2215431\ttotal: 5.99s\tremaining: 9.99s\n", + "375:\tlearn: 16996377.3351825\ttotal: 6.01s\tremaining: 9.97s\n", + "376:\tlearn: 16996037.5806770\ttotal: 6.02s\tremaining: 9.95s\n", + "377:\tlearn: 16991953.6478199\ttotal: 6.03s\tremaining: 9.93s\n", + "378:\tlearn: 16961328.6727692\ttotal: 6.05s\tremaining: 9.91s\n", + "379:\tlearn: 16957664.4831621\ttotal: 6.06s\tremaining: 9.89s\n", + "380:\tlearn: 16956856.4526881\ttotal: 6.08s\tremaining: 9.87s\n", + "381:\tlearn: 16947754.5891887\ttotal: 6.09s\tremaining: 9.86s\n", + "382:\tlearn: 16937471.3061729\ttotal: 6.11s\tremaining: 9.84s\n", + "383:\tlearn: 16910717.2697228\ttotal: 6.13s\tremaining: 9.84s\n", + "384:\tlearn: 16883021.8749316\ttotal: 6.15s\tremaining: 9.83s\n", + "385:\tlearn: 16874077.6620256\ttotal: 6.17s\tremaining: 9.82s\n", + "386:\tlearn: 16859663.0508862\ttotal: 6.19s\tremaining: 9.81s\n", + "387:\tlearn: 16843794.6984628\ttotal: 6.21s\tremaining: 9.79s\n", + "388:\tlearn: 16843670.2191430\ttotal: 6.22s\tremaining: 9.77s\n", + "389:\tlearn: 16833049.2556840\ttotal: 6.24s\tremaining: 9.76s\n", + "390:\tlearn: 16821522.4443567\ttotal: 6.26s\tremaining: 9.74s\n", + "391:\tlearn: 16818181.1766856\ttotal: 6.27s\tremaining: 9.73s\n", + "392:\tlearn: 16817749.5049150\ttotal: 6.29s\tremaining: 9.72s\n", + "393:\tlearn: 16817402.3614282\ttotal: 6.31s\tremaining: 9.71s\n", + "394:\tlearn: 16815679.7151727\ttotal: 6.33s\tremaining: 9.7s\n", + "395:\tlearn: 16810641.8717564\ttotal: 6.35s\tremaining: 9.68s\n", + "396:\tlearn: 16810291.1871768\ttotal: 6.36s\tremaining: 9.67s\n", + "397:\tlearn: 16808056.2422004\ttotal: 6.39s\tremaining: 9.67s\n", + "398:\tlearn: 16807804.2454334\ttotal: 6.42s\tremaining: 9.68s\n", + "399:\tlearn: 16799998.1957230\ttotal: 6.44s\tremaining: 9.66s\n", + "400:\tlearn: 16799220.2656080\ttotal: 6.46s\tremaining: 9.65s\n", + "401:\tlearn: 16798913.0252067\ttotal: 6.47s\tremaining: 9.63s\n", + "402:\tlearn: 16798319.2545577\ttotal: 6.49s\tremaining: 9.62s\n", + "403:\tlearn: 16796848.5752647\ttotal: 6.51s\tremaining: 9.6s\n", + "404:\tlearn: 16757656.8985529\ttotal: 6.52s\tremaining: 9.58s\n", + "405:\tlearn: 16745513.4381725\ttotal: 6.54s\tremaining: 9.56s\n", + "406:\tlearn: 16735416.8114581\ttotal: 6.55s\tremaining: 9.54s\n", + "407:\tlearn: 16734295.1424370\ttotal: 6.57s\tremaining: 9.53s\n", + "408:\tlearn: 16733140.3781664\ttotal: 6.58s\tremaining: 9.52s\n", + "409:\tlearn: 16723800.8980695\ttotal: 6.6s\tremaining: 9.5s\n", + "410:\tlearn: 16721200.9625357\ttotal: 6.62s\tremaining: 9.48s\n", + "411:\tlearn: 16720027.8472987\ttotal: 6.63s\tremaining: 9.46s\n", + "412:\tlearn: 16717199.5760035\ttotal: 6.65s\tremaining: 9.45s\n", + "413:\tlearn: 16713362.4492616\ttotal: 6.66s\tremaining: 9.43s\n", + "414:\tlearn: 16712806.0473182\ttotal: 6.67s\tremaining: 9.41s\n", + "415:\tlearn: 16711241.9902750\ttotal: 6.69s\tremaining: 9.39s\n", + "416:\tlearn: 16710626.7325455\ttotal: 6.71s\tremaining: 9.37s\n", + "417:\tlearn: 16644768.4542531\ttotal: 6.72s\tremaining: 9.36s\n", + "418:\tlearn: 16644403.8081224\ttotal: 6.74s\tremaining: 9.34s\n", + "419:\tlearn: 16644106.9601552\ttotal: 6.75s\tremaining: 9.32s\n", + "420:\tlearn: 16643628.6346956\ttotal: 6.77s\tremaining: 9.31s\n", + "421:\tlearn: 16640073.3813320\ttotal: 6.79s\tremaining: 9.29s\n", + "422:\tlearn: 16639549.7950808\ttotal: 6.8s\tremaining: 9.28s\n", + "423:\tlearn: 16639069.1006878\ttotal: 6.81s\tremaining: 9.25s\n", + "424:\tlearn: 16638481.2382327\ttotal: 6.82s\tremaining: 9.23s\n", + "425:\tlearn: 16638208.9073863\ttotal: 6.83s\tremaining: 9.21s\n", + "426:\tlearn: 16609090.9227109\ttotal: 6.85s\tremaining: 9.19s\n", + "427:\tlearn: 16607897.8537223\ttotal: 6.86s\tremaining: 9.17s\n", + "428:\tlearn: 16607613.0069443\ttotal: 6.88s\tremaining: 9.15s\n", + "429:\tlearn: 16603866.7848843\ttotal: 6.89s\tremaining: 9.13s\n", + "430:\tlearn: 16566652.4020620\ttotal: 6.91s\tremaining: 9.12s\n", + "431:\tlearn: 16566149.6048169\ttotal: 6.92s\tremaining: 9.1s\n", + "432:\tlearn: 16564672.1011733\ttotal: 6.93s\tremaining: 9.08s\n", + "433:\tlearn: 16564610.7741058\ttotal: 6.95s\tremaining: 9.06s\n", + "434:\tlearn: 16564198.8911273\ttotal: 6.96s\tremaining: 9.04s\n", + "435:\tlearn: 16559675.2968062\ttotal: 6.97s\tremaining: 9.02s\n", + "436:\tlearn: 16558753.2346339\ttotal: 6.99s\tremaining: 9s\n", + "437:\tlearn: 16558452.1907641\ttotal: 7s\tremaining: 8.98s\n", + "438:\tlearn: 16546587.2383006\ttotal: 7.01s\tremaining: 8.96s\n", + "439:\tlearn: 16543823.0847287\ttotal: 7.03s\tremaining: 8.95s\n", + "440:\tlearn: 16542126.8424469\ttotal: 7.04s\tremaining: 8.93s\n", + "441:\tlearn: 16541624.1632076\ttotal: 7.06s\tremaining: 8.91s\n", + "442:\tlearn: 16540326.5322872\ttotal: 7.07s\tremaining: 8.89s\n", + "443:\tlearn: 16530336.2084291\ttotal: 7.08s\tremaining: 8.87s\n", + "444:\tlearn: 16530167.9665629\ttotal: 7.1s\tremaining: 8.85s\n", + "445:\tlearn: 16528821.2477933\ttotal: 7.12s\tremaining: 8.84s\n", + "446:\tlearn: 16528766.2012617\ttotal: 7.13s\tremaining: 8.82s\n", + "447:\tlearn: 16518018.7193100\ttotal: 7.14s\tremaining: 8.8s\n", + "448:\tlearn: 16508723.6897544\ttotal: 7.16s\tremaining: 8.79s\n", + "449:\tlearn: 16508487.2637814\ttotal: 7.17s\tremaining: 8.77s\n", + "450:\tlearn: 16473955.6540161\ttotal: 7.19s\tremaining: 8.75s\n", + "451:\tlearn: 16453172.0203944\ttotal: 7.2s\tremaining: 8.73s\n", + "452:\tlearn: 16451483.6324413\ttotal: 7.21s\tremaining: 8.71s\n", + "453:\tlearn: 16451257.8036014\ttotal: 7.23s\tremaining: 8.69s\n", + "454:\tlearn: 16448369.9508352\ttotal: 7.24s\tremaining: 8.67s\n", + "455:\tlearn: 16446719.1385193\ttotal: 7.25s\tremaining: 8.65s\n", + "456:\tlearn: 16420736.1369659\ttotal: 7.27s\tremaining: 8.64s\n", + "457:\tlearn: 16420629.2824606\ttotal: 7.28s\tremaining: 8.62s\n", + "458:\tlearn: 16420336.6729748\ttotal: 7.29s\tremaining: 8.6s\n", + "459:\tlearn: 16420155.4584530\ttotal: 7.3s\tremaining: 8.57s\n", + "460:\tlearn: 16419734.8233202\ttotal: 7.32s\tremaining: 8.55s\n", + "461:\tlearn: 16419517.6225944\ttotal: 7.33s\tremaining: 8.53s\n", + "462:\tlearn: 16406145.7183320\ttotal: 7.34s\tremaining: 8.51s\n", + "463:\tlearn: 16404609.0651931\ttotal: 7.35s\tremaining: 8.49s\n", + "464:\tlearn: 16404332.0732862\ttotal: 7.36s\tremaining: 8.47s\n", + "465:\tlearn: 16404019.7507952\ttotal: 7.38s\tremaining: 8.45s\n", + "466:\tlearn: 16403507.0137349\ttotal: 7.39s\tremaining: 8.43s\n", + "467:\tlearn: 16402993.5886996\ttotal: 7.4s\tremaining: 8.41s\n", + "468:\tlearn: 16385955.8460101\ttotal: 7.42s\tremaining: 8.4s\n", + "469:\tlearn: 16373237.2004642\ttotal: 7.43s\tremaining: 8.38s\n", + "470:\tlearn: 16373038.3665164\ttotal: 7.44s\tremaining: 8.36s\n", + "471:\tlearn: 16372801.5860356\ttotal: 7.46s\tremaining: 8.35s\n", + "472:\tlearn: 16360759.6605520\ttotal: 7.48s\tremaining: 8.33s\n", + "473:\tlearn: 16360169.9657388\ttotal: 7.5s\tremaining: 8.32s\n", + "474:\tlearn: 16351841.0373273\ttotal: 7.52s\tremaining: 8.31s\n", + "475:\tlearn: 16349809.4004009\ttotal: 7.53s\tremaining: 8.29s\n", + "476:\tlearn: 16344483.1074475\ttotal: 7.55s\tremaining: 8.28s\n", + "477:\tlearn: 16340922.7262468\ttotal: 7.56s\tremaining: 8.26s\n", + "478:\tlearn: 16334736.4373107\ttotal: 7.58s\tremaining: 8.25s\n", + "479:\tlearn: 16334043.7402281\ttotal: 7.6s\tremaining: 8.23s\n", + "480:\tlearn: 16333745.0129155\ttotal: 7.62s\tremaining: 8.22s\n", + "481:\tlearn: 16332170.0024156\ttotal: 7.63s\tremaining: 8.2s\n", + "482:\tlearn: 16331680.4256261\ttotal: 7.64s\tremaining: 8.18s\n", + "483:\tlearn: 16321943.5880137\ttotal: 7.66s\tremaining: 8.16s\n", + "484:\tlearn: 16313566.1128530\ttotal: 7.67s\tremaining: 8.15s\n", + "485:\tlearn: 16312784.3783495\ttotal: 7.69s\tremaining: 8.13s\n", + "486:\tlearn: 16304256.8971602\ttotal: 7.7s\tremaining: 8.11s\n", + "487:\tlearn: 16299338.9360929\ttotal: 7.72s\tremaining: 8.1s\n", + "488:\tlearn: 16298399.2768748\ttotal: 7.73s\tremaining: 8.08s\n", + "489:\tlearn: 16282861.5959599\ttotal: 7.75s\tremaining: 8.06s\n", + "490:\tlearn: 16278027.7798172\ttotal: 7.76s\tremaining: 8.04s\n", + "491:\tlearn: 16262455.7433251\ttotal: 7.77s\tremaining: 8.03s\n", + "492:\tlearn: 16254609.6670435\ttotal: 7.79s\tremaining: 8.01s\n", + "493:\tlearn: 16250306.9197526\ttotal: 7.8s\tremaining: 7.99s\n", + "494:\tlearn: 16249855.9315045\ttotal: 7.81s\tremaining: 7.97s\n", + "495:\tlearn: 16248555.7562997\ttotal: 7.83s\tremaining: 7.95s\n", + "496:\tlearn: 16247555.1566330\ttotal: 7.84s\tremaining: 7.93s\n", + "497:\tlearn: 16247235.5993966\ttotal: 7.85s\tremaining: 7.91s\n", + "498:\tlearn: 16246264.7483105\ttotal: 7.86s\tremaining: 7.89s\n", + "499:\tlearn: 16246007.7491962\ttotal: 7.87s\tremaining: 7.87s\n", + "500:\tlearn: 16222867.6954421\ttotal: 7.88s\tremaining: 7.85s\n", + "501:\tlearn: 16222688.8853061\ttotal: 7.9s\tremaining: 7.83s\n", + "502:\tlearn: 16217885.3385915\ttotal: 7.91s\tremaining: 7.82s\n", + "503:\tlearn: 16217409.1580145\ttotal: 7.92s\tremaining: 7.79s\n", + "504:\tlearn: 16216838.3191240\ttotal: 7.93s\tremaining: 7.78s\n", + "505:\tlearn: 16216329.9777509\ttotal: 7.95s\tremaining: 7.76s\n", + "506:\tlearn: 16201534.4156055\ttotal: 7.96s\tremaining: 7.74s\n", + "507:\tlearn: 16198138.1904772\ttotal: 7.98s\tremaining: 7.72s\n", + "508:\tlearn: 16197904.2583705\ttotal: 7.99s\tremaining: 7.7s\n", + "509:\tlearn: 16193656.6407621\ttotal: 8s\tremaining: 7.69s\n", + "510:\tlearn: 16180805.8618897\ttotal: 8.01s\tremaining: 7.67s\n", + "511:\tlearn: 16176908.1769610\ttotal: 8.03s\tremaining: 7.65s\n", + "512:\tlearn: 16168261.0438871\ttotal: 8.04s\tremaining: 7.64s\n", + "513:\tlearn: 16167754.4165306\ttotal: 8.06s\tremaining: 7.62s\n", + "514:\tlearn: 16166295.0362243\ttotal: 8.08s\tremaining: 7.61s\n", + "515:\tlearn: 16166058.4053693\ttotal: 8.09s\tremaining: 7.59s\n", + "516:\tlearn: 16155412.7707338\ttotal: 8.11s\tremaining: 7.57s\n", + "517:\tlearn: 16152266.1742558\ttotal: 8.12s\tremaining: 7.56s\n", + "518:\tlearn: 16151552.8907870\ttotal: 8.14s\tremaining: 7.54s\n", + "519:\tlearn: 16140281.4351978\ttotal: 8.16s\tremaining: 7.53s\n", + "520:\tlearn: 16133450.4403783\ttotal: 8.18s\tremaining: 7.52s\n", + "521:\tlearn: 16132209.1334220\ttotal: 8.19s\tremaining: 7.5s\n", + "522:\tlearn: 16118104.6552795\ttotal: 8.21s\tremaining: 7.48s\n", + "523:\tlearn: 16108764.2393062\ttotal: 8.22s\tremaining: 7.47s\n", + "524:\tlearn: 16108234.0634605\ttotal: 8.24s\tremaining: 7.45s\n", + "525:\tlearn: 16107619.6760099\ttotal: 8.25s\tremaining: 7.44s\n", + "526:\tlearn: 16104870.7442280\ttotal: 8.27s\tremaining: 7.42s\n", + "527:\tlearn: 16102428.3934069\ttotal: 8.28s\tremaining: 7.4s\n", + "528:\tlearn: 16102157.2857565\ttotal: 8.29s\tremaining: 7.39s\n", + "529:\tlearn: 16101584.7403855\ttotal: 8.31s\tremaining: 7.37s\n", + "530:\tlearn: 16101480.3344969\ttotal: 8.32s\tremaining: 7.35s\n", + "531:\tlearn: 16100595.6548675\ttotal: 8.33s\tremaining: 7.33s\n", + "532:\tlearn: 16097511.0825233\ttotal: 8.35s\tremaining: 7.32s\n", + "533:\tlearn: 16096615.9743637\ttotal: 8.36s\tremaining: 7.3s\n", + "534:\tlearn: 16096369.6922988\ttotal: 8.37s\tremaining: 7.28s\n", + "535:\tlearn: 16095946.1647864\ttotal: 8.38s\tremaining: 7.26s\n", + "536:\tlearn: 16095637.6185090\ttotal: 8.4s\tremaining: 7.24s\n", + "537:\tlearn: 16094682.0243853\ttotal: 8.41s\tremaining: 7.22s\n", + "538:\tlearn: 16094291.9050311\ttotal: 8.42s\tremaining: 7.2s\n", + "539:\tlearn: 16093984.5280001\ttotal: 8.43s\tremaining: 7.18s\n", + "540:\tlearn: 16090374.6401334\ttotal: 8.45s\tremaining: 7.17s\n", + "541:\tlearn: 16090226.8772271\ttotal: 8.46s\tremaining: 7.15s\n", + "542:\tlearn: 16090050.1805201\ttotal: 8.47s\tremaining: 7.13s\n", + "543:\tlearn: 16069181.1048944\ttotal: 8.48s\tremaining: 7.11s\n", + "544:\tlearn: 16068504.9399291\ttotal: 8.5s\tremaining: 7.09s\n", + "545:\tlearn: 16068245.3744393\ttotal: 8.51s\tremaining: 7.08s\n", + "546:\tlearn: 16065773.4114093\ttotal: 8.52s\tremaining: 7.06s\n", + "547:\tlearn: 16051662.5046318\ttotal: 8.54s\tremaining: 7.04s\n", + "548:\tlearn: 16035327.2446945\ttotal: 8.55s\tremaining: 7.02s\n", + "549:\tlearn: 16035199.2858857\ttotal: 8.56s\tremaining: 7s\n", + "550:\tlearn: 16033842.9666151\ttotal: 8.57s\tremaining: 6.99s\n", + "551:\tlearn: 15995073.4381976\ttotal: 8.59s\tremaining: 6.97s\n", + "552:\tlearn: 15994812.5505379\ttotal: 8.6s\tremaining: 6.95s\n", + "553:\tlearn: 15994595.9921031\ttotal: 8.61s\tremaining: 6.93s\n", + "554:\tlearn: 15992248.3834318\ttotal: 8.62s\tremaining: 6.92s\n", + "555:\tlearn: 15992027.4484601\ttotal: 8.64s\tremaining: 6.9s\n", + "556:\tlearn: 15990566.0719983\ttotal: 8.65s\tremaining: 6.88s\n", + "557:\tlearn: 15985609.0920187\ttotal: 8.66s\tremaining: 6.86s\n", + "558:\tlearn: 15984517.8156083\ttotal: 8.68s\tremaining: 6.84s\n", + "559:\tlearn: 15958775.9803743\ttotal: 8.69s\tremaining: 6.83s\n", + "560:\tlearn: 15958166.8639855\ttotal: 8.7s\tremaining: 6.81s\n", + "561:\tlearn: 15949224.5334582\ttotal: 8.71s\tremaining: 6.79s\n", + "562:\tlearn: 15948769.9101270\ttotal: 8.73s\tremaining: 6.77s\n", + "563:\tlearn: 15930009.9576761\ttotal: 8.74s\tremaining: 6.76s\n", + "564:\tlearn: 15917439.6202170\ttotal: 8.75s\tremaining: 6.74s\n", + "565:\tlearn: 15908669.4567536\ttotal: 8.77s\tremaining: 6.72s\n", + "566:\tlearn: 15908084.2939630\ttotal: 8.78s\tremaining: 6.7s\n", + "567:\tlearn: 15906697.1590494\ttotal: 8.79s\tremaining: 6.69s\n", + "568:\tlearn: 15906522.4609846\ttotal: 8.8s\tremaining: 6.67s\n", + "569:\tlearn: 15906139.9138507\ttotal: 8.82s\tremaining: 6.65s\n", + "570:\tlearn: 15905855.0642382\ttotal: 8.83s\tremaining: 6.63s\n", + "571:\tlearn: 15897372.3501416\ttotal: 8.84s\tremaining: 6.62s\n", + "572:\tlearn: 15893536.4661240\ttotal: 8.86s\tremaining: 6.6s\n", + "573:\tlearn: 15893206.2810918\ttotal: 8.87s\tremaining: 6.58s\n", + "574:\tlearn: 15892918.2703602\ttotal: 8.88s\tremaining: 6.56s\n", + "575:\tlearn: 15892752.8029869\ttotal: 8.89s\tremaining: 6.55s\n", + "576:\tlearn: 15885169.7413434\ttotal: 8.91s\tremaining: 6.53s\n", + "577:\tlearn: 15884936.8745209\ttotal: 8.92s\tremaining: 6.51s\n", + "578:\tlearn: 15876877.1991641\ttotal: 8.93s\tremaining: 6.49s\n", + "579:\tlearn: 15865774.4061534\ttotal: 8.94s\tremaining: 6.48s\n", + "580:\tlearn: 15859212.9207966\ttotal: 8.96s\tremaining: 6.46s\n", + "581:\tlearn: 15858807.6511813\ttotal: 8.97s\tremaining: 6.44s\n", + "582:\tlearn: 15850129.9468116\ttotal: 8.98s\tremaining: 6.42s\n", + "583:\tlearn: 15845554.5689368\ttotal: 8.99s\tremaining: 6.41s\n", + "584:\tlearn: 15844986.6765475\ttotal: 9.01s\tremaining: 6.39s\n", + "585:\tlearn: 15844796.1180439\ttotal: 9.02s\tremaining: 6.37s\n", + "586:\tlearn: 15844586.1630771\ttotal: 9.03s\tremaining: 6.36s\n", + "587:\tlearn: 15827685.4584540\ttotal: 9.05s\tremaining: 6.34s\n", + "588:\tlearn: 15826910.6044821\ttotal: 9.06s\tremaining: 6.32s\n", + "589:\tlearn: 15824060.9875073\ttotal: 9.07s\tremaining: 6.3s\n", + "590:\tlearn: 15818523.6912985\ttotal: 9.09s\tremaining: 6.29s\n", + "591:\tlearn: 15810640.6921394\ttotal: 9.1s\tremaining: 6.27s\n", + "592:\tlearn: 15795481.4197185\ttotal: 9.12s\tremaining: 6.26s\n", + "593:\tlearn: 15795256.4491006\ttotal: 9.13s\tremaining: 6.24s\n", + "594:\tlearn: 15784420.7363473\ttotal: 9.14s\tremaining: 6.22s\n", + "595:\tlearn: 15784290.1819258\ttotal: 9.15s\tremaining: 6.21s\n", + "596:\tlearn: 15783955.0773924\ttotal: 9.17s\tremaining: 6.19s\n", + "597:\tlearn: 15781518.5372107\ttotal: 9.18s\tremaining: 6.17s\n", + "598:\tlearn: 15779547.4210947\ttotal: 9.2s\tremaining: 6.16s\n", + "599:\tlearn: 15777334.3663340\ttotal: 9.21s\tremaining: 6.14s\n", + "600:\tlearn: 15774774.7721883\ttotal: 9.23s\tremaining: 6.13s\n", + "601:\tlearn: 15774672.2356339\ttotal: 9.24s\tremaining: 6.11s\n", + "602:\tlearn: 15773528.0736833\ttotal: 9.26s\tremaining: 6.09s\n", + "603:\tlearn: 15768721.3649454\ttotal: 9.27s\tremaining: 6.08s\n", + "604:\tlearn: 15768502.0877019\ttotal: 9.28s\tremaining: 6.06s\n", + "605:\tlearn: 15768057.4929247\ttotal: 9.3s\tremaining: 6.05s\n", + "606:\tlearn: 15767950.4285043\ttotal: 9.31s\tremaining: 6.03s\n", + "607:\tlearn: 15767445.1324607\ttotal: 9.32s\tremaining: 6.01s\n", + "608:\tlearn: 15767269.0628064\ttotal: 9.33s\tremaining: 5.99s\n", + "609:\tlearn: 15767020.8174624\ttotal: 9.35s\tremaining: 5.97s\n", + "610:\tlearn: 15762309.5160245\ttotal: 9.36s\tremaining: 5.96s\n", + "611:\tlearn: 15757527.7718093\ttotal: 9.37s\tremaining: 5.94s\n", + "612:\tlearn: 15757150.7731734\ttotal: 9.38s\tremaining: 5.92s\n", + "613:\tlearn: 15756885.6252756\ttotal: 9.39s\tremaining: 5.91s\n", + "614:\tlearn: 15755584.6816303\ttotal: 9.41s\tremaining: 5.89s\n", + "615:\tlearn: 15755485.6737331\ttotal: 9.42s\tremaining: 5.87s\n", + "616:\tlearn: 15754432.0517599\ttotal: 9.43s\tremaining: 5.86s\n", + "617:\tlearn: 15744535.8203508\ttotal: 9.45s\tremaining: 5.84s\n", + "618:\tlearn: 15740683.5538600\ttotal: 9.46s\tremaining: 5.82s\n", + "619:\tlearn: 15736903.5667213\ttotal: 9.47s\tremaining: 5.8s\n", + "620:\tlearn: 15736355.4210963\ttotal: 9.48s\tremaining: 5.79s\n", + "621:\tlearn: 15729940.4032081\ttotal: 9.5s\tremaining: 5.77s\n", + "622:\tlearn: 15729775.7542976\ttotal: 9.51s\tremaining: 5.75s\n", + "623:\tlearn: 15726578.4125003\ttotal: 9.52s\tremaining: 5.74s\n", + "624:\tlearn: 15713451.5317183\ttotal: 9.54s\tremaining: 5.72s\n", + "625:\tlearn: 15712116.7478338\ttotal: 9.55s\tremaining: 5.71s\n", + "626:\tlearn: 15712039.0336448\ttotal: 9.56s\tremaining: 5.69s\n", + "627:\tlearn: 15711687.4136682\ttotal: 9.57s\tremaining: 5.67s\n", + "628:\tlearn: 15711298.6681597\ttotal: 9.59s\tremaining: 5.66s\n", + "629:\tlearn: 15705228.0021081\ttotal: 9.6s\tremaining: 5.64s\n", + "630:\tlearn: 15705060.0247650\ttotal: 9.62s\tremaining: 5.62s\n", + "631:\tlearn: 15702513.1910574\ttotal: 9.63s\tremaining: 5.61s\n", + "632:\tlearn: 15702203.6145508\ttotal: 9.64s\tremaining: 5.59s\n", + "633:\tlearn: 15698975.2951288\ttotal: 9.65s\tremaining: 5.57s\n", + "634:\tlearn: 15694674.2341421\ttotal: 9.67s\tremaining: 5.56s\n", + "635:\tlearn: 15686913.9104937\ttotal: 9.68s\tremaining: 5.54s\n", + "636:\tlearn: 15686809.9586513\ttotal: 9.7s\tremaining: 5.53s\n", + "637:\tlearn: 15685604.3787689\ttotal: 9.71s\tremaining: 5.51s\n", + "638:\tlearn: 15685081.4917552\ttotal: 9.72s\tremaining: 5.49s\n", + "639:\tlearn: 15676541.2826685\ttotal: 9.74s\tremaining: 5.48s\n", + "640:\tlearn: 15672855.0180760\ttotal: 9.75s\tremaining: 5.46s\n", + "641:\tlearn: 15666780.7593096\ttotal: 9.77s\tremaining: 5.45s\n", + "642:\tlearn: 15659438.3508408\ttotal: 9.78s\tremaining: 5.43s\n", + "643:\tlearn: 15653755.4598701\ttotal: 9.8s\tremaining: 5.42s\n", + "644:\tlearn: 15652034.4638985\ttotal: 9.81s\tremaining: 5.4s\n", + "645:\tlearn: 15645095.6489597\ttotal: 9.82s\tremaining: 5.38s\n", + "646:\tlearn: 15641226.9420905\ttotal: 9.84s\tremaining: 5.37s\n", + "647:\tlearn: 15639833.9184524\ttotal: 9.85s\tremaining: 5.35s\n", + "648:\tlearn: 15639581.1651510\ttotal: 9.87s\tremaining: 5.34s\n", + "649:\tlearn: 15635923.3848062\ttotal: 9.88s\tremaining: 5.32s\n", + "650:\tlearn: 15635813.1152459\ttotal: 9.89s\tremaining: 5.3s\n", + "651:\tlearn: 15635469.3555938\ttotal: 9.9s\tremaining: 5.29s\n", + "652:\tlearn: 15635363.9174910\ttotal: 9.92s\tremaining: 5.27s\n", + "653:\tlearn: 15633936.7433448\ttotal: 9.93s\tremaining: 5.25s\n", + "654:\tlearn: 15633839.2271448\ttotal: 9.94s\tremaining: 5.24s\n", + "655:\tlearn: 15633735.8610291\ttotal: 9.96s\tremaining: 5.22s\n", + "656:\tlearn: 15633309.8063070\ttotal: 9.97s\tremaining: 5.2s\n", + "657:\tlearn: 15632683.8986677\ttotal: 9.98s\tremaining: 5.19s\n", + "658:\tlearn: 15632461.2639014\ttotal: 9.99s\tremaining: 5.17s\n", + "659:\tlearn: 15627123.1765533\ttotal: 10s\tremaining: 5.16s\n", + "660:\tlearn: 15626996.0787558\ttotal: 10s\tremaining: 5.14s\n", + "661:\tlearn: 15624291.0204091\ttotal: 10s\tremaining: 5.12s\n", + "662:\tlearn: 15617684.3098363\ttotal: 10s\tremaining: 5.11s\n", + "663:\tlearn: 15611967.2176796\ttotal: 10.1s\tremaining: 5.09s\n", + "664:\tlearn: 15598472.2546786\ttotal: 10.1s\tremaining: 5.07s\n", + "665:\tlearn: 15597526.0470563\ttotal: 10.1s\tremaining: 5.06s\n", + "666:\tlearn: 15597430.3920481\ttotal: 10.1s\tremaining: 5.04s\n", + "667:\tlearn: 15596422.7059295\ttotal: 10.1s\tremaining: 5.03s\n", + "668:\tlearn: 15591400.2242411\ttotal: 10.1s\tremaining: 5.01s\n", + "669:\tlearn: 15585199.5277811\ttotal: 10.1s\tremaining: 4.99s\n", + "670:\tlearn: 15585003.5063693\ttotal: 10.2s\tremaining: 4.98s\n", + "671:\tlearn: 15578765.7193891\ttotal: 10.2s\tremaining: 4.96s\n", + "672:\tlearn: 15577252.0151364\ttotal: 10.2s\tremaining: 4.94s\n", + "673:\tlearn: 15576511.8797514\ttotal: 10.2s\tremaining: 4.93s\n", + "674:\tlearn: 15576120.7606092\ttotal: 10.2s\tremaining: 4.91s\n", + "675:\tlearn: 15574398.5273782\ttotal: 10.2s\tremaining: 4.89s\n", + "676:\tlearn: 15565660.7493905\ttotal: 10.2s\tremaining: 4.88s\n", + "677:\tlearn: 15561009.3437211\ttotal: 10.2s\tremaining: 4.86s\n", + "678:\tlearn: 15548878.4770401\ttotal: 10.3s\tremaining: 4.85s\n", + "679:\tlearn: 15527713.9632219\ttotal: 10.3s\tremaining: 4.83s\n", + "680:\tlearn: 15519745.2151864\ttotal: 10.3s\tremaining: 4.82s\n", + "681:\tlearn: 15519391.2760902\ttotal: 10.3s\tremaining: 4.8s\n", + "682:\tlearn: 15514461.8611265\ttotal: 10.3s\tremaining: 4.78s\n", + "683:\tlearn: 15514296.1001141\ttotal: 10.3s\tremaining: 4.77s\n", + "684:\tlearn: 15514204.8658979\ttotal: 10.3s\tremaining: 4.75s\n", + "685:\tlearn: 15513977.3554214\ttotal: 10.3s\tremaining: 4.74s\n", + "686:\tlearn: 15513906.5046745\ttotal: 10.4s\tremaining: 4.72s\n", + "687:\tlearn: 15513701.8112778\ttotal: 10.4s\tremaining: 4.7s\n", + "688:\tlearn: 15513602.4959013\ttotal: 10.4s\tremaining: 4.69s\n", + "689:\tlearn: 15513510.7910896\ttotal: 10.4s\tremaining: 4.67s\n", + "690:\tlearn: 15513352.2070048\ttotal: 10.4s\tremaining: 4.65s\n", + "691:\tlearn: 15513238.8204588\ttotal: 10.4s\tremaining: 4.64s\n", + "692:\tlearn: 15513154.0618557\ttotal: 10.4s\tremaining: 4.62s\n", + "693:\tlearn: 15512878.9114412\ttotal: 10.4s\tremaining: 4.6s\n", + "694:\tlearn: 15509248.2055515\ttotal: 10.5s\tremaining: 4.59s\n", + "695:\tlearn: 15508734.7327170\ttotal: 10.5s\tremaining: 4.57s\n", + "696:\tlearn: 15508495.7881550\ttotal: 10.5s\tremaining: 4.55s\n", + "697:\tlearn: 15508349.6872134\ttotal: 10.5s\tremaining: 4.54s\n", + "698:\tlearn: 15508190.6588965\ttotal: 10.5s\tremaining: 4.52s\n", + "699:\tlearn: 15508018.8419773\ttotal: 10.5s\tremaining: 4.5s\n", + "700:\tlearn: 15507826.3791202\ttotal: 10.5s\tremaining: 4.49s\n", + "701:\tlearn: 15507705.8100928\ttotal: 10.5s\tremaining: 4.47s\n", + "702:\tlearn: 15507533.3512682\ttotal: 10.5s\tremaining: 4.46s\n", + "703:\tlearn: 15501571.2913355\ttotal: 10.6s\tremaining: 4.44s\n", + "704:\tlearn: 15495921.0773672\ttotal: 10.6s\tremaining: 4.42s\n", + "705:\tlearn: 15495385.0875416\ttotal: 10.6s\tremaining: 4.41s\n", + "706:\tlearn: 15495191.8032918\ttotal: 10.6s\tremaining: 4.39s\n", + "707:\tlearn: 15494128.9589635\ttotal: 10.6s\tremaining: 4.38s\n", + "708:\tlearn: 15493806.9566177\ttotal: 10.6s\tremaining: 4.36s\n", + "709:\tlearn: 15493694.0465547\ttotal: 10.6s\tremaining: 4.34s\n", + "710:\tlearn: 15493305.1729869\ttotal: 10.6s\tremaining: 4.33s\n", + "711:\tlearn: 15487948.0399475\ttotal: 10.7s\tremaining: 4.31s\n", + "712:\tlearn: 15487843.0916850\ttotal: 10.7s\tremaining: 4.29s\n", + "713:\tlearn: 15482765.7669785\ttotal: 10.7s\tremaining: 4.28s\n", + "714:\tlearn: 15474767.3580796\ttotal: 10.7s\tremaining: 4.26s\n", + "715:\tlearn: 15472407.8166003\ttotal: 10.7s\tremaining: 4.25s\n", + "716:\tlearn: 15467592.3874842\ttotal: 10.7s\tremaining: 4.23s\n", + "717:\tlearn: 15467435.4901525\ttotal: 10.7s\tremaining: 4.21s\n", + "718:\tlearn: 15462871.2869120\ttotal: 10.7s\tremaining: 4.2s\n", + "719:\tlearn: 15462771.0380185\ttotal: 10.8s\tremaining: 4.18s\n", + "720:\tlearn: 15462475.2715024\ttotal: 10.8s\tremaining: 4.17s\n", + "721:\tlearn: 15454885.7938423\ttotal: 10.8s\tremaining: 4.15s\n", + "722:\tlearn: 15450557.5824215\ttotal: 10.8s\tremaining: 4.13s\n", + "723:\tlearn: 15446455.2317749\ttotal: 10.8s\tremaining: 4.12s\n", + "724:\tlearn: 15445711.0004476\ttotal: 10.8s\tremaining: 4.1s\n", + "725:\tlearn: 15441822.5331613\ttotal: 10.8s\tremaining: 4.09s\n", + "726:\tlearn: 15441182.6843715\ttotal: 10.8s\tremaining: 4.07s\n", + "727:\tlearn: 15441088.8915881\ttotal: 10.9s\tremaining: 4.06s\n", + "728:\tlearn: 15441002.5272406\ttotal: 10.9s\tremaining: 4.04s\n", + "729:\tlearn: 15440884.2830869\ttotal: 10.9s\tremaining: 4.02s\n", + "730:\tlearn: 15440734.2579995\ttotal: 10.9s\tremaining: 4.01s\n", + "731:\tlearn: 15440611.8887909\ttotal: 10.9s\tremaining: 3.99s\n", + "732:\tlearn: 15440249.4221271\ttotal: 10.9s\tremaining: 3.98s\n", + "733:\tlearn: 15440158.8154476\ttotal: 10.9s\tremaining: 3.96s\n", + "734:\tlearn: 15436472.7845071\ttotal: 10.9s\tremaining: 3.94s\n", + "735:\tlearn: 15433672.4484876\ttotal: 11s\tremaining: 3.93s\n", + "736:\tlearn: 15433490.1840146\ttotal: 11s\tremaining: 3.91s\n", + "737:\tlearn: 15433308.6881010\ttotal: 11s\tremaining: 3.9s\n", + "738:\tlearn: 15433042.7409848\ttotal: 11s\tremaining: 3.88s\n", + "739:\tlearn: 15432541.8769518\ttotal: 11s\tremaining: 3.87s\n", + "740:\tlearn: 15431856.6761047\ttotal: 11s\tremaining: 3.85s\n", + "741:\tlearn: 15431804.7359345\ttotal: 11s\tremaining: 3.83s\n", + "742:\tlearn: 15427355.2392047\ttotal: 11s\tremaining: 3.82s\n", + "743:\tlearn: 15427218.1028185\ttotal: 11.1s\tremaining: 3.8s\n", + "744:\tlearn: 15424332.1093472\ttotal: 11.1s\tremaining: 3.79s\n", + "745:\tlearn: 15388321.8033125\ttotal: 11.1s\tremaining: 3.77s\n", + "746:\tlearn: 15377267.9048803\ttotal: 11.1s\tremaining: 3.76s\n", + "747:\tlearn: 15374625.6198420\ttotal: 11.1s\tremaining: 3.74s\n", + "748:\tlearn: 15370386.0426691\ttotal: 11.1s\tremaining: 3.73s\n", + "749:\tlearn: 15359901.4299089\ttotal: 11.1s\tremaining: 3.71s\n", + "750:\tlearn: 15358774.7332579\ttotal: 11.1s\tremaining: 3.69s\n", + "751:\tlearn: 15358651.8711020\ttotal: 11.2s\tremaining: 3.68s\n", + "752:\tlearn: 15358300.8764559\ttotal: 11.2s\tremaining: 3.66s\n", + "753:\tlearn: 15357884.8170886\ttotal: 11.2s\tremaining: 3.65s\n", + "754:\tlearn: 15357643.0994172\ttotal: 11.2s\tremaining: 3.63s\n", + "755:\tlearn: 15357565.0887636\ttotal: 11.2s\tremaining: 3.62s\n", + "756:\tlearn: 15351820.0777339\ttotal: 11.2s\tremaining: 3.6s\n", + "757:\tlearn: 15351414.2517094\ttotal: 11.2s\tremaining: 3.58s\n", + "758:\tlearn: 15349501.7532204\ttotal: 11.2s\tremaining: 3.57s\n", + "759:\tlearn: 15348526.7586048\ttotal: 11.3s\tremaining: 3.56s\n", + "760:\tlearn: 15348352.7244253\ttotal: 11.3s\tremaining: 3.54s\n", + "761:\tlearn: 15347292.4488773\ttotal: 11.3s\tremaining: 3.52s\n", + "762:\tlearn: 15347207.8865499\ttotal: 11.3s\tremaining: 3.51s\n", + "763:\tlearn: 15342959.4790246\ttotal: 11.3s\tremaining: 3.49s\n", + "764:\tlearn: 15342697.6483068\ttotal: 11.3s\tremaining: 3.48s\n", + "765:\tlearn: 15313028.1511576\ttotal: 11.3s\tremaining: 3.46s\n", + "766:\tlearn: 15310947.6382018\ttotal: 11.3s\tremaining: 3.45s\n", + "767:\tlearn: 15263340.9642127\ttotal: 11.4s\tremaining: 3.43s\n", + "768:\tlearn: 15259807.2026083\ttotal: 11.4s\tremaining: 3.42s\n", + "769:\tlearn: 15259504.1148296\ttotal: 11.4s\tremaining: 3.4s\n", + "770:\tlearn: 15259395.2637694\ttotal: 11.4s\tremaining: 3.39s\n", + "771:\tlearn: 15256045.7141942\ttotal: 11.4s\tremaining: 3.37s\n", + "772:\tlearn: 15252870.9021417\ttotal: 11.4s\tremaining: 3.36s\n", + "773:\tlearn: 15248430.3201074\ttotal: 11.4s\tremaining: 3.34s\n", + "774:\tlearn: 15246752.4177458\ttotal: 11.5s\tremaining: 3.33s\n", + "775:\tlearn: 15245960.1417687\ttotal: 11.5s\tremaining: 3.31s\n", + "776:\tlearn: 15245917.6645107\ttotal: 11.5s\tremaining: 3.3s\n", + "777:\tlearn: 15245817.5185452\ttotal: 11.5s\tremaining: 3.28s\n", + "778:\tlearn: 15245619.8351855\ttotal: 11.5s\tremaining: 3.27s\n", + "779:\tlearn: 15244869.5667520\ttotal: 11.5s\tremaining: 3.25s\n", + "780:\tlearn: 15244818.8943236\ttotal: 11.5s\tremaining: 3.24s\n", + "781:\tlearn: 15244254.3637038\ttotal: 11.6s\tremaining: 3.22s\n", + "782:\tlearn: 15243818.2939855\ttotal: 11.6s\tremaining: 3.21s\n", + "783:\tlearn: 15243668.1645179\ttotal: 11.6s\tremaining: 3.19s\n", + "784:\tlearn: 15240656.8617467\ttotal: 11.6s\tremaining: 3.17s\n", + "785:\tlearn: 15237802.6637690\ttotal: 11.6s\tremaining: 3.16s\n", + "786:\tlearn: 15235097.3769887\ttotal: 11.6s\tremaining: 3.14s\n", + "787:\tlearn: 15231063.4576018\ttotal: 11.6s\tremaining: 3.13s\n", + "788:\tlearn: 15224406.4239600\ttotal: 11.6s\tremaining: 3.11s\n", + "789:\tlearn: 15220791.2846445\ttotal: 11.7s\tremaining: 3.1s\n", + "790:\tlearn: 15220221.4180094\ttotal: 11.7s\tremaining: 3.08s\n", + "791:\tlearn: 15220121.7499013\ttotal: 11.7s\tremaining: 3.07s\n", + "792:\tlearn: 15218396.9325757\ttotal: 11.7s\tremaining: 3.05s\n", + "793:\tlearn: 15213830.8844557\ttotal: 11.7s\tremaining: 3.04s\n", + "794:\tlearn: 15212644.4009126\ttotal: 11.7s\tremaining: 3.02s\n", + "795:\tlearn: 15212570.5272286\ttotal: 11.7s\tremaining: 3.01s\n", + "796:\tlearn: 15172874.1552397\ttotal: 11.7s\tremaining: 2.99s\n", + "797:\tlearn: 15164671.3501787\ttotal: 11.8s\tremaining: 2.98s\n", + "798:\tlearn: 15162711.8871221\ttotal: 11.8s\tremaining: 2.96s\n", + "799:\tlearn: 15162618.0050229\ttotal: 11.8s\tremaining: 2.94s\n", + "800:\tlearn: 15161186.8924011\ttotal: 11.8s\tremaining: 2.93s\n", + "801:\tlearn: 15160994.4738412\ttotal: 11.8s\tremaining: 2.91s\n", + "802:\tlearn: 15159385.3831268\ttotal: 11.8s\tremaining: 2.9s\n", + "803:\tlearn: 15159166.1576231\ttotal: 11.8s\tremaining: 2.88s\n", + "804:\tlearn: 15156764.1801770\ttotal: 11.8s\tremaining: 2.87s\n", + "805:\tlearn: 15146691.8394282\ttotal: 11.9s\tremaining: 2.85s\n", + "806:\tlearn: 15146533.6706853\ttotal: 11.9s\tremaining: 2.84s\n", + "807:\tlearn: 15146408.7773292\ttotal: 11.9s\tremaining: 2.82s\n", + "808:\tlearn: 15142359.7678728\ttotal: 11.9s\tremaining: 2.81s\n", + "809:\tlearn: 15142322.1248825\ttotal: 11.9s\tremaining: 2.79s\n", + "810:\tlearn: 15132770.1153732\ttotal: 11.9s\tremaining: 2.78s\n", + "811:\tlearn: 15101480.9924963\ttotal: 11.9s\tremaining: 2.77s\n", + "812:\tlearn: 15101445.1875248\ttotal: 12s\tremaining: 2.75s\n", + "813:\tlearn: 15075376.4419388\ttotal: 12s\tremaining: 2.73s\n", + "814:\tlearn: 15073160.0820287\ttotal: 12s\tremaining: 2.72s\n", + "815:\tlearn: 15072725.5140996\ttotal: 12s\tremaining: 2.71s\n", + "816:\tlearn: 15072585.7018342\ttotal: 12s\tremaining: 2.69s\n", + "817:\tlearn: 15071522.0001919\ttotal: 12s\tremaining: 2.67s\n", + "818:\tlearn: 15071382.2097110\ttotal: 12s\tremaining: 2.66s\n", + "819:\tlearn: 15071301.2886091\ttotal: 12s\tremaining: 2.64s\n", + "820:\tlearn: 15071025.2992144\ttotal: 12.1s\tremaining: 2.63s\n", + "821:\tlearn: 15069498.2268762\ttotal: 12.1s\tremaining: 2.61s\n", + "822:\tlearn: 15061575.7065075\ttotal: 12.1s\tremaining: 2.6s\n", + "823:\tlearn: 15061416.4068476\ttotal: 12.1s\tremaining: 2.58s\n", + "824:\tlearn: 15060945.6687130\ttotal: 12.1s\tremaining: 2.57s\n", + "825:\tlearn: 15051099.8538783\ttotal: 12.1s\tremaining: 2.55s\n", + "826:\tlearn: 15050450.9663299\ttotal: 12.1s\tremaining: 2.54s\n", + "827:\tlearn: 15049722.9751983\ttotal: 12.2s\tremaining: 2.52s\n", + "828:\tlearn: 15049467.7452535\ttotal: 12.2s\tremaining: 2.51s\n", + "829:\tlearn: 15049412.7697933\ttotal: 12.2s\tremaining: 2.49s\n", + "830:\tlearn: 15048891.7740041\ttotal: 12.2s\tremaining: 2.48s\n", + "831:\tlearn: 15048043.0994998\ttotal: 12.2s\tremaining: 2.46s\n", + "832:\tlearn: 15046697.3368860\ttotal: 12.2s\tremaining: 2.45s\n", + "833:\tlearn: 15038272.8803419\ttotal: 12.2s\tremaining: 2.43s\n", + "834:\tlearn: 15034639.9951102\ttotal: 12.2s\tremaining: 2.42s\n", + "835:\tlearn: 15030153.8614245\ttotal: 12.3s\tremaining: 2.4s\n", + "836:\tlearn: 15027964.0190757\ttotal: 12.3s\tremaining: 2.39s\n", + "837:\tlearn: 15023890.1409211\ttotal: 12.3s\tremaining: 2.37s\n", + "838:\tlearn: 15022954.0613643\ttotal: 12.3s\tremaining: 2.36s\n", + "839:\tlearn: 15022653.7321874\ttotal: 12.3s\tremaining: 2.34s\n", + "840:\tlearn: 15021763.5899870\ttotal: 12.3s\tremaining: 2.33s\n", + "841:\tlearn: 15021552.9666208\ttotal: 12.3s\tremaining: 2.31s\n", + "842:\tlearn: 15017213.3112838\ttotal: 12.3s\tremaining: 2.3s\n", + "843:\tlearn: 15006868.9919636\ttotal: 12.4s\tremaining: 2.28s\n", + "844:\tlearn: 15006047.4873296\ttotal: 12.4s\tremaining: 2.27s\n", + "845:\tlearn: 15003167.3995596\ttotal: 12.4s\tremaining: 2.25s\n", + "846:\tlearn: 15001516.1719277\ttotal: 12.4s\tremaining: 2.24s\n", + "847:\tlearn: 15000023.6971343\ttotal: 12.4s\tremaining: 2.22s\n", + "848:\tlearn: 14996097.9901016\ttotal: 12.4s\tremaining: 2.21s\n", + "849:\tlearn: 14995809.9617414\ttotal: 12.4s\tremaining: 2.19s\n", + "850:\tlearn: 14991694.0680204\ttotal: 12.4s\tremaining: 2.18s\n", + "851:\tlearn: 14990806.5441048\ttotal: 12.5s\tremaining: 2.16s\n", + "852:\tlearn: 14990539.4146062\ttotal: 12.5s\tremaining: 2.15s\n", + "853:\tlearn: 14990428.7975864\ttotal: 12.5s\tremaining: 2.13s\n", + "854:\tlearn: 14989061.7567162\ttotal: 12.5s\tremaining: 2.12s\n", + "855:\tlearn: 14983131.0103419\ttotal: 12.5s\tremaining: 2.1s\n", + "856:\tlearn: 14982655.3316759\ttotal: 12.5s\tremaining: 2.09s\n", + "857:\tlearn: 14977099.0608273\ttotal: 12.5s\tremaining: 2.07s\n", + "858:\tlearn: 14976713.9058693\ttotal: 12.6s\tremaining: 2.06s\n", + "859:\tlearn: 14976613.3364184\ttotal: 12.6s\tremaining: 2.04s\n", + "860:\tlearn: 14964115.6999829\ttotal: 12.6s\tremaining: 2.03s\n", + "861:\tlearn: 14961152.7626425\ttotal: 12.6s\tremaining: 2.02s\n", + "862:\tlearn: 14960316.8698796\ttotal: 12.6s\tremaining: 2s\n", + "863:\tlearn: 14960206.5805103\ttotal: 12.6s\tremaining: 1.99s\n", + "864:\tlearn: 14948350.2065232\ttotal: 12.6s\tremaining: 1.97s\n", + "865:\tlearn: 14948237.8225238\ttotal: 12.6s\tremaining: 1.96s\n", + "866:\tlearn: 14948145.1280412\ttotal: 12.7s\tremaining: 1.94s\n", + "867:\tlearn: 14947479.9936319\ttotal: 12.7s\tremaining: 1.93s\n", + "868:\tlearn: 14946706.9144290\ttotal: 12.7s\tremaining: 1.91s\n", + "869:\tlearn: 14946008.4529886\ttotal: 12.7s\tremaining: 1.9s\n", + "870:\tlearn: 14938382.9733130\ttotal: 12.7s\tremaining: 1.88s\n", + "871:\tlearn: 14935923.0018589\ttotal: 12.7s\tremaining: 1.87s\n", + "872:\tlearn: 14935763.9386719\ttotal: 12.7s\tremaining: 1.85s\n", + "873:\tlearn: 14935390.9032799\ttotal: 12.7s\tremaining: 1.84s\n", + "874:\tlearn: 14924136.4999000\ttotal: 12.8s\tremaining: 1.82s\n", + "875:\tlearn: 14923231.4975181\ttotal: 12.8s\tremaining: 1.81s\n", + "876:\tlearn: 14920764.7489123\ttotal: 12.8s\tremaining: 1.79s\n", + "877:\tlearn: 14920619.6869935\ttotal: 12.8s\tremaining: 1.78s\n", + "878:\tlearn: 14920259.9887151\ttotal: 12.8s\tremaining: 1.76s\n", + "879:\tlearn: 14918671.6063618\ttotal: 12.8s\tremaining: 1.75s\n", + "880:\tlearn: 14909484.3446534\ttotal: 12.8s\tremaining: 1.73s\n", + "881:\tlearn: 14909331.3722806\ttotal: 12.8s\tremaining: 1.72s\n", + "882:\tlearn: 14909000.4744294\ttotal: 12.9s\tremaining: 1.7s\n", + "883:\tlearn: 14907810.4215534\ttotal: 12.9s\tremaining: 1.69s\n", + "884:\tlearn: 14907739.9399244\ttotal: 12.9s\tremaining: 1.68s\n", + "885:\tlearn: 14907643.8234156\ttotal: 12.9s\tremaining: 1.66s\n", + "886:\tlearn: 14902234.6414918\ttotal: 12.9s\tremaining: 1.65s\n", + "887:\tlearn: 14899639.1808572\ttotal: 12.9s\tremaining: 1.63s\n", + "888:\tlearn: 14898572.6302420\ttotal: 12.9s\tremaining: 1.62s\n", + "889:\tlearn: 14898481.9011232\ttotal: 13s\tremaining: 1.6s\n", + "890:\tlearn: 14898396.0569341\ttotal: 13s\tremaining: 1.59s\n", + "891:\tlearn: 14897825.8306216\ttotal: 13s\tremaining: 1.57s\n", + "892:\tlearn: 14897667.5383103\ttotal: 13s\tremaining: 1.56s\n", + "893:\tlearn: 14894977.2967186\ttotal: 13s\tremaining: 1.54s\n", + "894:\tlearn: 14894451.1015405\ttotal: 13s\tremaining: 1.53s\n", + "895:\tlearn: 14894302.6357933\ttotal: 13s\tremaining: 1.51s\n", + "896:\tlearn: 14892042.1460828\ttotal: 13s\tremaining: 1.5s\n", + "897:\tlearn: 14891913.2181914\ttotal: 13.1s\tremaining: 1.48s\n", + "898:\tlearn: 14891548.8875063\ttotal: 13.1s\tremaining: 1.47s\n", + "899:\tlearn: 14891535.2806629\ttotal: 13.1s\tremaining: 1.45s\n", + "900:\tlearn: 14885531.4554658\ttotal: 13.1s\tremaining: 1.44s\n", + "901:\tlearn: 14885300.5490787\ttotal: 13.1s\tremaining: 1.42s\n", + "902:\tlearn: 14882237.8064495\ttotal: 13.1s\tremaining: 1.41s\n", + "903:\tlearn: 14871534.0201501\ttotal: 13.1s\tremaining: 1.39s\n", + "904:\tlearn: 14870231.5883229\ttotal: 13.1s\tremaining: 1.38s\n", + "905:\tlearn: 14870176.1224648\ttotal: 13.2s\tremaining: 1.36s\n", + "906:\tlearn: 14869680.0980501\ttotal: 13.2s\tremaining: 1.35s\n", + "907:\tlearn: 14869457.0180442\ttotal: 13.2s\tremaining: 1.33s\n", + "908:\tlearn: 14868933.3692698\ttotal: 13.2s\tremaining: 1.32s\n", + "909:\tlearn: 14863961.6816683\ttotal: 13.2s\tremaining: 1.31s\n", + "910:\tlearn: 14863914.5911026\ttotal: 13.2s\tremaining: 1.29s\n", + "911:\tlearn: 14855476.3946477\ttotal: 13.2s\tremaining: 1.28s\n", + "912:\tlearn: 14855166.2145584\ttotal: 13.3s\tremaining: 1.26s\n", + "913:\tlearn: 14850934.1124148\ttotal: 13.3s\tremaining: 1.25s\n", + "914:\tlearn: 14846483.1667184\ttotal: 13.3s\tremaining: 1.23s\n", + "915:\tlearn: 14837736.3329095\ttotal: 13.3s\tremaining: 1.22s\n", + "916:\tlearn: 14837595.6179847\ttotal: 13.3s\tremaining: 1.2s\n", + "917:\tlearn: 14830817.0386636\ttotal: 13.3s\tremaining: 1.19s\n", + "918:\tlearn: 14830573.0969752\ttotal: 13.3s\tremaining: 1.18s\n", + "919:\tlearn: 14830470.0025192\ttotal: 13.3s\tremaining: 1.16s\n", + "920:\tlearn: 14829815.6285131\ttotal: 13.4s\tremaining: 1.15s\n", + "921:\tlearn: 14825434.2189552\ttotal: 13.4s\tremaining: 1.13s\n", + "922:\tlearn: 14819445.0181126\ttotal: 13.4s\tremaining: 1.12s\n", + "923:\tlearn: 14819398.6279561\ttotal: 13.4s\tremaining: 1.1s\n", + "924:\tlearn: 14818813.5923928\ttotal: 13.4s\tremaining: 1.09s\n", + "925:\tlearn: 14815956.9177135\ttotal: 13.4s\tremaining: 1.07s\n", + "926:\tlearn: 14815788.2679741\ttotal: 13.4s\tremaining: 1.06s\n", + "927:\tlearn: 14815448.9260298\ttotal: 13.5s\tremaining: 1.04s\n", + "928:\tlearn: 14815074.5372959\ttotal: 13.5s\tremaining: 1.03s\n", + "929:\tlearn: 14814229.2585638\ttotal: 13.5s\tremaining: 1.01s\n", + "930:\tlearn: 14804039.0241152\ttotal: 13.5s\tremaining: 1s\n", + "931:\tlearn: 14803929.0738285\ttotal: 13.5s\tremaining: 987ms\n", + "932:\tlearn: 14803822.0317935\ttotal: 13.5s\tremaining: 972ms\n", + "933:\tlearn: 14802847.7620639\ttotal: 13.6s\tremaining: 958ms\n", + "934:\tlearn: 14802644.0143811\ttotal: 13.6s\tremaining: 944ms\n", + "935:\tlearn: 14801390.4240818\ttotal: 13.6s\tremaining: 929ms\n", + "936:\tlearn: 14801273.4225706\ttotal: 13.6s\tremaining: 915ms\n", + "937:\tlearn: 14801021.8599058\ttotal: 13.6s\tremaining: 900ms\n", + "938:\tlearn: 14800715.3029627\ttotal: 13.6s\tremaining: 885ms\n", + "939:\tlearn: 14799551.6406369\ttotal: 13.6s\tremaining: 871ms\n", + "940:\tlearn: 14795598.5613345\ttotal: 13.7s\tremaining: 856ms\n", + "941:\tlearn: 14794465.0278834\ttotal: 13.7s\tremaining: 842ms\n", + "942:\tlearn: 14794259.7563387\ttotal: 13.7s\tremaining: 827ms\n", + "943:\tlearn: 14794156.6214413\ttotal: 13.7s\tremaining: 813ms\n", + "944:\tlearn: 14792982.4344262\ttotal: 13.7s\tremaining: 798ms\n", + "945:\tlearn: 14792468.8012658\ttotal: 13.7s\tremaining: 783ms\n", + "946:\tlearn: 14792139.9923168\ttotal: 13.7s\tremaining: 769ms\n", + "947:\tlearn: 14786295.7938911\ttotal: 13.8s\tremaining: 754ms\n", + "948:\tlearn: 14783272.8762359\ttotal: 13.8s\tremaining: 740ms\n", + "949:\tlearn: 14782689.5513664\ttotal: 13.8s\tremaining: 725ms\n", + "950:\tlearn: 14782664.1266181\ttotal: 13.8s\tremaining: 710ms\n", + "951:\tlearn: 14774339.7873426\ttotal: 13.8s\tremaining: 696ms\n", + "952:\tlearn: 14769038.5403572\ttotal: 13.8s\tremaining: 681ms\n", + "953:\tlearn: 14761312.5028488\ttotal: 13.8s\tremaining: 667ms\n", + "954:\tlearn: 14760414.3496721\ttotal: 13.8s\tremaining: 652ms\n", + "955:\tlearn: 14752950.6860631\ttotal: 13.9s\tremaining: 638ms\n", + "956:\tlearn: 14752691.2767919\ttotal: 13.9s\tremaining: 623ms\n", + "957:\tlearn: 14752196.9268404\ttotal: 13.9s\tremaining: 608ms\n", + "958:\tlearn: 14750149.4071752\ttotal: 13.9s\tremaining: 594ms\n", + "959:\tlearn: 14749691.8632556\ttotal: 13.9s\tremaining: 579ms\n", + "960:\tlearn: 14749496.6164671\ttotal: 13.9s\tremaining: 565ms\n", + "961:\tlearn: 14749338.8909588\ttotal: 13.9s\tremaining: 550ms\n", + "962:\tlearn: 14749291.1143099\ttotal: 13.9s\tremaining: 536ms\n", + "963:\tlearn: 14739582.8672605\ttotal: 14s\tremaining: 521ms\n", + "964:\tlearn: 14739542.4442574\ttotal: 14s\tremaining: 506ms\n", + "965:\tlearn: 14739306.4391584\ttotal: 14s\tremaining: 492ms\n", + "966:\tlearn: 14708630.7344942\ttotal: 14s\tremaining: 477ms\n", + "967:\tlearn: 14680984.5320050\ttotal: 14s\tremaining: 463ms\n", + "968:\tlearn: 14677319.7390898\ttotal: 14s\tremaining: 448ms\n", + "969:\tlearn: 14677109.6650390\ttotal: 14s\tremaining: 434ms\n", + "970:\tlearn: 14676355.6242387\ttotal: 14s\tremaining: 419ms\n", + "971:\tlearn: 14667109.8761175\ttotal: 14.1s\tremaining: 405ms\n", + "972:\tlearn: 14643934.7657464\ttotal: 14.1s\tremaining: 390ms\n", + "973:\tlearn: 14642633.8347823\ttotal: 14.1s\tremaining: 376ms\n", + "974:\tlearn: 14642167.8071045\ttotal: 14.1s\tremaining: 362ms\n", + "975:\tlearn: 14641845.3365852\ttotal: 14.1s\tremaining: 347ms\n", + "976:\tlearn: 14640306.5144587\ttotal: 14.1s\tremaining: 333ms\n", + "977:\tlearn: 14640184.9147582\ttotal: 14.1s\tremaining: 318ms\n", + "978:\tlearn: 14640008.8483425\ttotal: 14.2s\tremaining: 304ms\n", + "979:\tlearn: 14636344.5519428\ttotal: 14.2s\tremaining: 289ms\n", + "980:\tlearn: 14633576.2663352\ttotal: 14.2s\tremaining: 275ms\n", + "981:\tlearn: 14624796.2134142\ttotal: 14.2s\tremaining: 260ms\n", + "982:\tlearn: 14624250.6774943\ttotal: 14.2s\tremaining: 246ms\n", + "983:\tlearn: 14615917.0650799\ttotal: 14.2s\tremaining: 231ms\n", + "984:\tlearn: 14615782.8388140\ttotal: 14.2s\tremaining: 217ms\n", + "985:\tlearn: 14612724.2754075\ttotal: 14.3s\tremaining: 203ms\n", + "986:\tlearn: 14609973.2837772\ttotal: 14.3s\tremaining: 188ms\n", + "987:\tlearn: 14605203.8050795\ttotal: 14.3s\tremaining: 174ms\n", + "988:\tlearn: 14605011.6874159\ttotal: 14.3s\tremaining: 159ms\n", + "989:\tlearn: 14594552.7887146\ttotal: 14.3s\tremaining: 145ms\n", + "990:\tlearn: 14591881.9316489\ttotal: 14.3s\tremaining: 130ms\n", + "991:\tlearn: 14581962.3358039\ttotal: 14.3s\tremaining: 116ms\n", + "992:\tlearn: 14581829.5331587\ttotal: 14.4s\tremaining: 101ms\n", + "993:\tlearn: 14581669.7347033\ttotal: 14.4s\tremaining: 86.8ms\n", + "994:\tlearn: 14577373.3119596\ttotal: 14.4s\tremaining: 72.3ms\n", + "995:\tlearn: 14577196.6327960\ttotal: 14.4s\tremaining: 57.8ms\n", + "996:\tlearn: 14577122.1536884\ttotal: 14.4s\tremaining: 43.4ms\n", + "997:\tlearn: 14574702.1336653\ttotal: 14.4s\tremaining: 28.9ms\n", + "998:\tlearn: 14574660.6060510\ttotal: 14.4s\tremaining: 14.4ms\n", + "999:\tlearn: 14574625.9856659\ttotal: 14.4s\tremaining: 0us\n" + ] + }, { "data": { "text/html": [ @@ -836,31 +1826,28 @@ " 'area', 'kitchen_area']),\n", " ('cat',\n", " OrdinalEncoder(handle_unknown='use_encoded_value',\n", - " unknown_value=999),\n", + " unknown_value=99999999),\n", " ['region', 'building_type',\n", - " 'object_type',\n", - " 'floor_level'])])),\n", + " 'object_type'])])),\n", " ('model',\n", - " RandomForestRegressor(max_depth=10, n_estimators=20))])In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org." ], "text/plain": [ "Pipeline(steps=[('preprocessor',\n", @@ -870,12 +1857,11 @@ " 'area', 'kitchen_area']),\n", " ('cat',\n", " OrdinalEncoder(handle_unknown='use_encoded_value',\n", - " unknown_value=999),\n", + " unknown_value=99999999),\n", " ['region', 'building_type',\n", - " 'object_type',\n", - " 'floor_level'])])),\n", + " 'object_type'])])),\n", " ('model',\n", - " RandomForestRegressor(max_depth=10, n_estimators=20))])" + " )])" ] }, "execution_count": 10, @@ -899,9 +1885,9 @@ { "data": { "text/plain": [ - "{'mae': np.float64(1276343.108894747),\n", - " 'mape': np.float64(0.35471390164231303),\n", - " 'mse': np.float64(174567675833231.12)}" + "{'mae': 1447931.3425270966,\n", + " 'mape': 1.6294525363466488e+18,\n", + " 'mse': 281898017343454.56}" ] }, "execution_count": 11, @@ -967,7 +1953,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "/home/andrey/work/institute/MLE/assets/mlflow/.venv_lec_mlflow/lib/python3.10/site-packages/mlflow/types/utils.py:407: UserWarning: Hint: Inferred schema contains integer column(s). Integer columns in Python cannot represent missing values. If your input data contains missing values at inference time, it will be encoded as floats and will cause a schema enforcement error. The best way to avoid this problem is to infer the model schema based on a realistic data sample (training dataset) that includes missing values. Alternatively, you can declare integer columns as doubles (float64) whenever these columns may have missing values. See `Handling Integers With Missing Values `_ for more details.\n", + "/home/andrey/work/institute/MLE/assets/mlflow/.venv_ml2/lib/python3.10/site-packages/mlflow/types/utils.py:407: UserWarning: Hint: Inferred schema contains integer column(s). Integer columns in Python cannot represent missing values. If your input data contains missing values at inference time, it will be encoded as floats and will cause a schema enforcement error. The best way to avoid this problem is to infer the model schema based on a realistic data sample (training dataset) that includes missing values. Alternatively, you can declare integer columns as doubles (float64) whenever these columns may have missing values. See `Handling Integers With Missing Values `_ for more details.\n", " warnings.warn(\n" ] } @@ -1011,8 +1997,8 @@ "name": "stderr", "output_type": "stream", "text": [ - "2024/10/03 18:59:13 INFO mlflow.tracking._tracking_service.client: 🏃 View run baseline model at: http://127.0.0.1:5000/#/experiments/1/runs/24e41bb582554f42953fe6dc2b6b190e.\n", - "2024/10/03 18:59:13 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" + "2024/10/10 13:34:49 INFO mlflow.tracking._tracking_service.client: 🏃 View run baseline model at: http://127.0.0.1:5000/#/experiments/1/runs/06fa7ec1f1b74aedb3509c88dc4ee1c0.\n", + "2024/10/10 13:34:49 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" ] } ], @@ -1091,85 +2077,85 @@ " artifact_uri\n", " start_time\n", " end_time\n", - " metrics.mae\n", " metrics.mape\n", + " metrics.mae\n", " metrics.mse\n", - " params.preprocessor__cat__handle_unknown\n", + " params.preprocessor__verbose_feature_names_out\n", " ...\n", - " params.model__max_samples\n", - " params.preprocessor__transformers\n", - " params.model__monotonic_cst\n", - " params.model__warm_start\n", - " params.preprocessor__remainder\n", - " tags.mlflow.user\n", + " params.preprocessor__num\n", + " params.preprocessor__force_int_remainder_cols\n", + " params.preprocessor__cat__unknown_value\n", + " params.memory\n", + " params.preprocessor__cat__categories\n", " tags.mlflow.source.type\n", - " tags.mlflow.runName\n", - " tags.mlflow.source.name\n", + " tags.mlflow.user\n", " tags.mlflow.log-model.history\n", + " tags.mlflow.source.name\n", + " tags.mlflow.runName\n", " \n", " \n", " \n", " \n", " 0\n", - " 24e41bb582554f42953fe6dc2b6b190e\n", + " 06fa7ec1f1b74aedb3509c88dc4ee1c0\n", " 1\n", " FINISHED\n", - " mlflow-artifacts:/1/24e41bb582554f42953fe6dc2b...\n", - " 2024-10-03 15:59:12.732000+00:00\n", - " 2024-10-03 15:59:13.921000+00:00\n", - " 1.276343e+06\n", - " 0.354714\n", - " 1.745677e+14\n", - " use_encoded_value\n", + " mlflow-artifacts:/1/06fa7ec1f1b74aedb3509c88dc...\n", + " 2024-10-10 10:34:49.202000+00:00\n", + " 2024-10-10 10:34:49.765000+00:00\n", + " 1.629453e+18\n", + " 1.447931e+06\n", + " 2.818980e+14\n", + " True\n", " ...\n", + " StandardScaler()\n", + " True\n", + " 99999999\n", " None\n", - " [('num', StandardScaler(), ['geo_lat', 'geo_lo...\n", - " None\n", - " False\n", - " drop\n", - " andrey\n", + " auto\n", " LOCAL\n", - " baseline model\n", + " andrey\n", + " [{\"run_id\": \"06fa7ec1f1b74aedb3509c88dc4ee1c0\"...\n", " /home/andrey/work/institute/MLE/assets/mlflow/...\n", - " [{\"run_id\": \"24e41bb582554f42953fe6dc2b6b190e\"...\n", + " baseline model\n", " \n", " \n", "\n", - "

1 rows × 57 columns

\n", + "

1 rows × 40 columns

\n", "" ], "text/plain": [ " run_id experiment_id status \\\n", - "0 24e41bb582554f42953fe6dc2b6b190e 1 FINISHED \n", + "0 06fa7ec1f1b74aedb3509c88dc4ee1c0 1 FINISHED \n", "\n", " artifact_uri \\\n", - "0 mlflow-artifacts:/1/24e41bb582554f42953fe6dc2b... \n", + "0 mlflow-artifacts:/1/06fa7ec1f1b74aedb3509c88dc... \n", "\n", " start_time end_time \\\n", - "0 2024-10-03 15:59:12.732000+00:00 2024-10-03 15:59:13.921000+00:00 \n", + "0 2024-10-10 10:34:49.202000+00:00 2024-10-10 10:34:49.765000+00:00 \n", "\n", - " metrics.mae metrics.mape metrics.mse \\\n", - "0 1.276343e+06 0.354714 1.745677e+14 \n", + " metrics.mape metrics.mae metrics.mse \\\n", + "0 1.629453e+18 1.447931e+06 2.818980e+14 \n", "\n", - " params.preprocessor__cat__handle_unknown ... params.model__max_samples \\\n", - "0 use_encoded_value ... None \n", + " params.preprocessor__verbose_feature_names_out ... \\\n", + "0 True ... \n", "\n", - " params.preprocessor__transformers \\\n", - "0 [('num', StandardScaler(), ['geo_lat', 'geo_lo... \n", + " params.preprocessor__num params.preprocessor__force_int_remainder_cols \\\n", + "0 StandardScaler() True \n", "\n", - " params.model__monotonic_cst params.model__warm_start \\\n", - "0 None False \n", + " params.preprocessor__cat__unknown_value params.memory \\\n", + "0 99999999 None \n", "\n", - " params.preprocessor__remainder tags.mlflow.user tags.mlflow.source.type \\\n", - "0 drop andrey LOCAL \n", + " params.preprocessor__cat__categories tags.mlflow.source.type \\\n", + "0 auto LOCAL \n", "\n", - " tags.mlflow.runName tags.mlflow.source.name \\\n", - "0 baseline model /home/andrey/work/institute/MLE/assets/mlflow/... \n", + " tags.mlflow.user tags.mlflow.log-model.history \\\n", + "0 andrey [{\"run_id\": \"06fa7ec1f1b74aedb3509c88dc4ee1c0\"... \n", "\n", - " tags.mlflow.log-model.history \n", - "0 [{\"run_id\": \"24e41bb582554f42953fe6dc2b6b190e\"... \n", + " tags.mlflow.source.name tags.mlflow.runName \n", + "0 /home/andrey/work/institute/MLE/assets/mlflow/... baseline model \n", "\n", - "[1 rows x 57 columns]" + "[1 rows x 40 columns]" ] }, "execution_count": 19, @@ -1223,11 +2209,1024 @@ "name": "stderr", "output_type": "stream", "text": [ - "2024/10/03 18:59:14 WARNING mlflow.utils.autologging_utils: MLflow sklearn autologging is known to be compatible with 0.24.1 <= scikit-learn <= 1.5.1, but the installed version is 1.5.2. If you encounter errors during autologging, try upgrading / downgrading scikit-learn to a compatible version, or try upgrading MLflow.\n", - "2024/10/03 19:02:16 WARNING mlflow.utils.autologging_utils: MLflow autologging encountered a warning: \"/home/andrey/work/institute/MLE/assets/mlflow/.venv_lec_mlflow/lib/python3.10/site-packages/mlflow/types/utils.py:407: UserWarning: Hint: Inferred schema contains integer column(s). Integer columns in Python cannot represent missing values. If your input data contains missing values at inference time, it will be encoded as floats and will cause a schema enforcement error. The best way to avoid this problem is to infer the model schema based on a realistic data sample (training dataset) that includes missing values. Alternatively, you can declare integer columns as doubles (float64) whenever these columns may have missing values. See `Handling Integers With Missing Values `_ for more details.\"\n", - "2024/10/03 19:02:40 WARNING mlflow.utils.autologging_utils: MLflow autologging encountered a warning: \"/home/andrey/work/institute/MLE/assets/mlflow/.venv_lec_mlflow/lib/python3.10/site-packages/mlflow/types/utils.py:407: UserWarning: Hint: Inferred schema contains integer column(s). Integer columns in Python cannot represent missing values. If your input data contains missing values at inference time, it will be encoded as floats and will cause a schema enforcement error. The best way to avoid this problem is to infer the model schema based on a realistic data sample (training dataset) that includes missing values. Alternatively, you can declare integer columns as doubles (float64) whenever these columns may have missing values. See `Handling Integers With Missing Values `_ for more details.\"\n", - "2024/10/03 19:02:42 INFO mlflow.tracking._tracking_service.client: 🏃 View run auto at: http://127.0.0.1:5000/#/experiments/1/runs/2ced09116c264623b89d8df7fe33cb10.\n", - "2024/10/03 19:02:42 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" + "2024/10/10 13:34:49 WARNING mlflow.utils.autologging_utils: MLflow sklearn autologging is known to be compatible with 0.24.1 <= scikit-learn <= 1.5.1, but the installed version is 1.5.2. If you encounter errors during autologging, try upgrading / downgrading scikit-learn to a compatible version, or try upgrading MLflow.\n", + "2024/10/10 13:36:26 WARNING mlflow.utils.autologging_utils: MLflow autologging encountered a warning: \"/home/andrey/work/institute/MLE/assets/mlflow/.venv_ml2/lib/python3.10/site-packages/mlflow/types/utils.py:407: UserWarning: Hint: Inferred schema contains integer column(s). Integer columns in Python cannot represent missing values. If your input data contains missing values at inference time, it will be encoded as floats and will cause a schema enforcement error. The best way to avoid this problem is to infer the model schema based on a realistic data sample (training dataset) that includes missing values. Alternatively, you can declare integer columns as doubles (float64) whenever these columns may have missing values. See `Handling Integers With Missing Values `_ for more details.\"\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Learning rate set to 0.105957\n", + "0:\tlearn: 22102085.4544239\ttotal: 12.9ms\tremaining: 12.9s\n", + "1:\tlearn: 21994630.3403412\ttotal: 24.9ms\tremaining: 12.4s\n", + "2:\tlearn: 21906687.8196027\ttotal: 36.3ms\tremaining: 12.1s\n", + "3:\tlearn: 21834890.5050552\ttotal: 47.9ms\tremaining: 11.9s\n", + "4:\tlearn: 21770820.6751194\ttotal: 59ms\tremaining: 11.7s\n", + "5:\tlearn: 21719543.9330108\ttotal: 70.8ms\tremaining: 11.7s\n", + "6:\tlearn: 21676510.1666598\ttotal: 83.7ms\tremaining: 11.9s\n", + "7:\tlearn: 21641355.8079016\ttotal: 95.1ms\tremaining: 11.8s\n", + "8:\tlearn: 21612289.0494648\ttotal: 107ms\tremaining: 11.8s\n", + "9:\tlearn: 21583808.7061085\ttotal: 119ms\tremaining: 11.7s\n", + "10:\tlearn: 21559288.9618040\ttotal: 129ms\tremaining: 11.6s\n", + "11:\tlearn: 21537048.9920531\ttotal: 141ms\tremaining: 11.6s\n", + "12:\tlearn: 21444526.1629239\ttotal: 153ms\tremaining: 11.6s\n", + "13:\tlearn: 21426349.3370315\ttotal: 165ms\tremaining: 11.6s\n", + "14:\tlearn: 21411901.2338278\ttotal: 178ms\tremaining: 11.7s\n", + "15:\tlearn: 21399279.8023459\ttotal: 190ms\tremaining: 11.7s\n", + "16:\tlearn: 21299421.1434822\ttotal: 203ms\tremaining: 11.8s\n", + "17:\tlearn: 21288560.2595435\ttotal: 216ms\tremaining: 11.8s\n", + "18:\tlearn: 21277368.8876877\ttotal: 228ms\tremaining: 11.8s\n", + "19:\tlearn: 21229205.2938305\ttotal: 241ms\tremaining: 11.8s\n", + "20:\tlearn: 21220238.4828158\ttotal: 253ms\tremaining: 11.8s\n", + "21:\tlearn: 21212849.7885410\ttotal: 265ms\tremaining: 11.8s\n", + "22:\tlearn: 21205304.4132821\ttotal: 278ms\tremaining: 11.8s\n", + "23:\tlearn: 21198813.8508479\ttotal: 290ms\tremaining: 11.8s\n", + "24:\tlearn: 21184627.2326983\ttotal: 300ms\tremaining: 11.7s\n", + "25:\tlearn: 21172748.3410688\ttotal: 314ms\tremaining: 11.7s\n", + "26:\tlearn: 21103305.4766520\ttotal: 327ms\tremaining: 11.8s\n", + "27:\tlearn: 21096636.4037750\ttotal: 338ms\tremaining: 11.7s\n", + "28:\tlearn: 21082202.2892557\ttotal: 351ms\tremaining: 11.8s\n", + "29:\tlearn: 21077185.5274954\ttotal: 363ms\tremaining: 11.7s\n", + "30:\tlearn: 21071613.1691098\ttotal: 375ms\tremaining: 11.7s\n", + "31:\tlearn: 21067654.8502386\ttotal: 386ms\tremaining: 11.7s\n", + "32:\tlearn: 21053425.8947843\ttotal: 399ms\tremaining: 11.7s\n", + "33:\tlearn: 21038024.0563140\ttotal: 410ms\tremaining: 11.7s\n", + "34:\tlearn: 20961357.9814339\ttotal: 422ms\tremaining: 11.6s\n", + "35:\tlearn: 20946027.4479676\ttotal: 437ms\tremaining: 11.7s\n", + "36:\tlearn: 20866676.4104322\ttotal: 450ms\tremaining: 11.7s\n", + "37:\tlearn: 20863078.3182449\ttotal: 461ms\tremaining: 11.7s\n", + "38:\tlearn: 20859910.3609500\ttotal: 474ms\tremaining: 11.7s\n", + "39:\tlearn: 20853462.2703730\ttotal: 484ms\tremaining: 11.6s\n", + "40:\tlearn: 20851610.3209036\ttotal: 496ms\tremaining: 11.6s\n", + "41:\tlearn: 20847674.0809285\ttotal: 508ms\tremaining: 11.6s\n", + "42:\tlearn: 20845384.9263391\ttotal: 521ms\tremaining: 11.6s\n", + "43:\tlearn: 20843256.7428906\ttotal: 535ms\tremaining: 11.6s\n", + "44:\tlearn: 20841580.8594834\ttotal: 548ms\tremaining: 11.6s\n", + "45:\tlearn: 20819301.2718345\ttotal: 564ms\tremaining: 11.7s\n", + "46:\tlearn: 20812094.5913582\ttotal: 580ms\tremaining: 11.8s\n", + "47:\tlearn: 20808932.0866915\ttotal: 594ms\tremaining: 11.8s\n", + "48:\tlearn: 20763172.9200413\ttotal: 612ms\tremaining: 11.9s\n", + "49:\tlearn: 20729084.6574594\ttotal: 626ms\tremaining: 11.9s\n", + "50:\tlearn: 20721820.5403996\ttotal: 640ms\tremaining: 11.9s\n", + "51:\tlearn: 20715664.3732084\ttotal: 653ms\tremaining: 11.9s\n", + "52:\tlearn: 20712658.7025295\ttotal: 663ms\tremaining: 11.9s\n", + "53:\tlearn: 20704254.1704930\ttotal: 675ms\tremaining: 11.8s\n", + "54:\tlearn: 20690967.9220470\ttotal: 685ms\tremaining: 11.8s\n", + "55:\tlearn: 20686546.8978473\ttotal: 696ms\tremaining: 11.7s\n", + "56:\tlearn: 20682362.4255777\ttotal: 708ms\tremaining: 11.7s\n", + "57:\tlearn: 20680744.8113421\ttotal: 719ms\tremaining: 11.7s\n", + "58:\tlearn: 20677926.0871267\ttotal: 730ms\tremaining: 11.6s\n", + "59:\tlearn: 20658478.3098789\ttotal: 743ms\tremaining: 11.6s\n", + "60:\tlearn: 20641964.4472246\ttotal: 756ms\tremaining: 11.6s\n", + "61:\tlearn: 20639551.4216654\ttotal: 767ms\tremaining: 11.6s\n", + "62:\tlearn: 20638344.8919341\ttotal: 778ms\tremaining: 11.6s\n", + "63:\tlearn: 20635991.3894815\ttotal: 790ms\tremaining: 11.6s\n", + "64:\tlearn: 20595846.8116432\ttotal: 802ms\tremaining: 11.5s\n", + "65:\tlearn: 20592198.9483046\ttotal: 813ms\tremaining: 11.5s\n", + "66:\tlearn: 20565316.0060422\ttotal: 827ms\tremaining: 11.5s\n", + "67:\tlearn: 20563073.6783517\ttotal: 838ms\tremaining: 11.5s\n", + "68:\tlearn: 20553650.4649650\ttotal: 851ms\tremaining: 11.5s\n", + "69:\tlearn: 20545510.8230653\ttotal: 862ms\tremaining: 11.4s\n", + "70:\tlearn: 20544114.9272186\ttotal: 872ms\tremaining: 11.4s\n", + "71:\tlearn: 20541689.8802451\ttotal: 884ms\tremaining: 11.4s\n", + "72:\tlearn: 20538792.7074671\ttotal: 896ms\tremaining: 11.4s\n", + "73:\tlearn: 20517134.0713648\ttotal: 909ms\tremaining: 11.4s\n", + "74:\tlearn: 20510477.9089445\ttotal: 922ms\tremaining: 11.4s\n", + "75:\tlearn: 20494649.9067257\ttotal: 934ms\tremaining: 11.4s\n", + "76:\tlearn: 20490851.9879851\ttotal: 946ms\tremaining: 11.3s\n", + "77:\tlearn: 20488939.9621874\ttotal: 958ms\tremaining: 11.3s\n", + "78:\tlearn: 20432532.8171644\ttotal: 970ms\tremaining: 11.3s\n", + "79:\tlearn: 20428397.7107150\ttotal: 982ms\tremaining: 11.3s\n", + "80:\tlearn: 20421638.7734419\ttotal: 995ms\tremaining: 11.3s\n", + "81:\tlearn: 20421021.7388457\ttotal: 1.01s\tremaining: 11.3s\n", + "82:\tlearn: 20406404.2376730\ttotal: 1.02s\tremaining: 11.3s\n", + "83:\tlearn: 20021682.5008511\ttotal: 1.03s\tremaining: 11.3s\n", + "84:\tlearn: 20018322.6048631\ttotal: 1.05s\tremaining: 11.3s\n", + "85:\tlearn: 20004841.3476490\ttotal: 1.06s\tremaining: 11.2s\n", + "86:\tlearn: 19985666.0092745\ttotal: 1.07s\tremaining: 11.2s\n", + "87:\tlearn: 19983778.1947243\ttotal: 1.08s\tremaining: 11.2s\n", + "88:\tlearn: 19982460.1107908\ttotal: 1.09s\tremaining: 11.2s\n", + "89:\tlearn: 19979128.5494690\ttotal: 1.11s\tremaining: 11.2s\n", + "90:\tlearn: 19974094.9707357\ttotal: 1.11s\tremaining: 11.1s\n", + "91:\tlearn: 19972006.9431031\ttotal: 1.13s\tremaining: 11.1s\n", + "92:\tlearn: 19970846.2845466\ttotal: 1.14s\tremaining: 11.1s\n", + "93:\tlearn: 19968858.0073042\ttotal: 1.15s\tremaining: 11.1s\n", + "94:\tlearn: 19921720.6252972\ttotal: 1.17s\tremaining: 11.1s\n", + "95:\tlearn: 19916568.5707839\ttotal: 1.18s\tremaining: 11.1s\n", + "96:\tlearn: 19913228.5247508\ttotal: 1.19s\tremaining: 11.1s\n", + "97:\tlearn: 19901982.4625895\ttotal: 1.2s\tremaining: 11s\n", + "98:\tlearn: 19836107.7247888\ttotal: 1.21s\tremaining: 11s\n", + "99:\tlearn: 19834724.7455166\ttotal: 1.23s\tremaining: 11s\n", + "100:\tlearn: 19832811.9745741\ttotal: 1.24s\tremaining: 11s\n", + "101:\tlearn: 19818491.2851567\ttotal: 1.25s\tremaining: 11s\n", + "102:\tlearn: 19815779.3719026\ttotal: 1.26s\tremaining: 11s\n", + "103:\tlearn: 19814215.0962787\ttotal: 1.27s\tremaining: 11s\n", + "104:\tlearn: 19782274.6892663\ttotal: 1.29s\tremaining: 11s\n", + "105:\tlearn: 19777945.6507456\ttotal: 1.3s\tremaining: 11s\n", + "106:\tlearn: 19770488.9772154\ttotal: 1.31s\tremaining: 11s\n", + "107:\tlearn: 19769758.0023174\ttotal: 1.32s\tremaining: 10.9s\n", + "108:\tlearn: 19767541.9303017\ttotal: 1.34s\tremaining: 10.9s\n", + "109:\tlearn: 19766992.0126300\ttotal: 1.35s\tremaining: 10.9s\n", + "110:\tlearn: 19765032.8837298\ttotal: 1.36s\tremaining: 10.9s\n", + "111:\tlearn: 19705204.6771073\ttotal: 1.38s\tremaining: 10.9s\n", + "112:\tlearn: 19703649.0394020\ttotal: 1.39s\tremaining: 10.9s\n", + "113:\tlearn: 19693038.0415419\ttotal: 1.4s\tremaining: 10.9s\n", + "114:\tlearn: 19690294.4304072\ttotal: 1.42s\tremaining: 10.9s\n", + "115:\tlearn: 19686529.4709294\ttotal: 1.43s\tremaining: 10.9s\n", + "116:\tlearn: 19684887.8267152\ttotal: 1.44s\tremaining: 10.9s\n", + "117:\tlearn: 19369465.6970761\ttotal: 1.45s\tremaining: 10.9s\n", + "118:\tlearn: 19368868.0416380\ttotal: 1.47s\tremaining: 10.9s\n", + "119:\tlearn: 19334590.5868513\ttotal: 1.48s\tremaining: 10.9s\n", + "120:\tlearn: 19332200.0832597\ttotal: 1.49s\tremaining: 10.8s\n", + "121:\tlearn: 19320130.9244745\ttotal: 1.5s\tremaining: 10.8s\n", + "122:\tlearn: 19318220.9448337\ttotal: 1.51s\tremaining: 10.8s\n", + "123:\tlearn: 18941546.2095714\ttotal: 1.53s\tremaining: 10.8s\n", + "124:\tlearn: 18941056.2836883\ttotal: 1.54s\tremaining: 10.8s\n", + "125:\tlearn: 18939637.9662976\ttotal: 1.55s\tremaining: 10.7s\n", + "126:\tlearn: 18938172.4621610\ttotal: 1.56s\tremaining: 10.8s\n", + "127:\tlearn: 18935889.3619752\ttotal: 1.58s\tremaining: 10.8s\n", + "128:\tlearn: 18928784.7025346\ttotal: 1.59s\tremaining: 10.8s\n", + "129:\tlearn: 18926981.6933453\ttotal: 1.6s\tremaining: 10.7s\n", + "130:\tlearn: 18830178.3173696\ttotal: 1.61s\tremaining: 10.7s\n", + "131:\tlearn: 18828102.3918672\ttotal: 1.63s\tremaining: 10.7s\n", + "132:\tlearn: 18825755.9987015\ttotal: 1.64s\tremaining: 10.7s\n", + "133:\tlearn: 18793049.5462155\ttotal: 1.65s\tremaining: 10.7s\n", + "134:\tlearn: 18791452.8400128\ttotal: 1.66s\tremaining: 10.7s\n", + "135:\tlearn: 18484591.4924421\ttotal: 1.68s\tremaining: 10.6s\n", + "136:\tlearn: 18482373.1605741\ttotal: 1.69s\tremaining: 10.6s\n", + "137:\tlearn: 18414571.2543321\ttotal: 1.7s\tremaining: 10.6s\n", + "138:\tlearn: 18412913.4160574\ttotal: 1.71s\tremaining: 10.6s\n", + "139:\tlearn: 18409214.1141794\ttotal: 1.72s\tremaining: 10.6s\n", + "140:\tlearn: 18395140.1008086\ttotal: 1.74s\tremaining: 10.6s\n", + "141:\tlearn: 18390939.2248151\ttotal: 1.75s\tremaining: 10.6s\n", + "142:\tlearn: 18377925.8298573\ttotal: 1.76s\tremaining: 10.6s\n", + "143:\tlearn: 18371775.1291009\ttotal: 1.77s\tremaining: 10.5s\n", + "144:\tlearn: 18370251.1042623\ttotal: 1.78s\tremaining: 10.5s\n", + "145:\tlearn: 18332707.1499911\ttotal: 1.8s\tremaining: 10.5s\n", + "146:\tlearn: 18330693.2665230\ttotal: 1.81s\tremaining: 10.5s\n", + "147:\tlearn: 18329408.2952767\ttotal: 1.82s\tremaining: 10.5s\n", + "148:\tlearn: 18321783.9892793\ttotal: 1.83s\tremaining: 10.5s\n", + "149:\tlearn: 18321270.4958267\ttotal: 1.85s\tremaining: 10.5s\n", + "150:\tlearn: 18310325.1681801\ttotal: 1.86s\tremaining: 10.5s\n", + "151:\tlearn: 18299986.9413893\ttotal: 1.87s\tremaining: 10.4s\n", + "152:\tlearn: 18290217.7479708\ttotal: 1.89s\tremaining: 10.4s\n", + "153:\tlearn: 18280975.8537910\ttotal: 1.9s\tremaining: 10.4s\n", + "154:\tlearn: 18272215.6509019\ttotal: 1.91s\tremaining: 10.4s\n", + "155:\tlearn: 18263878.2178516\ttotal: 1.92s\tremaining: 10.4s\n", + "156:\tlearn: 18256009.4859248\ttotal: 1.94s\tremaining: 10.4s\n", + "157:\tlearn: 18248529.7799856\ttotal: 1.95s\tremaining: 10.4s\n", + "158:\tlearn: 18241388.0845094\ttotal: 1.96s\tremaining: 10.4s\n", + "159:\tlearn: 18234700.5127085\ttotal: 1.97s\tremaining: 10.3s\n", + "160:\tlearn: 18228095.5839778\ttotal: 1.98s\tremaining: 10.3s\n", + "161:\tlearn: 18222087.5153066\ttotal: 2s\tremaining: 10.3s\n", + "162:\tlearn: 18215963.2971261\ttotal: 2.01s\tremaining: 10.3s\n", + "163:\tlearn: 18210272.5545163\ttotal: 2.02s\tremaining: 10.3s\n", + "164:\tlearn: 18208920.7703569\ttotal: 2.03s\tremaining: 10.3s\n", + "165:\tlearn: 18204704.7145239\ttotal: 2.04s\tremaining: 10.3s\n", + "166:\tlearn: 18187135.8260335\ttotal: 2.05s\tremaining: 10.2s\n", + "167:\tlearn: 18183064.7135734\ttotal: 2.06s\tremaining: 10.2s\n", + "168:\tlearn: 18177887.1670860\ttotal: 2.08s\tremaining: 10.2s\n", + "169:\tlearn: 18173022.2110313\ttotal: 2.09s\tremaining: 10.2s\n", + "170:\tlearn: 18168573.4167384\ttotal: 2.1s\tremaining: 10.2s\n", + "171:\tlearn: 18165036.1971623\ttotal: 2.11s\tremaining: 10.2s\n", + "172:\tlearn: 18161841.9822954\ttotal: 2.13s\tremaining: 10.2s\n", + "173:\tlearn: 18129860.2061383\ttotal: 2.14s\tremaining: 10.1s\n", + "174:\tlearn: 18127931.5161091\ttotal: 2.15s\tremaining: 10.1s\n", + "175:\tlearn: 18124997.7778403\ttotal: 2.16s\tremaining: 10.1s\n", + "176:\tlearn: 18122975.2084322\ttotal: 2.17s\tremaining: 10.1s\n", + "177:\tlearn: 18120855.5325733\ttotal: 2.18s\tremaining: 10.1s\n", + "178:\tlearn: 18117907.6019994\ttotal: 2.2s\tremaining: 10.1s\n", + "179:\tlearn: 18116674.0864027\ttotal: 2.21s\tremaining: 10.1s\n", + "180:\tlearn: 18114086.9287957\ttotal: 2.22s\tremaining: 10s\n", + "181:\tlearn: 18087100.0827926\ttotal: 2.23s\tremaining: 10s\n", + "182:\tlearn: 18071944.2213105\ttotal: 2.24s\tremaining: 10s\n", + "183:\tlearn: 17952691.4261792\ttotal: 2.26s\tremaining: 10s\n", + "184:\tlearn: 17950298.6715866\ttotal: 2.27s\tremaining: 9.99s\n", + "185:\tlearn: 17949031.8169417\ttotal: 2.28s\tremaining: 9.98s\n", + "186:\tlearn: 17937943.5186847\ttotal: 2.29s\tremaining: 9.97s\n", + "187:\tlearn: 17937014.8027177\ttotal: 2.3s\tremaining: 9.96s\n", + "188:\tlearn: 17936493.5945773\ttotal: 2.31s\tremaining: 9.94s\n", + "189:\tlearn: 17935386.0093649\ttotal: 2.33s\tremaining: 9.92s\n", + "190:\tlearn: 17934203.8644718\ttotal: 2.34s\tremaining: 9.91s\n", + "191:\tlearn: 17928336.5184065\ttotal: 2.35s\tremaining: 9.89s\n", + "192:\tlearn: 17925443.1940046\ttotal: 2.36s\tremaining: 9.88s\n", + "193:\tlearn: 17924535.5533845\ttotal: 2.37s\tremaining: 9.86s\n", + "194:\tlearn: 17917225.8802206\ttotal: 2.39s\tremaining: 9.85s\n", + "195:\tlearn: 17904437.4148190\ttotal: 2.4s\tremaining: 9.84s\n", + "196:\tlearn: 17902915.3467923\ttotal: 2.41s\tremaining: 9.83s\n", + "197:\tlearn: 17900924.7512305\ttotal: 2.42s\tremaining: 9.81s\n", + "198:\tlearn: 17899976.2262471\ttotal: 2.43s\tremaining: 9.79s\n", + "199:\tlearn: 17896573.5977064\ttotal: 2.45s\tremaining: 9.79s\n", + "200:\tlearn: 17894480.1301072\ttotal: 2.46s\tremaining: 9.77s\n", + "201:\tlearn: 17891369.5414483\ttotal: 2.47s\tremaining: 9.76s\n", + "202:\tlearn: 17853776.3679239\ttotal: 2.48s\tremaining: 9.75s\n", + "203:\tlearn: 17851457.0828592\ttotal: 2.49s\tremaining: 9.73s\n", + "204:\tlearn: 17849621.6767992\ttotal: 2.5s\tremaining: 9.72s\n", + "205:\tlearn: 17848392.5509482\ttotal: 2.52s\tremaining: 9.7s\n", + "206:\tlearn: 17845597.2428619\ttotal: 2.53s\tremaining: 9.69s\n", + "207:\tlearn: 17841951.2763157\ttotal: 2.54s\tremaining: 9.68s\n", + "208:\tlearn: 17829332.8912371\ttotal: 2.55s\tremaining: 9.66s\n", + "209:\tlearn: 17825984.1152963\ttotal: 2.56s\tremaining: 9.65s\n", + "210:\tlearn: 17821360.2498463\ttotal: 2.58s\tremaining: 9.64s\n", + "211:\tlearn: 17816041.9633158\ttotal: 2.59s\tremaining: 9.63s\n", + "212:\tlearn: 17815089.0154101\ttotal: 2.6s\tremaining: 9.62s\n", + "213:\tlearn: 17812260.4222221\ttotal: 2.62s\tremaining: 9.61s\n", + "214:\tlearn: 17811642.1796060\ttotal: 2.63s\tremaining: 9.6s\n", + "215:\tlearn: 17811104.8656724\ttotal: 2.64s\tremaining: 9.58s\n", + "216:\tlearn: 17810456.2984828\ttotal: 2.65s\tremaining: 9.57s\n", + "217:\tlearn: 17809982.4909707\ttotal: 2.66s\tremaining: 9.55s\n", + "218:\tlearn: 17809543.7803178\ttotal: 2.67s\tremaining: 9.54s\n", + "219:\tlearn: 17809136.8325569\ttotal: 2.68s\tremaining: 9.52s\n", + "220:\tlearn: 17808758.7315278\ttotal: 2.7s\tremaining: 9.5s\n", + "221:\tlearn: 17808406.9145618\ttotal: 2.71s\tremaining: 9.49s\n", + "222:\tlearn: 17806754.0179687\ttotal: 2.72s\tremaining: 9.49s\n", + "223:\tlearn: 17806262.4885592\ttotal: 2.73s\tremaining: 9.47s\n", + "224:\tlearn: 17805319.3776209\ttotal: 2.75s\tremaining: 9.46s\n", + "225:\tlearn: 17805011.6013482\ttotal: 2.76s\tremaining: 9.44s\n", + "226:\tlearn: 17804724.0362310\ttotal: 2.77s\tremaining: 9.42s\n", + "227:\tlearn: 17793961.7547867\ttotal: 2.78s\tremaining: 9.41s\n", + "228:\tlearn: 17793044.3976904\ttotal: 2.79s\tremaining: 9.4s\n", + "229:\tlearn: 17791876.3449986\ttotal: 2.8s\tremaining: 9.39s\n", + "230:\tlearn: 17770039.2877531\ttotal: 2.82s\tremaining: 9.38s\n", + "231:\tlearn: 17769759.3423197\ttotal: 2.83s\tremaining: 9.36s\n", + "232:\tlearn: 17769498.1846872\ttotal: 2.84s\tremaining: 9.35s\n", + "233:\tlearn: 17769106.6516586\ttotal: 2.85s\tremaining: 9.33s\n", + "234:\tlearn: 17765866.7512613\ttotal: 2.86s\tremaining: 9.32s\n", + "235:\tlearn: 17763818.0836765\ttotal: 2.87s\tremaining: 9.3s\n", + "236:\tlearn: 17761637.5687877\ttotal: 2.89s\tremaining: 9.29s\n", + "237:\tlearn: 17755293.6166299\ttotal: 2.9s\tremaining: 9.28s\n", + "238:\tlearn: 17749597.6285121\ttotal: 2.91s\tremaining: 9.27s\n", + "239:\tlearn: 17731193.4780969\ttotal: 2.92s\tremaining: 9.26s\n", + "240:\tlearn: 17730941.1840209\ttotal: 2.94s\tremaining: 9.24s\n", + "241:\tlearn: 17730651.4109866\ttotal: 2.94s\tremaining: 9.23s\n", + "242:\tlearn: 17729951.1772204\ttotal: 2.96s\tremaining: 9.21s\n", + "243:\tlearn: 17725674.6169533\ttotal: 2.97s\tremaining: 9.2s\n", + "244:\tlearn: 17724397.3837970\ttotal: 2.98s\tremaining: 9.19s\n", + "245:\tlearn: 17723085.9667878\ttotal: 2.99s\tremaining: 9.18s\n", + "246:\tlearn: 17716068.0643361\ttotal: 3.01s\tremaining: 9.16s\n", + "247:\tlearn: 17685621.7941613\ttotal: 3.02s\tremaining: 9.15s\n", + "248:\tlearn: 17684272.6716694\ttotal: 3.03s\tremaining: 9.14s\n", + "249:\tlearn: 17683390.0888279\ttotal: 3.04s\tremaining: 9.13s\n", + "250:\tlearn: 17683052.4845925\ttotal: 3.05s\tremaining: 9.11s\n", + "251:\tlearn: 17678624.0868252\ttotal: 3.06s\tremaining: 9.1s\n", + "252:\tlearn: 17665657.9640584\ttotal: 3.08s\tremaining: 9.09s\n", + "253:\tlearn: 17664624.5487132\ttotal: 3.09s\tremaining: 9.08s\n", + "254:\tlearn: 17663925.0646167\ttotal: 3.1s\tremaining: 9.07s\n", + "255:\tlearn: 17653813.6196925\ttotal: 3.12s\tremaining: 9.06s\n", + "256:\tlearn: 17636698.5157040\ttotal: 3.13s\tremaining: 9.05s\n", + "257:\tlearn: 17634671.9750893\ttotal: 3.14s\tremaining: 9.04s\n", + "258:\tlearn: 17633930.6422340\ttotal: 3.15s\tremaining: 9.02s\n", + "259:\tlearn: 17633026.0861171\ttotal: 3.17s\tremaining: 9.01s\n", + "260:\tlearn: 17632489.1254856\ttotal: 3.18s\tremaining: 8.99s\n", + "261:\tlearn: 17628474.9187765\ttotal: 3.19s\tremaining: 8.98s\n", + "262:\tlearn: 17627320.9817928\ttotal: 3.2s\tremaining: 8.97s\n", + "263:\tlearn: 17626116.4772868\ttotal: 3.21s\tremaining: 8.95s\n", + "264:\tlearn: 17623329.0754817\ttotal: 3.22s\tremaining: 8.94s\n", + "265:\tlearn: 17622243.1901613\ttotal: 3.24s\tremaining: 8.93s\n", + "266:\tlearn: 17550321.8250878\ttotal: 3.25s\tremaining: 8.92s\n", + "267:\tlearn: 17549755.3651767\ttotal: 3.26s\tremaining: 8.91s\n", + "268:\tlearn: 17545607.1212430\ttotal: 3.28s\tremaining: 8.9s\n", + "269:\tlearn: 17541242.2629221\ttotal: 3.29s\tremaining: 8.89s\n", + "270:\tlearn: 17499407.7313592\ttotal: 3.3s\tremaining: 8.88s\n", + "271:\tlearn: 17499145.8282321\ttotal: 3.31s\tremaining: 8.86s\n", + "272:\tlearn: 17498934.5535116\ttotal: 3.32s\tremaining: 8.85s\n", + "273:\tlearn: 17498347.2546318\ttotal: 3.33s\tremaining: 8.83s\n", + "274:\tlearn: 17498149.7061684\ttotal: 3.34s\tremaining: 8.81s\n", + "275:\tlearn: 17497860.3337909\ttotal: 3.35s\tremaining: 8.8s\n", + "276:\tlearn: 17497134.2565818\ttotal: 3.37s\tremaining: 8.78s\n", + "277:\tlearn: 17496943.1446578\ttotal: 3.38s\tremaining: 8.77s\n", + "278:\tlearn: 17495461.7397646\ttotal: 3.39s\tremaining: 8.76s\n", + "279:\tlearn: 17492860.8467310\ttotal: 3.4s\tremaining: 8.75s\n", + "280:\tlearn: 17492256.7750564\ttotal: 3.41s\tremaining: 8.74s\n", + "281:\tlearn: 17491315.8920024\ttotal: 3.42s\tremaining: 8.72s\n", + "282:\tlearn: 17488802.8492737\ttotal: 3.44s\tremaining: 8.71s\n", + "283:\tlearn: 17479802.6541152\ttotal: 3.45s\tremaining: 8.7s\n", + "284:\tlearn: 17477169.5331720\ttotal: 3.46s\tremaining: 8.69s\n", + "285:\tlearn: 17474743.6190942\ttotal: 3.47s\tremaining: 8.67s\n", + "286:\tlearn: 17468342.7955232\ttotal: 3.49s\tremaining: 8.67s\n", + "287:\tlearn: 17467579.9985437\ttotal: 3.5s\tremaining: 8.66s\n", + "288:\tlearn: 17467009.9684055\ttotal: 3.51s\tremaining: 8.64s\n", + "289:\tlearn: 17464125.0260113\ttotal: 3.52s\tremaining: 8.63s\n", + "290:\tlearn: 17463508.0564477\ttotal: 3.54s\tremaining: 8.62s\n", + "291:\tlearn: 17453183.2620432\ttotal: 3.55s\tremaining: 8.61s\n", + "292:\tlearn: 17452971.0671546\ttotal: 3.56s\tremaining: 8.6s\n", + "293:\tlearn: 17452198.5884342\ttotal: 3.58s\tremaining: 8.6s\n", + "294:\tlearn: 17450925.6159031\ttotal: 3.59s\tremaining: 8.59s\n", + "295:\tlearn: 17450685.1155343\ttotal: 3.6s\tremaining: 8.57s\n", + "296:\tlearn: 17447975.7379237\ttotal: 3.62s\tremaining: 8.56s\n", + "297:\tlearn: 17446417.7251561\ttotal: 3.63s\tremaining: 8.55s\n", + "298:\tlearn: 17446166.7629704\ttotal: 3.64s\tremaining: 8.54s\n", + "299:\tlearn: 17445963.1442260\ttotal: 3.65s\tremaining: 8.52s\n", + "300:\tlearn: 17445745.7958927\ttotal: 3.66s\tremaining: 8.5s\n", + "301:\tlearn: 17444963.9290154\ttotal: 3.67s\tremaining: 8.49s\n", + "302:\tlearn: 17432650.1591210\ttotal: 3.69s\tremaining: 8.48s\n", + "303:\tlearn: 17430525.1210288\ttotal: 3.7s\tremaining: 8.47s\n", + "304:\tlearn: 17418414.4601453\ttotal: 3.71s\tremaining: 8.46s\n", + "305:\tlearn: 17417977.4735651\ttotal: 3.72s\tremaining: 8.44s\n", + "306:\tlearn: 17335624.2943914\ttotal: 3.73s\tremaining: 8.43s\n", + "307:\tlearn: 17323558.9233681\ttotal: 3.75s\tremaining: 8.42s\n", + "308:\tlearn: 17323047.3527617\ttotal: 3.76s\tremaining: 8.41s\n", + "309:\tlearn: 17322403.3488620\ttotal: 3.77s\tremaining: 8.39s\n", + "310:\tlearn: 17322187.6973801\ttotal: 3.78s\tremaining: 8.38s\n", + "311:\tlearn: 17320898.8497406\ttotal: 3.8s\tremaining: 8.37s\n", + "312:\tlearn: 17312668.7000429\ttotal: 3.81s\tremaining: 8.36s\n", + "313:\tlearn: 17299277.5985403\ttotal: 3.82s\tremaining: 8.35s\n", + "314:\tlearn: 17298175.9786240\ttotal: 3.83s\tremaining: 8.34s\n", + "315:\tlearn: 17296005.0430765\ttotal: 3.85s\tremaining: 8.33s\n", + "316:\tlearn: 17295834.3986842\ttotal: 3.86s\tremaining: 8.32s\n", + "317:\tlearn: 17295646.8271436\ttotal: 3.87s\tremaining: 8.3s\n", + "318:\tlearn: 17295412.2240763\ttotal: 3.88s\tremaining: 8.29s\n", + "319:\tlearn: 17295269.3891063\ttotal: 3.89s\tremaining: 8.27s\n", + "320:\tlearn: 17294720.1427139\ttotal: 3.9s\tremaining: 8.26s\n", + "321:\tlearn: 17280405.8179874\ttotal: 3.92s\tremaining: 8.25s\n", + "322:\tlearn: 17279788.6705542\ttotal: 3.93s\tremaining: 8.23s\n", + "323:\tlearn: 17259578.2219214\ttotal: 3.94s\tremaining: 8.22s\n", + "324:\tlearn: 17258995.8851109\ttotal: 3.95s\tremaining: 8.2s\n", + "325:\tlearn: 17256802.0040208\ttotal: 3.96s\tremaining: 8.19s\n", + "326:\tlearn: 17245667.9352932\ttotal: 3.97s\tremaining: 8.18s\n", + "327:\tlearn: 17245157.2383849\ttotal: 3.99s\tremaining: 8.17s\n", + "328:\tlearn: 17244420.0505767\ttotal: 4s\tremaining: 8.16s\n", + "329:\tlearn: 17240620.9311856\ttotal: 4.01s\tremaining: 8.14s\n", + "330:\tlearn: 17240126.6382259\ttotal: 4.02s\tremaining: 8.13s\n", + "331:\tlearn: 17239554.3263042\ttotal: 4.04s\tremaining: 8.12s\n", + "332:\tlearn: 17239249.4122676\ttotal: 4.05s\tremaining: 8.11s\n", + "333:\tlearn: 17237315.5959603\ttotal: 4.06s\tremaining: 8.1s\n", + "334:\tlearn: 17237170.4183008\ttotal: 4.07s\tremaining: 8.09s\n", + "335:\tlearn: 17235498.1709182\ttotal: 4.08s\tremaining: 8.07s\n", + "336:\tlearn: 17154286.9322136\ttotal: 4.1s\tremaining: 8.06s\n", + "337:\tlearn: 17152860.5403583\ttotal: 4.11s\tremaining: 8.05s\n", + "338:\tlearn: 17139897.5803445\ttotal: 4.12s\tremaining: 8.04s\n", + "339:\tlearn: 17139685.6194353\ttotal: 4.13s\tremaining: 8.03s\n", + "340:\tlearn: 17129406.8909698\ttotal: 4.15s\tremaining: 8.01s\n", + "341:\tlearn: 17126386.5318429\ttotal: 4.16s\tremaining: 8.01s\n", + "342:\tlearn: 17125338.5826429\ttotal: 4.17s\tremaining: 8s\n", + "343:\tlearn: 17124937.1764028\ttotal: 4.19s\tremaining: 7.99s\n", + "344:\tlearn: 17124773.5128614\ttotal: 4.2s\tremaining: 7.97s\n", + "345:\tlearn: 17123822.0085471\ttotal: 4.21s\tremaining: 7.97s\n", + "346:\tlearn: 17122604.8415169\ttotal: 4.23s\tremaining: 7.95s\n", + "347:\tlearn: 17121767.5370013\ttotal: 4.24s\tremaining: 7.94s\n", + "348:\tlearn: 17109471.1428348\ttotal: 4.25s\tremaining: 7.93s\n", + "349:\tlearn: 17092688.7777393\ttotal: 4.27s\tremaining: 7.92s\n", + "350:\tlearn: 17081854.5539987\ttotal: 4.28s\tremaining: 7.92s\n", + "351:\tlearn: 17081117.2220910\ttotal: 4.29s\tremaining: 7.91s\n", + "352:\tlearn: 17079431.1991192\ttotal: 4.31s\tremaining: 7.9s\n", + "353:\tlearn: 17065749.4676464\ttotal: 4.32s\tremaining: 7.89s\n", + "354:\tlearn: 17050839.2238400\ttotal: 4.34s\tremaining: 7.88s\n", + "355:\tlearn: 17050106.8831270\ttotal: 4.35s\tremaining: 7.87s\n", + "356:\tlearn: 17046033.2332065\ttotal: 4.37s\tremaining: 7.86s\n", + "357:\tlearn: 17043704.2415802\ttotal: 4.38s\tremaining: 7.85s\n", + "358:\tlearn: 17034226.2631681\ttotal: 4.39s\tremaining: 7.84s\n", + "359:\tlearn: 17019515.6806659\ttotal: 4.41s\tremaining: 7.83s\n", + "360:\tlearn: 17018472.9763746\ttotal: 4.42s\tremaining: 7.82s\n", + "361:\tlearn: 17017909.7121151\ttotal: 4.43s\tremaining: 7.81s\n", + "362:\tlearn: 17017463.3942640\ttotal: 4.44s\tremaining: 7.8s\n", + "363:\tlearn: 17016467.4317116\ttotal: 4.46s\tremaining: 7.79s\n", + "364:\tlearn: 17016320.3746025\ttotal: 4.47s\tremaining: 7.77s\n", + "365:\tlearn: 17014043.0108512\ttotal: 4.48s\tremaining: 7.76s\n", + "366:\tlearn: 17013536.3710672\ttotal: 4.5s\tremaining: 7.75s\n", + "367:\tlearn: 17011993.2014165\ttotal: 4.51s\tremaining: 7.75s\n", + "368:\tlearn: 17011849.5641841\ttotal: 4.52s\tremaining: 7.73s\n", + "369:\tlearn: 17011403.7126883\ttotal: 4.53s\tremaining: 7.72s\n", + "370:\tlearn: 17009763.5741945\ttotal: 4.55s\tremaining: 7.71s\n", + "371:\tlearn: 17009382.7519630\ttotal: 4.56s\tremaining: 7.7s\n", + "372:\tlearn: 17008464.7915054\ttotal: 4.57s\tremaining: 7.68s\n", + "373:\tlearn: 17008143.8161261\ttotal: 4.58s\tremaining: 7.67s\n", + "374:\tlearn: 16996814.2215431\ttotal: 4.6s\tremaining: 7.66s\n", + "375:\tlearn: 16996377.3351825\ttotal: 4.61s\tremaining: 7.65s\n", + "376:\tlearn: 16996037.5806770\ttotal: 4.62s\tremaining: 7.64s\n", + "377:\tlearn: 16991953.6478199\ttotal: 4.63s\tremaining: 7.63s\n", + "378:\tlearn: 16961328.6727692\ttotal: 4.65s\tremaining: 7.62s\n", + "379:\tlearn: 16957664.4831621\ttotal: 4.66s\tremaining: 7.61s\n", + "380:\tlearn: 16956856.4526881\ttotal: 4.67s\tremaining: 7.6s\n", + "381:\tlearn: 16947754.5891887\ttotal: 4.69s\tremaining: 7.59s\n", + "382:\tlearn: 16937471.3061729\ttotal: 4.7s\tremaining: 7.58s\n", + "383:\tlearn: 16910717.2697228\ttotal: 4.72s\tremaining: 7.57s\n", + "384:\tlearn: 16883021.8749316\ttotal: 4.73s\tremaining: 7.56s\n", + "385:\tlearn: 16874077.6620256\ttotal: 4.75s\tremaining: 7.55s\n", + "386:\tlearn: 16859663.0508862\ttotal: 4.76s\tremaining: 7.54s\n", + "387:\tlearn: 16843794.6984628\ttotal: 4.78s\tremaining: 7.54s\n", + "388:\tlearn: 16843670.2191430\ttotal: 4.79s\tremaining: 7.53s\n", + "389:\tlearn: 16833049.2556840\ttotal: 4.8s\tremaining: 7.52s\n", + "390:\tlearn: 16821522.4443567\ttotal: 4.82s\tremaining: 7.51s\n", + "391:\tlearn: 16818181.1766856\ttotal: 4.84s\tremaining: 7.5s\n", + "392:\tlearn: 16817749.5049150\ttotal: 4.85s\tremaining: 7.49s\n", + "393:\tlearn: 16817402.3614282\ttotal: 4.87s\tremaining: 7.49s\n", + "394:\tlearn: 16815679.7151727\ttotal: 4.88s\tremaining: 7.48s\n", + "395:\tlearn: 16810641.8717564\ttotal: 4.9s\tremaining: 7.47s\n", + "396:\tlearn: 16810291.1871768\ttotal: 4.91s\tremaining: 7.45s\n", + "397:\tlearn: 16808056.2422004\ttotal: 4.92s\tremaining: 7.44s\n", + "398:\tlearn: 16807804.2454334\ttotal: 4.93s\tremaining: 7.43s\n", + "399:\tlearn: 16799998.1957230\ttotal: 4.95s\tremaining: 7.42s\n", + "400:\tlearn: 16799220.2656080\ttotal: 4.96s\tremaining: 7.41s\n", + "401:\tlearn: 16798913.0252067\ttotal: 4.97s\tremaining: 7.4s\n", + "402:\tlearn: 16798319.2545577\ttotal: 4.99s\tremaining: 7.39s\n", + "403:\tlearn: 16796848.5752647\ttotal: 5s\tremaining: 7.38s\n", + "404:\tlearn: 16757656.8985529\ttotal: 5.02s\tremaining: 7.37s\n", + "405:\tlearn: 16745513.4381725\ttotal: 5.03s\tremaining: 7.36s\n", + "406:\tlearn: 16735416.8114581\ttotal: 5.04s\tremaining: 7.35s\n", + "407:\tlearn: 16734295.1424370\ttotal: 5.06s\tremaining: 7.34s\n", + "408:\tlearn: 16733140.3781664\ttotal: 5.07s\tremaining: 7.33s\n", + "409:\tlearn: 16723800.8980695\ttotal: 5.09s\tremaining: 7.33s\n", + "410:\tlearn: 16721200.9625357\ttotal: 5.11s\tremaining: 7.32s\n", + "411:\tlearn: 16720027.8472987\ttotal: 5.12s\tremaining: 7.31s\n", + "412:\tlearn: 16717199.5760035\ttotal: 5.13s\tremaining: 7.3s\n", + "413:\tlearn: 16713362.4492616\ttotal: 5.15s\tremaining: 7.29s\n", + "414:\tlearn: 16712806.0473182\ttotal: 5.16s\tremaining: 7.28s\n", + "415:\tlearn: 16711241.9902750\ttotal: 5.18s\tremaining: 7.27s\n", + "416:\tlearn: 16710626.7325455\ttotal: 5.19s\tremaining: 7.25s\n", + "417:\tlearn: 16644768.4542531\ttotal: 5.2s\tremaining: 7.24s\n", + "418:\tlearn: 16644403.8081224\ttotal: 5.21s\tremaining: 7.23s\n", + "419:\tlearn: 16644106.9601552\ttotal: 5.22s\tremaining: 7.21s\n", + "420:\tlearn: 16643628.6346956\ttotal: 5.24s\tremaining: 7.2s\n", + "421:\tlearn: 16640073.3813320\ttotal: 5.25s\tremaining: 7.19s\n", + "422:\tlearn: 16639549.7950808\ttotal: 5.26s\tremaining: 7.18s\n", + "423:\tlearn: 16639069.1006878\ttotal: 5.27s\tremaining: 7.17s\n", + "424:\tlearn: 16638481.2382327\ttotal: 5.29s\tremaining: 7.15s\n", + "425:\tlearn: 16638208.9073863\ttotal: 5.3s\tremaining: 7.14s\n", + "426:\tlearn: 16609090.9227109\ttotal: 5.31s\tremaining: 7.13s\n", + "427:\tlearn: 16607897.8537223\ttotal: 5.33s\tremaining: 7.13s\n", + "428:\tlearn: 16607613.0069443\ttotal: 5.34s\tremaining: 7.11s\n", + "429:\tlearn: 16603866.7848843\ttotal: 5.36s\tremaining: 7.11s\n", + "430:\tlearn: 16566652.4020620\ttotal: 5.37s\tremaining: 7.09s\n", + "431:\tlearn: 16566149.6048169\ttotal: 5.39s\tremaining: 7.08s\n", + "432:\tlearn: 16564672.1011733\ttotal: 5.4s\tremaining: 7.07s\n", + "433:\tlearn: 16564610.7741058\ttotal: 5.41s\tremaining: 7.06s\n", + "434:\tlearn: 16564198.8911273\ttotal: 5.42s\tremaining: 7.04s\n", + "435:\tlearn: 16559675.2968062\ttotal: 5.43s\tremaining: 7.03s\n", + "436:\tlearn: 16558753.2346339\ttotal: 5.45s\tremaining: 7.02s\n", + "437:\tlearn: 16558452.1907641\ttotal: 5.46s\tremaining: 7.01s\n", + "438:\tlearn: 16546587.2383006\ttotal: 5.47s\tremaining: 6.99s\n", + "439:\tlearn: 16543823.0847287\ttotal: 5.49s\tremaining: 6.98s\n", + "440:\tlearn: 16542126.8424469\ttotal: 5.5s\tremaining: 6.97s\n", + "441:\tlearn: 16541624.1632076\ttotal: 5.51s\tremaining: 6.96s\n", + "442:\tlearn: 16540326.5322872\ttotal: 5.52s\tremaining: 6.95s\n", + "443:\tlearn: 16530336.2084291\ttotal: 5.54s\tremaining: 6.93s\n", + "444:\tlearn: 16530167.9665629\ttotal: 5.55s\tremaining: 6.92s\n", + "445:\tlearn: 16528821.2477933\ttotal: 5.56s\tremaining: 6.91s\n", + "446:\tlearn: 16528766.2012617\ttotal: 5.57s\tremaining: 6.89s\n", + "447:\tlearn: 16518018.7193100\ttotal: 5.59s\tremaining: 6.88s\n", + "448:\tlearn: 16508723.6897544\ttotal: 5.6s\tremaining: 6.87s\n", + "449:\tlearn: 16508487.2637814\ttotal: 5.61s\tremaining: 6.86s\n", + "450:\tlearn: 16473955.6540161\ttotal: 5.63s\tremaining: 6.85s\n", + "451:\tlearn: 16453172.0203944\ttotal: 5.64s\tremaining: 6.84s\n", + "452:\tlearn: 16451483.6324413\ttotal: 5.65s\tremaining: 6.82s\n", + "453:\tlearn: 16451257.8036014\ttotal: 5.67s\tremaining: 6.81s\n", + "454:\tlearn: 16448369.9508352\ttotal: 5.68s\tremaining: 6.8s\n", + "455:\tlearn: 16446719.1385193\ttotal: 5.69s\tremaining: 6.79s\n", + "456:\tlearn: 16420736.1369659\ttotal: 5.71s\tremaining: 6.78s\n", + "457:\tlearn: 16420629.2824606\ttotal: 5.72s\tremaining: 6.77s\n", + "458:\tlearn: 16420336.6729748\ttotal: 5.73s\tremaining: 6.76s\n", + "459:\tlearn: 16420155.4584530\ttotal: 5.74s\tremaining: 6.74s\n", + "460:\tlearn: 16419734.8233202\ttotal: 5.76s\tremaining: 6.73s\n", + "461:\tlearn: 16419517.6225944\ttotal: 5.77s\tremaining: 6.72s\n", + "462:\tlearn: 16406145.7183320\ttotal: 5.78s\tremaining: 6.71s\n", + "463:\tlearn: 16404609.0651931\ttotal: 5.8s\tremaining: 6.7s\n", + "464:\tlearn: 16404332.0732862\ttotal: 5.81s\tremaining: 6.68s\n", + "465:\tlearn: 16404019.7507952\ttotal: 5.82s\tremaining: 6.67s\n", + "466:\tlearn: 16403507.0137349\ttotal: 5.83s\tremaining: 6.66s\n", + "467:\tlearn: 16402993.5886996\ttotal: 5.84s\tremaining: 6.64s\n", + "468:\tlearn: 16385955.8460101\ttotal: 5.86s\tremaining: 6.63s\n", + "469:\tlearn: 16373237.2004642\ttotal: 5.87s\tremaining: 6.62s\n", + "470:\tlearn: 16373038.3665164\ttotal: 5.88s\tremaining: 6.6s\n", + "471:\tlearn: 16372801.5860356\ttotal: 5.89s\tremaining: 6.59s\n", + "472:\tlearn: 16360759.6605520\ttotal: 5.9s\tremaining: 6.58s\n", + "473:\tlearn: 16360169.9657388\ttotal: 5.92s\tremaining: 6.57s\n", + "474:\tlearn: 16351841.0373273\ttotal: 5.93s\tremaining: 6.55s\n", + "475:\tlearn: 16349809.4004009\ttotal: 5.94s\tremaining: 6.54s\n", + "476:\tlearn: 16344483.1074475\ttotal: 5.96s\tremaining: 6.53s\n", + "477:\tlearn: 16340922.7262468\ttotal: 5.97s\tremaining: 6.52s\n", + "478:\tlearn: 16334736.4373107\ttotal: 5.98s\tremaining: 6.5s\n", + "479:\tlearn: 16334043.7402281\ttotal: 5.99s\tremaining: 6.49s\n", + "480:\tlearn: 16333745.0129155\ttotal: 6s\tremaining: 6.48s\n", + "481:\tlearn: 16332170.0024156\ttotal: 6.02s\tremaining: 6.46s\n", + "482:\tlearn: 16331680.4256261\ttotal: 6.03s\tremaining: 6.45s\n", + "483:\tlearn: 16321943.5880137\ttotal: 6.04s\tremaining: 6.44s\n", + "484:\tlearn: 16313566.1128530\ttotal: 6.05s\tremaining: 6.43s\n", + "485:\tlearn: 16312784.3783495\ttotal: 6.07s\tremaining: 6.42s\n", + "486:\tlearn: 16304256.8971602\ttotal: 6.08s\tremaining: 6.4s\n", + "487:\tlearn: 16299338.9360929\ttotal: 6.09s\tremaining: 6.39s\n", + "488:\tlearn: 16298399.2768748\ttotal: 6.11s\tremaining: 6.38s\n", + "489:\tlearn: 16282861.5959599\ttotal: 6.12s\tremaining: 6.37s\n", + "490:\tlearn: 16278027.7798172\ttotal: 6.13s\tremaining: 6.36s\n", + "491:\tlearn: 16262455.7433251\ttotal: 6.15s\tremaining: 6.35s\n", + "492:\tlearn: 16254609.6670435\ttotal: 6.16s\tremaining: 6.33s\n", + "493:\tlearn: 16250306.9197526\ttotal: 6.17s\tremaining: 6.32s\n", + "494:\tlearn: 16249855.9315045\ttotal: 6.19s\tremaining: 6.31s\n", + "495:\tlearn: 16248555.7562997\ttotal: 6.2s\tremaining: 6.3s\n", + "496:\tlearn: 16247555.1566330\ttotal: 6.21s\tremaining: 6.29s\n", + "497:\tlearn: 16247235.5993966\ttotal: 6.22s\tremaining: 6.27s\n", + "498:\tlearn: 16246264.7483105\ttotal: 6.23s\tremaining: 6.26s\n", + "499:\tlearn: 16246007.7491962\ttotal: 6.24s\tremaining: 6.24s\n", + "500:\tlearn: 16222867.6954421\ttotal: 6.25s\tremaining: 6.23s\n", + "501:\tlearn: 16222688.8853061\ttotal: 6.27s\tremaining: 6.22s\n", + "502:\tlearn: 16217885.3385915\ttotal: 6.28s\tremaining: 6.2s\n", + "503:\tlearn: 16217409.1580145\ttotal: 6.29s\tremaining: 6.19s\n", + "504:\tlearn: 16216838.3191240\ttotal: 6.3s\tremaining: 6.18s\n", + "505:\tlearn: 16216329.9777509\ttotal: 6.32s\tremaining: 6.17s\n", + "506:\tlearn: 16201534.4156055\ttotal: 6.33s\tremaining: 6.16s\n", + "507:\tlearn: 16198138.1904772\ttotal: 6.34s\tremaining: 6.14s\n", + "508:\tlearn: 16197904.2583705\ttotal: 6.35s\tremaining: 6.13s\n", + "509:\tlearn: 16193656.6407621\ttotal: 6.37s\tremaining: 6.12s\n", + "510:\tlearn: 16180805.8618897\ttotal: 6.38s\tremaining: 6.1s\n", + "511:\tlearn: 16176908.1769610\ttotal: 6.39s\tremaining: 6.09s\n", + "512:\tlearn: 16168261.0438871\ttotal: 6.4s\tremaining: 6.08s\n", + "513:\tlearn: 16167754.4165306\ttotal: 6.42s\tremaining: 6.07s\n", + "514:\tlearn: 16166295.0362243\ttotal: 6.43s\tremaining: 6.05s\n", + "515:\tlearn: 16166058.4053693\ttotal: 6.44s\tremaining: 6.04s\n", + "516:\tlearn: 16155412.7707338\ttotal: 6.45s\tremaining: 6.03s\n", + "517:\tlearn: 16152266.1742558\ttotal: 6.46s\tremaining: 6.02s\n", + "518:\tlearn: 16151552.8907870\ttotal: 6.48s\tremaining: 6s\n", + "519:\tlearn: 16140281.4351978\ttotal: 6.49s\tremaining: 5.99s\n", + "520:\tlearn: 16133450.4403783\ttotal: 6.5s\tremaining: 5.98s\n", + "521:\tlearn: 16132209.1334220\ttotal: 6.51s\tremaining: 5.97s\n", + "522:\tlearn: 16118104.6552795\ttotal: 6.53s\tremaining: 5.95s\n", + "523:\tlearn: 16108764.2393062\ttotal: 6.54s\tremaining: 5.94s\n", + "524:\tlearn: 16108234.0634605\ttotal: 6.55s\tremaining: 5.93s\n", + "525:\tlearn: 16107619.6760099\ttotal: 6.57s\tremaining: 5.92s\n", + "526:\tlearn: 16104870.7442280\ttotal: 6.58s\tremaining: 5.9s\n", + "527:\tlearn: 16102428.3934069\ttotal: 6.59s\tremaining: 5.89s\n", + "528:\tlearn: 16102157.2857565\ttotal: 6.6s\tremaining: 5.88s\n", + "529:\tlearn: 16101584.7403855\ttotal: 6.61s\tremaining: 5.86s\n", + "530:\tlearn: 16101480.3344969\ttotal: 6.63s\tremaining: 5.85s\n", + "531:\tlearn: 16100595.6548675\ttotal: 6.64s\tremaining: 5.84s\n", + "532:\tlearn: 16097511.0825233\ttotal: 6.65s\tremaining: 5.83s\n", + "533:\tlearn: 16096615.9743637\ttotal: 6.66s\tremaining: 5.81s\n", + "534:\tlearn: 16096369.6922988\ttotal: 6.67s\tremaining: 5.8s\n", + "535:\tlearn: 16095946.1647864\ttotal: 6.69s\tremaining: 5.79s\n", + "536:\tlearn: 16095637.6185090\ttotal: 6.7s\tremaining: 5.78s\n", + "537:\tlearn: 16094682.0243853\ttotal: 6.71s\tremaining: 5.76s\n", + "538:\tlearn: 16094291.9050311\ttotal: 6.72s\tremaining: 5.75s\n", + "539:\tlearn: 16093984.5280001\ttotal: 6.73s\tremaining: 5.73s\n", + "540:\tlearn: 16090374.6401334\ttotal: 6.75s\tremaining: 5.72s\n", + "541:\tlearn: 16090226.8772271\ttotal: 6.76s\tremaining: 5.71s\n", + "542:\tlearn: 16090050.1805201\ttotal: 6.77s\tremaining: 5.7s\n", + "543:\tlearn: 16069181.1048944\ttotal: 6.78s\tremaining: 5.68s\n", + "544:\tlearn: 16068504.9399291\ttotal: 6.79s\tremaining: 5.67s\n", + "545:\tlearn: 16068245.3744393\ttotal: 6.8s\tremaining: 5.66s\n", + "546:\tlearn: 16065773.4114093\ttotal: 6.82s\tremaining: 5.65s\n", + "547:\tlearn: 16051662.5046318\ttotal: 6.83s\tremaining: 5.63s\n", + "548:\tlearn: 16035327.2446945\ttotal: 6.84s\tremaining: 5.62s\n", + "549:\tlearn: 16035199.2858857\ttotal: 6.86s\tremaining: 5.61s\n", + "550:\tlearn: 16033842.9666151\ttotal: 6.87s\tremaining: 5.6s\n", + "551:\tlearn: 15995073.4381976\ttotal: 6.88s\tremaining: 5.58s\n", + "552:\tlearn: 15994812.5505379\ttotal: 6.89s\tremaining: 5.57s\n", + "553:\tlearn: 15994595.9921031\ttotal: 6.91s\tremaining: 5.56s\n", + "554:\tlearn: 15992248.3834318\ttotal: 6.92s\tremaining: 5.55s\n", + "555:\tlearn: 15992027.4484601\ttotal: 6.93s\tremaining: 5.54s\n", + "556:\tlearn: 15990566.0719983\ttotal: 6.95s\tremaining: 5.52s\n", + "557:\tlearn: 15985609.0920187\ttotal: 6.96s\tremaining: 5.51s\n", + "558:\tlearn: 15984517.8156083\ttotal: 6.97s\tremaining: 5.5s\n", + "559:\tlearn: 15958775.9803743\ttotal: 6.99s\tremaining: 5.49s\n", + "560:\tlearn: 15958166.8639855\ttotal: 7s\tremaining: 5.48s\n", + "561:\tlearn: 15949224.5334582\ttotal: 7.01s\tremaining: 5.47s\n", + "562:\tlearn: 15948769.9101270\ttotal: 7.03s\tremaining: 5.45s\n", + "563:\tlearn: 15930009.9576761\ttotal: 7.04s\tremaining: 5.44s\n", + "564:\tlearn: 15917439.6202170\ttotal: 7.05s\tremaining: 5.43s\n", + "565:\tlearn: 15908669.4567536\ttotal: 7.07s\tremaining: 5.42s\n", + "566:\tlearn: 15908084.2939630\ttotal: 7.08s\tremaining: 5.41s\n", + "567:\tlearn: 15906697.1590494\ttotal: 7.09s\tremaining: 5.39s\n", + "568:\tlearn: 15906522.4609846\ttotal: 7.1s\tremaining: 5.38s\n", + "569:\tlearn: 15906139.9138507\ttotal: 7.12s\tremaining: 5.37s\n", + "570:\tlearn: 15905855.0642382\ttotal: 7.13s\tremaining: 5.35s\n", + "571:\tlearn: 15897372.3501416\ttotal: 7.14s\tremaining: 5.34s\n", + "572:\tlearn: 15893536.4661240\ttotal: 7.15s\tremaining: 5.33s\n", + "573:\tlearn: 15893206.2810918\ttotal: 7.17s\tremaining: 5.32s\n", + "574:\tlearn: 15892918.2703602\ttotal: 7.18s\tremaining: 5.3s\n", + "575:\tlearn: 15892752.8029869\ttotal: 7.19s\tremaining: 5.29s\n", + "576:\tlearn: 15885169.7413434\ttotal: 7.2s\tremaining: 5.28s\n", + "577:\tlearn: 15884936.8745209\ttotal: 7.21s\tremaining: 5.27s\n", + "578:\tlearn: 15876877.1991641\ttotal: 7.22s\tremaining: 5.25s\n", + "579:\tlearn: 15865774.4061534\ttotal: 7.24s\tremaining: 5.24s\n", + "580:\tlearn: 15859212.9207966\ttotal: 7.25s\tremaining: 5.23s\n", + "581:\tlearn: 15858807.6511813\ttotal: 7.26s\tremaining: 5.21s\n", + "582:\tlearn: 15850129.9468116\ttotal: 7.27s\tremaining: 5.2s\n", + "583:\tlearn: 15845554.5689368\ttotal: 7.29s\tremaining: 5.19s\n", + "584:\tlearn: 15844986.6765475\ttotal: 7.3s\tremaining: 5.18s\n", + "585:\tlearn: 15844796.1180439\ttotal: 7.31s\tremaining: 5.17s\n", + "586:\tlearn: 15844586.1630771\ttotal: 7.33s\tremaining: 5.15s\n", + "587:\tlearn: 15827685.4584540\ttotal: 7.34s\tremaining: 5.14s\n", + "588:\tlearn: 15826910.6044821\ttotal: 7.35s\tremaining: 5.13s\n", + "589:\tlearn: 15824060.9875073\ttotal: 7.36s\tremaining: 5.12s\n", + "590:\tlearn: 15818523.6912985\ttotal: 7.38s\tremaining: 5.1s\n", + "591:\tlearn: 15810640.6921394\ttotal: 7.39s\tremaining: 5.09s\n", + "592:\tlearn: 15795481.4197185\ttotal: 7.4s\tremaining: 5.08s\n", + "593:\tlearn: 15795256.4491006\ttotal: 7.41s\tremaining: 5.07s\n", + "594:\tlearn: 15784420.7363473\ttotal: 7.43s\tremaining: 5.05s\n", + "595:\tlearn: 15784290.1819258\ttotal: 7.44s\tremaining: 5.04s\n", + "596:\tlearn: 15783955.0773924\ttotal: 7.45s\tremaining: 5.03s\n", + "597:\tlearn: 15781518.5372107\ttotal: 7.46s\tremaining: 5.02s\n", + "598:\tlearn: 15779547.4210947\ttotal: 7.48s\tremaining: 5s\n", + "599:\tlearn: 15777334.3663340\ttotal: 7.49s\tremaining: 4.99s\n", + "600:\tlearn: 15774774.7721883\ttotal: 7.5s\tremaining: 4.98s\n", + "601:\tlearn: 15774672.2356339\ttotal: 7.51s\tremaining: 4.97s\n", + "602:\tlearn: 15773528.0736833\ttotal: 7.53s\tremaining: 4.96s\n", + "603:\tlearn: 15768721.3649454\ttotal: 7.54s\tremaining: 4.94s\n", + "604:\tlearn: 15768502.0877019\ttotal: 7.55s\tremaining: 4.93s\n", + "605:\tlearn: 15768057.4929247\ttotal: 7.57s\tremaining: 4.92s\n", + "606:\tlearn: 15767950.4285043\ttotal: 7.58s\tremaining: 4.91s\n", + "607:\tlearn: 15767445.1324607\ttotal: 7.59s\tremaining: 4.89s\n", + "608:\tlearn: 15767269.0628064\ttotal: 7.6s\tremaining: 4.88s\n", + "609:\tlearn: 15767020.8174624\ttotal: 7.61s\tremaining: 4.87s\n", + "610:\tlearn: 15762309.5160245\ttotal: 7.62s\tremaining: 4.85s\n", + "611:\tlearn: 15757527.7718093\ttotal: 7.63s\tremaining: 4.84s\n", + "612:\tlearn: 15757150.7731734\ttotal: 7.65s\tremaining: 4.83s\n", + "613:\tlearn: 15756885.6252756\ttotal: 7.66s\tremaining: 4.81s\n", + "614:\tlearn: 15755584.6816303\ttotal: 7.67s\tremaining: 4.8s\n", + "615:\tlearn: 15755485.6737331\ttotal: 7.68s\tremaining: 4.79s\n", + "616:\tlearn: 15754432.0517599\ttotal: 7.7s\tremaining: 4.78s\n", + "617:\tlearn: 15744535.8203508\ttotal: 7.71s\tremaining: 4.76s\n", + "618:\tlearn: 15740683.5538600\ttotal: 7.72s\tremaining: 4.75s\n", + "619:\tlearn: 15736903.5667213\ttotal: 7.73s\tremaining: 4.74s\n", + "620:\tlearn: 15736355.4210963\ttotal: 7.75s\tremaining: 4.73s\n", + "621:\tlearn: 15729940.4032081\ttotal: 7.76s\tremaining: 4.71s\n", + "622:\tlearn: 15729775.7542976\ttotal: 7.77s\tremaining: 4.7s\n", + "623:\tlearn: 15726578.4125003\ttotal: 7.78s\tremaining: 4.69s\n", + "624:\tlearn: 15713451.5317183\ttotal: 7.8s\tremaining: 4.68s\n", + "625:\tlearn: 15712116.7478338\ttotal: 7.81s\tremaining: 4.67s\n", + "626:\tlearn: 15712039.0336448\ttotal: 7.82s\tremaining: 4.65s\n", + "627:\tlearn: 15711687.4136682\ttotal: 7.84s\tremaining: 4.64s\n", + "628:\tlearn: 15711298.6681597\ttotal: 7.85s\tremaining: 4.63s\n", + "629:\tlearn: 15705228.0021081\ttotal: 7.86s\tremaining: 4.62s\n", + "630:\tlearn: 15705060.0247650\ttotal: 7.88s\tremaining: 4.61s\n", + "631:\tlearn: 15702513.1910574\ttotal: 7.89s\tremaining: 4.6s\n", + "632:\tlearn: 15702203.6145508\ttotal: 7.91s\tremaining: 4.58s\n", + "633:\tlearn: 15698975.2951288\ttotal: 7.92s\tremaining: 4.57s\n", + "634:\tlearn: 15694674.2341421\ttotal: 7.94s\tremaining: 4.56s\n", + "635:\tlearn: 15686913.9104937\ttotal: 7.95s\tremaining: 4.55s\n", + "636:\tlearn: 15686809.9586513\ttotal: 7.96s\tremaining: 4.54s\n", + "637:\tlearn: 15685604.3787689\ttotal: 7.98s\tremaining: 4.53s\n", + "638:\tlearn: 15685081.4917552\ttotal: 7.99s\tremaining: 4.51s\n", + "639:\tlearn: 15676541.2826685\ttotal: 8s\tremaining: 4.5s\n", + "640:\tlearn: 15672855.0180760\ttotal: 8.02s\tremaining: 4.49s\n", + "641:\tlearn: 15666780.7593096\ttotal: 8.03s\tremaining: 4.48s\n", + "642:\tlearn: 15659438.3508408\ttotal: 8.04s\tremaining: 4.46s\n", + "643:\tlearn: 15653755.4598701\ttotal: 8.05s\tremaining: 4.45s\n", + "644:\tlearn: 15652034.4638985\ttotal: 8.07s\tremaining: 4.44s\n", + "645:\tlearn: 15645095.6489597\ttotal: 8.08s\tremaining: 4.43s\n", + "646:\tlearn: 15641226.9420905\ttotal: 8.09s\tremaining: 4.42s\n", + "647:\tlearn: 15639833.9184524\ttotal: 8.11s\tremaining: 4.4s\n", + "648:\tlearn: 15639581.1651510\ttotal: 8.12s\tremaining: 4.39s\n", + "649:\tlearn: 15635923.3848062\ttotal: 8.13s\tremaining: 4.38s\n", + "650:\tlearn: 15635813.1152459\ttotal: 8.14s\tremaining: 4.37s\n", + "651:\tlearn: 15635469.3555938\ttotal: 8.16s\tremaining: 4.35s\n", + "652:\tlearn: 15635363.9174910\ttotal: 8.17s\tremaining: 4.34s\n", + "653:\tlearn: 15633936.7433448\ttotal: 8.18s\tremaining: 4.33s\n", + "654:\tlearn: 15633839.2271448\ttotal: 8.19s\tremaining: 4.31s\n", + "655:\tlearn: 15633735.8610291\ttotal: 8.21s\tremaining: 4.3s\n", + "656:\tlearn: 15633309.8063070\ttotal: 8.22s\tremaining: 4.29s\n", + "657:\tlearn: 15632683.8986677\ttotal: 8.23s\tremaining: 4.28s\n", + "658:\tlearn: 15632461.2639014\ttotal: 8.24s\tremaining: 4.26s\n", + "659:\tlearn: 15627123.1765533\ttotal: 8.26s\tremaining: 4.25s\n", + "660:\tlearn: 15626996.0787558\ttotal: 8.27s\tremaining: 4.24s\n", + "661:\tlearn: 15624291.0204091\ttotal: 8.28s\tremaining: 4.23s\n", + "662:\tlearn: 15617684.3098363\ttotal: 8.29s\tremaining: 4.21s\n", + "663:\tlearn: 15611967.2176796\ttotal: 8.31s\tremaining: 4.2s\n", + "664:\tlearn: 15598472.2546786\ttotal: 8.32s\tremaining: 4.19s\n", + "665:\tlearn: 15597526.0470563\ttotal: 8.33s\tremaining: 4.18s\n", + "666:\tlearn: 15597430.3920481\ttotal: 8.34s\tremaining: 4.16s\n", + "667:\tlearn: 15596422.7059295\ttotal: 8.35s\tremaining: 4.15s\n", + "668:\tlearn: 15591400.2242411\ttotal: 8.37s\tremaining: 4.14s\n", + "669:\tlearn: 15585199.5277811\ttotal: 8.38s\tremaining: 4.13s\n", + "670:\tlearn: 15585003.5063693\ttotal: 8.39s\tremaining: 4.11s\n", + "671:\tlearn: 15578765.7193891\ttotal: 8.4s\tremaining: 4.1s\n", + "672:\tlearn: 15577252.0151364\ttotal: 8.41s\tremaining: 4.09s\n", + "673:\tlearn: 15576511.8797514\ttotal: 8.43s\tremaining: 4.08s\n", + "674:\tlearn: 15576120.7606092\ttotal: 8.44s\tremaining: 4.06s\n", + "675:\tlearn: 15574398.5273782\ttotal: 8.45s\tremaining: 4.05s\n", + "676:\tlearn: 15565660.7493905\ttotal: 8.46s\tremaining: 4.04s\n", + "677:\tlearn: 15561009.3437211\ttotal: 8.48s\tremaining: 4.03s\n", + "678:\tlearn: 15548878.4770401\ttotal: 8.49s\tremaining: 4.01s\n", + "679:\tlearn: 15527713.9632219\ttotal: 8.5s\tremaining: 4s\n", + "680:\tlearn: 15519745.2151864\ttotal: 8.51s\tremaining: 3.99s\n", + "681:\tlearn: 15519391.2760902\ttotal: 8.53s\tremaining: 3.98s\n", + "682:\tlearn: 15514461.8611265\ttotal: 8.54s\tremaining: 3.96s\n", + "683:\tlearn: 15514296.1001141\ttotal: 8.55s\tremaining: 3.95s\n", + "684:\tlearn: 15514204.8658979\ttotal: 8.56s\tremaining: 3.94s\n", + "685:\tlearn: 15513977.3554214\ttotal: 8.57s\tremaining: 3.92s\n", + "686:\tlearn: 15513906.5046745\ttotal: 8.59s\tremaining: 3.91s\n", + "687:\tlearn: 15513701.8112778\ttotal: 8.6s\tremaining: 3.9s\n", + "688:\tlearn: 15513602.4959013\ttotal: 8.61s\tremaining: 3.88s\n", + "689:\tlearn: 15513510.7910896\ttotal: 8.62s\tremaining: 3.87s\n", + "690:\tlearn: 15513352.2070048\ttotal: 8.64s\tremaining: 3.86s\n", + "691:\tlearn: 15513238.8204588\ttotal: 8.65s\tremaining: 3.85s\n", + "692:\tlearn: 15513154.0618557\ttotal: 8.66s\tremaining: 3.83s\n", + "693:\tlearn: 15512878.9114412\ttotal: 8.67s\tremaining: 3.82s\n", + "694:\tlearn: 15509248.2055515\ttotal: 8.68s\tremaining: 3.81s\n", + "695:\tlearn: 15508734.7327170\ttotal: 8.69s\tremaining: 3.8s\n", + "696:\tlearn: 15508495.7881550\ttotal: 8.7s\tremaining: 3.78s\n", + "697:\tlearn: 15508349.6872134\ttotal: 8.72s\tremaining: 3.77s\n", + "698:\tlearn: 15508190.6588965\ttotal: 8.73s\tremaining: 3.76s\n", + "699:\tlearn: 15508018.8419773\ttotal: 8.74s\tremaining: 3.75s\n", + "700:\tlearn: 15507826.3791202\ttotal: 8.75s\tremaining: 3.73s\n", + "701:\tlearn: 15507705.8100928\ttotal: 8.76s\tremaining: 3.72s\n", + "702:\tlearn: 15507533.3512682\ttotal: 8.77s\tremaining: 3.71s\n", + "703:\tlearn: 15501571.2913355\ttotal: 8.78s\tremaining: 3.69s\n", + "704:\tlearn: 15495921.0773672\ttotal: 8.8s\tremaining: 3.68s\n", + "705:\tlearn: 15495385.0875416\ttotal: 8.81s\tremaining: 3.67s\n", + "706:\tlearn: 15495191.8032918\ttotal: 8.82s\tremaining: 3.65s\n", + "707:\tlearn: 15494128.9589635\ttotal: 8.83s\tremaining: 3.64s\n", + "708:\tlearn: 15493806.9566177\ttotal: 8.84s\tremaining: 3.63s\n", + "709:\tlearn: 15493694.0465547\ttotal: 8.85s\tremaining: 3.62s\n", + "710:\tlearn: 15493305.1729869\ttotal: 8.87s\tremaining: 3.6s\n", + "711:\tlearn: 15487948.0399475\ttotal: 8.88s\tremaining: 3.59s\n", + "712:\tlearn: 15487843.0916850\ttotal: 8.89s\tremaining: 3.58s\n", + "713:\tlearn: 15482765.7669785\ttotal: 8.9s\tremaining: 3.56s\n", + "714:\tlearn: 15474767.3580796\ttotal: 8.91s\tremaining: 3.55s\n", + "715:\tlearn: 15472407.8166003\ttotal: 8.93s\tremaining: 3.54s\n", + "716:\tlearn: 15467592.3874842\ttotal: 8.94s\tremaining: 3.53s\n", + "717:\tlearn: 15467435.4901525\ttotal: 8.95s\tremaining: 3.51s\n", + "718:\tlearn: 15462871.2869120\ttotal: 8.96s\tremaining: 3.5s\n", + "719:\tlearn: 15462771.0380185\ttotal: 8.97s\tremaining: 3.49s\n", + "720:\tlearn: 15462475.2715024\ttotal: 8.98s\tremaining: 3.48s\n", + "721:\tlearn: 15454885.7938423\ttotal: 8.99s\tremaining: 3.46s\n", + "722:\tlearn: 15450557.5824215\ttotal: 9.01s\tremaining: 3.45s\n", + "723:\tlearn: 15446455.2317749\ttotal: 9.02s\tremaining: 3.44s\n", + "724:\tlearn: 15445711.0004476\ttotal: 9.03s\tremaining: 3.43s\n", + "725:\tlearn: 15441822.5331613\ttotal: 9.05s\tremaining: 3.41s\n", + "726:\tlearn: 15441182.6843715\ttotal: 9.06s\tremaining: 3.4s\n", + "727:\tlearn: 15441088.8915881\ttotal: 9.07s\tremaining: 3.39s\n", + "728:\tlearn: 15441002.5272406\ttotal: 9.09s\tremaining: 3.38s\n", + "729:\tlearn: 15440884.2830869\ttotal: 9.1s\tremaining: 3.37s\n", + "730:\tlearn: 15440734.2579995\ttotal: 9.11s\tremaining: 3.35s\n", + "731:\tlearn: 15440611.8887909\ttotal: 9.13s\tremaining: 3.34s\n", + "732:\tlearn: 15440249.4221271\ttotal: 9.14s\tremaining: 3.33s\n", + "733:\tlearn: 15440158.8154476\ttotal: 9.15s\tremaining: 3.32s\n", + "734:\tlearn: 15436472.7845071\ttotal: 9.16s\tremaining: 3.3s\n", + "735:\tlearn: 15433672.4484876\ttotal: 9.18s\tremaining: 3.29s\n", + "736:\tlearn: 15433490.1840146\ttotal: 9.19s\tremaining: 3.28s\n", + "737:\tlearn: 15433308.6881010\ttotal: 9.2s\tremaining: 3.27s\n", + "738:\tlearn: 15433042.7409848\ttotal: 9.21s\tremaining: 3.25s\n", + "739:\tlearn: 15432541.8769518\ttotal: 9.23s\tremaining: 3.24s\n", + "740:\tlearn: 15431856.6761047\ttotal: 9.24s\tremaining: 3.23s\n", + "741:\tlearn: 15431804.7359345\ttotal: 9.25s\tremaining: 3.22s\n", + "742:\tlearn: 15427355.2392047\ttotal: 9.26s\tremaining: 3.2s\n", + "743:\tlearn: 15427218.1028185\ttotal: 9.27s\tremaining: 3.19s\n", + "744:\tlearn: 15424332.1093472\ttotal: 9.29s\tremaining: 3.18s\n", + "745:\tlearn: 15388321.8033125\ttotal: 9.3s\tremaining: 3.17s\n", + "746:\tlearn: 15377267.9048803\ttotal: 9.31s\tremaining: 3.15s\n", + "747:\tlearn: 15374625.6198420\ttotal: 9.32s\tremaining: 3.14s\n", + "748:\tlearn: 15370386.0426691\ttotal: 9.34s\tremaining: 3.13s\n", + "749:\tlearn: 15359901.4299089\ttotal: 9.35s\tremaining: 3.12s\n", + "750:\tlearn: 15358774.7332579\ttotal: 9.36s\tremaining: 3.1s\n", + "751:\tlearn: 15358651.8711020\ttotal: 9.37s\tremaining: 3.09s\n", + "752:\tlearn: 15358300.8764559\ttotal: 9.38s\tremaining: 3.08s\n", + "753:\tlearn: 15357884.8170886\ttotal: 9.4s\tremaining: 3.06s\n", + "754:\tlearn: 15357643.0994172\ttotal: 9.41s\tremaining: 3.05s\n", + "755:\tlearn: 15357565.0887636\ttotal: 9.42s\tremaining: 3.04s\n", + "756:\tlearn: 15351820.0777339\ttotal: 9.43s\tremaining: 3.03s\n", + "757:\tlearn: 15351414.2517094\ttotal: 9.44s\tremaining: 3.01s\n", + "758:\tlearn: 15349501.7532204\ttotal: 9.46s\tremaining: 3s\n", + "759:\tlearn: 15348526.7586048\ttotal: 9.47s\tremaining: 2.99s\n", + "760:\tlearn: 15348352.7244253\ttotal: 9.48s\tremaining: 2.98s\n", + "761:\tlearn: 15347292.4488773\ttotal: 9.49s\tremaining: 2.96s\n", + "762:\tlearn: 15347207.8865499\ttotal: 9.51s\tremaining: 2.95s\n", + "763:\tlearn: 15342959.4790246\ttotal: 9.52s\tremaining: 2.94s\n", + "764:\tlearn: 15342697.6483068\ttotal: 9.53s\tremaining: 2.93s\n", + "765:\tlearn: 15313028.1511576\ttotal: 9.54s\tremaining: 2.92s\n", + "766:\tlearn: 15310947.6382018\ttotal: 9.56s\tremaining: 2.9s\n", + "767:\tlearn: 15263340.9642127\ttotal: 9.57s\tremaining: 2.89s\n", + "768:\tlearn: 15259807.2026083\ttotal: 9.58s\tremaining: 2.88s\n", + "769:\tlearn: 15259504.1148296\ttotal: 9.59s\tremaining: 2.87s\n", + "770:\tlearn: 15259395.2637694\ttotal: 9.6s\tremaining: 2.85s\n", + "771:\tlearn: 15256045.7141942\ttotal: 9.61s\tremaining: 2.84s\n", + "772:\tlearn: 15252870.9021417\ttotal: 9.63s\tremaining: 2.83s\n", + "773:\tlearn: 15248430.3201074\ttotal: 9.64s\tremaining: 2.81s\n", + "774:\tlearn: 15246752.4177458\ttotal: 9.65s\tremaining: 2.8s\n", + "775:\tlearn: 15245960.1417687\ttotal: 9.67s\tremaining: 2.79s\n", + "776:\tlearn: 15245917.6645107\ttotal: 9.68s\tremaining: 2.78s\n", + "777:\tlearn: 15245817.5185452\ttotal: 9.69s\tremaining: 2.77s\n", + "778:\tlearn: 15245619.8351855\ttotal: 9.7s\tremaining: 2.75s\n", + "779:\tlearn: 15244869.5667520\ttotal: 9.71s\tremaining: 2.74s\n", + "780:\tlearn: 15244818.8943236\ttotal: 9.72s\tremaining: 2.73s\n", + "781:\tlearn: 15244254.3637038\ttotal: 9.74s\tremaining: 2.71s\n", + "782:\tlearn: 15243818.2939855\ttotal: 9.75s\tremaining: 2.7s\n", + "783:\tlearn: 15243668.1645179\ttotal: 9.76s\tremaining: 2.69s\n", + "784:\tlearn: 15240656.8617467\ttotal: 9.77s\tremaining: 2.68s\n", + "785:\tlearn: 15237802.6637690\ttotal: 9.78s\tremaining: 2.66s\n", + "786:\tlearn: 15235097.3769887\ttotal: 9.79s\tremaining: 2.65s\n", + "787:\tlearn: 15231063.4576018\ttotal: 9.81s\tremaining: 2.64s\n", + "788:\tlearn: 15224406.4239600\ttotal: 9.82s\tremaining: 2.63s\n", + "789:\tlearn: 15220791.2846445\ttotal: 9.83s\tremaining: 2.61s\n", + "790:\tlearn: 15220221.4180094\ttotal: 9.85s\tremaining: 2.6s\n", + "791:\tlearn: 15220121.7499013\ttotal: 9.86s\tremaining: 2.59s\n", + "792:\tlearn: 15218396.9325757\ttotal: 9.87s\tremaining: 2.58s\n", + "793:\tlearn: 15213830.8844557\ttotal: 9.88s\tremaining: 2.56s\n", + "794:\tlearn: 15212644.4009126\ttotal: 9.89s\tremaining: 2.55s\n", + "795:\tlearn: 15212570.5272286\ttotal: 9.9s\tremaining: 2.54s\n", + "796:\tlearn: 15172874.1552397\ttotal: 9.92s\tremaining: 2.52s\n", + "797:\tlearn: 15164671.3501787\ttotal: 9.93s\tremaining: 2.51s\n", + "798:\tlearn: 15162711.8871221\ttotal: 9.94s\tremaining: 2.5s\n", + "799:\tlearn: 15162618.0050229\ttotal: 9.95s\tremaining: 2.49s\n", + "800:\tlearn: 15161186.8924011\ttotal: 9.96s\tremaining: 2.48s\n", + "801:\tlearn: 15160994.4738412\ttotal: 9.98s\tremaining: 2.46s\n", + "802:\tlearn: 15159385.3831268\ttotal: 9.99s\tremaining: 2.45s\n", + "803:\tlearn: 15159166.1576231\ttotal: 10s\tremaining: 2.44s\n", + "804:\tlearn: 15156764.1801770\ttotal: 10s\tremaining: 2.43s\n", + "805:\tlearn: 15146691.8394282\ttotal: 10s\tremaining: 2.41s\n", + "806:\tlearn: 15146533.6706853\ttotal: 10s\tremaining: 2.4s\n", + "807:\tlearn: 15146408.7773292\ttotal: 10.1s\tremaining: 2.39s\n", + "808:\tlearn: 15142359.7678728\ttotal: 10.1s\tremaining: 2.38s\n", + "809:\tlearn: 15142322.1248825\ttotal: 10.1s\tremaining: 2.37s\n", + "810:\tlearn: 15132770.1153732\ttotal: 10.1s\tremaining: 2.35s\n", + "811:\tlearn: 15101480.9924963\ttotal: 10.1s\tremaining: 2.34s\n", + "812:\tlearn: 15101445.1875248\ttotal: 10.1s\tremaining: 2.33s\n", + "813:\tlearn: 15075376.4419388\ttotal: 10.1s\tremaining: 2.32s\n", + "814:\tlearn: 15073160.0820287\ttotal: 10.2s\tremaining: 2.3s\n", + "815:\tlearn: 15072725.5140996\ttotal: 10.2s\tremaining: 2.29s\n", + "816:\tlearn: 15072585.7018342\ttotal: 10.2s\tremaining: 2.28s\n", + "817:\tlearn: 15071522.0001919\ttotal: 10.2s\tremaining: 2.27s\n", + "818:\tlearn: 15071382.2097110\ttotal: 10.2s\tremaining: 2.25s\n", + "819:\tlearn: 15071301.2886091\ttotal: 10.2s\tremaining: 2.24s\n", + "820:\tlearn: 15071025.2992144\ttotal: 10.2s\tremaining: 2.23s\n", + "821:\tlearn: 15069498.2268762\ttotal: 10.2s\tremaining: 2.22s\n", + "822:\tlearn: 15061575.7065075\ttotal: 10.2s\tremaining: 2.2s\n", + "823:\tlearn: 15061416.4068476\ttotal: 10.3s\tremaining: 2.19s\n", + "824:\tlearn: 15060945.6687130\ttotal: 10.3s\tremaining: 2.18s\n", + "825:\tlearn: 15051099.8538783\ttotal: 10.3s\tremaining: 2.17s\n", + "826:\tlearn: 15050450.9663299\ttotal: 10.3s\tremaining: 2.15s\n", + "827:\tlearn: 15049722.9751983\ttotal: 10.3s\tremaining: 2.14s\n", + "828:\tlearn: 15049467.7452535\ttotal: 10.3s\tremaining: 2.13s\n", + "829:\tlearn: 15049412.7697933\ttotal: 10.3s\tremaining: 2.12s\n", + "830:\tlearn: 15048891.7740041\ttotal: 10.3s\tremaining: 2.1s\n", + "831:\tlearn: 15048043.0994998\ttotal: 10.4s\tremaining: 2.09s\n", + "832:\tlearn: 15046697.3368860\ttotal: 10.4s\tremaining: 2.08s\n", + "833:\tlearn: 15038272.8803419\ttotal: 10.4s\tremaining: 2.06s\n", + "834:\tlearn: 15034639.9951102\ttotal: 10.4s\tremaining: 2.05s\n", + "835:\tlearn: 15030153.8614245\ttotal: 10.4s\tremaining: 2.04s\n", + "836:\tlearn: 15027964.0190757\ttotal: 10.4s\tremaining: 2.03s\n", + "837:\tlearn: 15023890.1409211\ttotal: 10.4s\tremaining: 2.02s\n", + "838:\tlearn: 15022954.0613643\ttotal: 10.4s\tremaining: 2s\n", + "839:\tlearn: 15022653.7321874\ttotal: 10.5s\tremaining: 1.99s\n", + "840:\tlearn: 15021763.5899870\ttotal: 10.5s\tremaining: 1.98s\n", + "841:\tlearn: 15021552.9666208\ttotal: 10.5s\tremaining: 1.97s\n", + "842:\tlearn: 15017213.3112838\ttotal: 10.5s\tremaining: 1.95s\n", + "843:\tlearn: 15006868.9919636\ttotal: 10.5s\tremaining: 1.94s\n", + "844:\tlearn: 15006047.4873296\ttotal: 10.5s\tremaining: 1.93s\n", + "845:\tlearn: 15003167.3995596\ttotal: 10.5s\tremaining: 1.92s\n", + "846:\tlearn: 15001516.1719277\ttotal: 10.5s\tremaining: 1.9s\n", + "847:\tlearn: 15000023.6971343\ttotal: 10.6s\tremaining: 1.89s\n", + "848:\tlearn: 14996097.9901016\ttotal: 10.6s\tremaining: 1.88s\n", + "849:\tlearn: 14995809.9617414\ttotal: 10.6s\tremaining: 1.87s\n", + "850:\tlearn: 14991694.0680204\ttotal: 10.6s\tremaining: 1.85s\n", + "851:\tlearn: 14990806.5441048\ttotal: 10.6s\tremaining: 1.84s\n", + "852:\tlearn: 14990539.4146062\ttotal: 10.6s\tremaining: 1.83s\n", + "853:\tlearn: 14990428.7975864\ttotal: 10.6s\tremaining: 1.82s\n", + "854:\tlearn: 14989061.7567162\ttotal: 10.6s\tremaining: 1.8s\n", + "855:\tlearn: 14983131.0103419\ttotal: 10.6s\tremaining: 1.79s\n", + "856:\tlearn: 14982655.3316759\ttotal: 10.7s\tremaining: 1.78s\n", + "857:\tlearn: 14977099.0608273\ttotal: 10.7s\tremaining: 1.77s\n", + "858:\tlearn: 14976713.9058693\ttotal: 10.7s\tremaining: 1.75s\n", + "859:\tlearn: 14976613.3364184\ttotal: 10.7s\tremaining: 1.74s\n", + "860:\tlearn: 14964115.6999829\ttotal: 10.7s\tremaining: 1.73s\n", + "861:\tlearn: 14961152.7626425\ttotal: 10.7s\tremaining: 1.72s\n", + "862:\tlearn: 14960316.8698796\ttotal: 10.7s\tremaining: 1.7s\n", + "863:\tlearn: 14960206.5805103\ttotal: 10.8s\tremaining: 1.69s\n", + "864:\tlearn: 14948350.2065232\ttotal: 10.8s\tremaining: 1.68s\n", + "865:\tlearn: 14948237.8225238\ttotal: 10.8s\tremaining: 1.67s\n", + "866:\tlearn: 14948145.1280412\ttotal: 10.8s\tremaining: 1.66s\n", + "867:\tlearn: 14947479.9936319\ttotal: 10.8s\tremaining: 1.64s\n", + "868:\tlearn: 14946706.9144290\ttotal: 10.8s\tremaining: 1.63s\n", + "869:\tlearn: 14946008.4529886\ttotal: 10.8s\tremaining: 1.62s\n", + "870:\tlearn: 14938382.9733130\ttotal: 10.8s\tremaining: 1.6s\n", + "871:\tlearn: 14935923.0018589\ttotal: 10.9s\tremaining: 1.59s\n", + "872:\tlearn: 14935763.9386719\ttotal: 10.9s\tremaining: 1.58s\n", + "873:\tlearn: 14935390.9032799\ttotal: 10.9s\tremaining: 1.57s\n", + "874:\tlearn: 14924136.4999000\ttotal: 10.9s\tremaining: 1.55s\n", + "875:\tlearn: 14923231.4975181\ttotal: 10.9s\tremaining: 1.54s\n", + "876:\tlearn: 14920764.7489123\ttotal: 10.9s\tremaining: 1.53s\n", + "877:\tlearn: 14920619.6869935\ttotal: 10.9s\tremaining: 1.52s\n", + "878:\tlearn: 14920259.9887151\ttotal: 10.9s\tremaining: 1.51s\n", + "879:\tlearn: 14918671.6063618\ttotal: 11s\tremaining: 1.49s\n", + "880:\tlearn: 14909484.3446534\ttotal: 11s\tremaining: 1.48s\n", + "881:\tlearn: 14909331.3722806\ttotal: 11s\tremaining: 1.47s\n", + "882:\tlearn: 14909000.4744294\ttotal: 11s\tremaining: 1.46s\n", + "883:\tlearn: 14907810.4215534\ttotal: 11s\tremaining: 1.44s\n", + "884:\tlearn: 14907739.9399244\ttotal: 11s\tremaining: 1.43s\n", + "885:\tlearn: 14907643.8234156\ttotal: 11s\tremaining: 1.42s\n", + "886:\tlearn: 14902234.6414918\ttotal: 11s\tremaining: 1.41s\n", + "887:\tlearn: 14899639.1808572\ttotal: 11.1s\tremaining: 1.39s\n", + "888:\tlearn: 14898572.6302420\ttotal: 11.1s\tremaining: 1.38s\n", + "889:\tlearn: 14898481.9011232\ttotal: 11.1s\tremaining: 1.37s\n", + "890:\tlearn: 14898396.0569341\ttotal: 11.1s\tremaining: 1.36s\n", + "891:\tlearn: 14897825.8306216\ttotal: 11.1s\tremaining: 1.34s\n", + "892:\tlearn: 14897667.5383103\ttotal: 11.1s\tremaining: 1.33s\n", + "893:\tlearn: 14894977.2967186\ttotal: 11.1s\tremaining: 1.32s\n", + "894:\tlearn: 14894451.1015405\ttotal: 11.1s\tremaining: 1.31s\n", + "895:\tlearn: 14894302.6357933\ttotal: 11.1s\tremaining: 1.29s\n", + "896:\tlearn: 14892042.1460828\ttotal: 11.2s\tremaining: 1.28s\n", + "897:\tlearn: 14891913.2181914\ttotal: 11.2s\tremaining: 1.27s\n", + "898:\tlearn: 14891548.8875063\ttotal: 11.2s\tremaining: 1.26s\n", + "899:\tlearn: 14891535.2806629\ttotal: 11.2s\tremaining: 1.24s\n", + "900:\tlearn: 14885531.4554658\ttotal: 11.2s\tremaining: 1.23s\n", + "901:\tlearn: 14885300.5490787\ttotal: 11.2s\tremaining: 1.22s\n", + "902:\tlearn: 14882237.8064495\ttotal: 11.2s\tremaining: 1.21s\n", + "903:\tlearn: 14871534.0201501\ttotal: 11.3s\tremaining: 1.2s\n", + "904:\tlearn: 14870231.5883229\ttotal: 11.3s\tremaining: 1.18s\n", + "905:\tlearn: 14870176.1224648\ttotal: 11.3s\tremaining: 1.17s\n", + "906:\tlearn: 14869680.0980501\ttotal: 11.3s\tremaining: 1.16s\n", + "907:\tlearn: 14869457.0180442\ttotal: 11.3s\tremaining: 1.15s\n", + "908:\tlearn: 14868933.3692698\ttotal: 11.3s\tremaining: 1.13s\n", + "909:\tlearn: 14863961.6816683\ttotal: 11.3s\tremaining: 1.12s\n", + "910:\tlearn: 14863914.5911026\ttotal: 11.4s\tremaining: 1.11s\n", + "911:\tlearn: 14855476.3946477\ttotal: 11.4s\tremaining: 1.1s\n", + "912:\tlearn: 14855166.2145584\ttotal: 11.4s\tremaining: 1.08s\n", + "913:\tlearn: 14850934.1124148\ttotal: 11.4s\tremaining: 1.07s\n", + "914:\tlearn: 14846483.1667184\ttotal: 11.4s\tremaining: 1.06s\n", + "915:\tlearn: 14837736.3329095\ttotal: 11.4s\tremaining: 1.05s\n", + "916:\tlearn: 14837595.6179847\ttotal: 11.4s\tremaining: 1.03s\n", + "917:\tlearn: 14830817.0386636\ttotal: 11.4s\tremaining: 1.02s\n", + "918:\tlearn: 14830573.0969752\ttotal: 11.5s\tremaining: 1.01s\n", + "919:\tlearn: 14830470.0025192\ttotal: 11.5s\tremaining: 997ms\n", + "920:\tlearn: 14829815.6285131\ttotal: 11.5s\tremaining: 985ms\n", + "921:\tlearn: 14825434.2189552\ttotal: 11.5s\tremaining: 973ms\n", + "922:\tlearn: 14819445.0181126\ttotal: 11.5s\tremaining: 960ms\n", + "923:\tlearn: 14819398.6279561\ttotal: 11.5s\tremaining: 948ms\n", + "924:\tlearn: 14818813.5923928\ttotal: 11.5s\tremaining: 935ms\n", + "925:\tlearn: 14815956.9177135\ttotal: 11.5s\tremaining: 923ms\n", + "926:\tlearn: 14815788.2679741\ttotal: 11.6s\tremaining: 910ms\n", + "927:\tlearn: 14815448.9260298\ttotal: 11.6s\tremaining: 897ms\n", + "928:\tlearn: 14815074.5372959\ttotal: 11.6s\tremaining: 885ms\n", + "929:\tlearn: 14814229.2585638\ttotal: 11.6s\tremaining: 872ms\n", + "930:\tlearn: 14804039.0241152\ttotal: 11.6s\tremaining: 860ms\n", + "931:\tlearn: 14803929.0738285\ttotal: 11.6s\tremaining: 847ms\n", + "932:\tlearn: 14803822.0317935\ttotal: 11.6s\tremaining: 835ms\n", + "933:\tlearn: 14802847.7620639\ttotal: 11.6s\tremaining: 822ms\n", + "934:\tlearn: 14802644.0143811\ttotal: 11.7s\tremaining: 810ms\n", + "935:\tlearn: 14801390.4240818\ttotal: 11.7s\tremaining: 798ms\n", + "936:\tlearn: 14801273.4225706\ttotal: 11.7s\tremaining: 785ms\n", + "937:\tlearn: 14801021.8599058\ttotal: 11.7s\tremaining: 772ms\n", + "938:\tlearn: 14800715.3029627\ttotal: 11.7s\tremaining: 760ms\n", + "939:\tlearn: 14799551.6406369\ttotal: 11.7s\tremaining: 748ms\n", + "940:\tlearn: 14795598.5613345\ttotal: 11.7s\tremaining: 735ms\n", + "941:\tlearn: 14794465.0278834\ttotal: 11.7s\tremaining: 723ms\n", + "942:\tlearn: 14794259.7563387\ttotal: 11.8s\tremaining: 710ms\n", + "943:\tlearn: 14794156.6214413\ttotal: 11.8s\tremaining: 698ms\n", + "944:\tlearn: 14792982.4344262\ttotal: 11.8s\tremaining: 685ms\n", + "945:\tlearn: 14792468.8012658\ttotal: 11.8s\tremaining: 673ms\n", + "946:\tlearn: 14792139.9923168\ttotal: 11.8s\tremaining: 661ms\n", + "947:\tlearn: 14786295.7938911\ttotal: 11.8s\tremaining: 648ms\n", + "948:\tlearn: 14783272.8762359\ttotal: 11.8s\tremaining: 636ms\n", + "949:\tlearn: 14782689.5513664\ttotal: 11.8s\tremaining: 623ms\n", + "950:\tlearn: 14782664.1266181\ttotal: 11.9s\tremaining: 611ms\n", + "951:\tlearn: 14774339.7873426\ttotal: 11.9s\tremaining: 598ms\n", + "952:\tlearn: 14769038.5403572\ttotal: 11.9s\tremaining: 586ms\n", + "953:\tlearn: 14761312.5028488\ttotal: 11.9s\tremaining: 573ms\n", + "954:\tlearn: 14760414.3496721\ttotal: 11.9s\tremaining: 561ms\n", + "955:\tlearn: 14752950.6860631\ttotal: 11.9s\tremaining: 548ms\n", + "956:\tlearn: 14752691.2767919\ttotal: 11.9s\tremaining: 536ms\n", + "957:\tlearn: 14752196.9268404\ttotal: 11.9s\tremaining: 524ms\n", + "958:\tlearn: 14750149.4071752\ttotal: 12s\tremaining: 511ms\n", + "959:\tlearn: 14749691.8632556\ttotal: 12s\tremaining: 499ms\n", + "960:\tlearn: 14749496.6164671\ttotal: 12s\tremaining: 486ms\n", + "961:\tlearn: 14749338.8909588\ttotal: 12s\tremaining: 474ms\n", + "962:\tlearn: 14749291.1143099\ttotal: 12s\tremaining: 461ms\n", + "963:\tlearn: 14739582.8672605\ttotal: 12s\tremaining: 449ms\n", + "964:\tlearn: 14739542.4442574\ttotal: 12s\tremaining: 436ms\n", + "965:\tlearn: 14739306.4391584\ttotal: 12s\tremaining: 424ms\n", + "966:\tlearn: 14708630.7344942\ttotal: 12.1s\tremaining: 411ms\n", + "967:\tlearn: 14680984.5320050\ttotal: 12.1s\tremaining: 399ms\n", + "968:\tlearn: 14677319.7390898\ttotal: 12.1s\tremaining: 386ms\n", + "969:\tlearn: 14677109.6650390\ttotal: 12.1s\tremaining: 374ms\n", + "970:\tlearn: 14676355.6242387\ttotal: 12.1s\tremaining: 361ms\n", + "971:\tlearn: 14667109.8761175\ttotal: 12.1s\tremaining: 349ms\n", + "972:\tlearn: 14643934.7657464\ttotal: 12.1s\tremaining: 337ms\n", + "973:\tlearn: 14642633.8347823\ttotal: 12.1s\tremaining: 324ms\n", + "974:\tlearn: 14642167.8071045\ttotal: 12.2s\tremaining: 312ms\n", + "975:\tlearn: 14641845.3365852\ttotal: 12.2s\tremaining: 299ms\n", + "976:\tlearn: 14640306.5144587\ttotal: 12.2s\tremaining: 287ms\n", + "977:\tlearn: 14640184.9147582\ttotal: 12.2s\tremaining: 274ms\n", + "978:\tlearn: 14640008.8483425\ttotal: 12.2s\tremaining: 262ms\n", + "979:\tlearn: 14636344.5519428\ttotal: 12.2s\tremaining: 249ms\n", + "980:\tlearn: 14633576.2663352\ttotal: 12.2s\tremaining: 237ms\n", + "981:\tlearn: 14624796.2134142\ttotal: 12.2s\tremaining: 225ms\n", + "982:\tlearn: 14624250.6774943\ttotal: 12.3s\tremaining: 212ms\n", + "983:\tlearn: 14615917.0650799\ttotal: 12.3s\tremaining: 200ms\n", + "984:\tlearn: 14615782.8388140\ttotal: 12.3s\tremaining: 187ms\n", + "985:\tlearn: 14612724.2754075\ttotal: 12.3s\tremaining: 175ms\n", + "986:\tlearn: 14609973.2837772\ttotal: 12.3s\tremaining: 162ms\n", + "987:\tlearn: 14605203.8050795\ttotal: 12.3s\tremaining: 150ms\n", + "988:\tlearn: 14605011.6874159\ttotal: 12.3s\tremaining: 137ms\n", + "989:\tlearn: 14594552.7887146\ttotal: 12.4s\tremaining: 125ms\n", + "990:\tlearn: 14591881.9316489\ttotal: 12.4s\tremaining: 112ms\n", + "991:\tlearn: 14581962.3358039\ttotal: 12.4s\tremaining: 99.8ms\n", + "992:\tlearn: 14581829.5331587\ttotal: 12.4s\tremaining: 87.3ms\n", + "993:\tlearn: 14581669.7347033\ttotal: 12.4s\tremaining: 74.9ms\n", + "994:\tlearn: 14577373.3119596\ttotal: 12.4s\tremaining: 62.4ms\n", + "995:\tlearn: 14577196.6327960\ttotal: 12.4s\tremaining: 49.9ms\n", + "996:\tlearn: 14577122.1536884\ttotal: 12.4s\tremaining: 37.4ms\n", + "997:\tlearn: 14574702.1336653\ttotal: 12.4s\tremaining: 25ms\n", + "998:\tlearn: 14574660.6060510\ttotal: 12.5s\tremaining: 12.5ms\n", + "999:\tlearn: 14574625.9856659\ttotal: 12.5s\tremaining: 0us\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024/10/10 13:36:41 WARNING mlflow.utils.autologging_utils: MLflow autologging encountered a warning: \"/home/andrey/work/institute/MLE/assets/mlflow/.venv_ml2/lib/python3.10/site-packages/mlflow/types/utils.py:407: UserWarning: Hint: Inferred schema contains integer column(s). Integer columns in Python cannot represent missing values. If your input data contains missing values at inference time, it will be encoded as floats and will cause a schema enforcement error. The best way to avoid this problem is to infer the model schema based on a realistic data sample (training dataset) that includes missing values. Alternatively, you can declare integer columns as doubles (float64) whenever these columns may have missing values. See `Handling Integers With Missing Values `_ for more details.\"\n", + "2024/10/10 13:36:42 INFO mlflow.tracking._tracking_service.client: 🏃 View run auto at: http://127.0.0.1:5000/#/experiments/1/runs/9dd5dd7674d44c44be662e3c9bdbb5a4.\n", + "2024/10/10 13:36:42 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" ] } ], @@ -1684,10 +3683,9 @@ " 'area', 'kitchen_area']),\n", " ('cat',\n", " OrdinalEncoder(handle_unknown='use_encoded_value',\n", - " unknown_value=999),\n", + " unknown_value=99999999),\n", " ['region', 'building_type',\n", - " 'object_type',\n", - " 'floor_level'])])),\n", + " 'object_type'])])),\n", " ('model', RandomForestRegressor(max_depth=6, n_estimators=10))])In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org." + " unknown_value=99999999),\n", + " ['region', 'building_type', 'object_type'])]) 
['geo_lat', 'geo_lon', 'level', 'levels', 'rooms', 'area', 'kitchen_area']
StandardScaler()
['region', 'building_type', 'object_type']
OrdinalEncoder(handle_unknown='use_encoded_value', unknown_value=99999999)
RandomForestRegressor(max_depth=6, n_estimators=10)
" ], "text/plain": [ "Pipeline(steps=[('preprocessor',\n", @@ -1716,10 +3712,9 @@ " 'area', 'kitchen_area']),\n", " ('cat',\n", " OrdinalEncoder(handle_unknown='use_encoded_value',\n", - " unknown_value=999),\n", + " unknown_value=99999999),\n", " ['region', 'building_type',\n", - " 'object_type',\n", - " 'floor_level'])])),\n", + " 'object_type'])])),\n", " ('model', RandomForestRegressor(max_depth=6, n_estimators=10))])" ] }, @@ -1743,9 +3738,9 @@ { "data": { "text/plain": [ - "{'mae': np.float64(1536543.887713661),\n", - " 'mape': np.float64(0.42528854535519156),\n", - " 'mse': np.float64(210549541556055.7)}" + "{'mae': 1714377.2944153566,\n", + " 'mape': 2.2781106914490788e+18,\n", + " 'mse': 406946525225456.6}" ] }, "execution_count": 25, @@ -1772,8 +3767,8 @@ "name": "stderr", "output_type": "stream", "text": [ - "2024/10/03 19:02:51 INFO mlflow.tracking._tracking_service.client: 🏃 View run smaller_model at: http://127.0.0.1:5000/#/experiments/1/runs/20f66bd4c3754a04b5e47ecc0f577e76.\n", - "2024/10/03 19:02:51 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" + "2024/10/10 13:36:50 INFO mlflow.tracking._tracking_service.client: 🏃 View run smaller_model at: http://127.0.0.1:5000/#/experiments/1/runs/638e0e7ad9f94ceaa01cacd1916771e2.\n", + "2024/10/10 13:36:50 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" ] } ], @@ -1782,10 +3777,7 @@ "\n", "\n", "RUN_NAME = 'smaller_model'\n", - "# Когда создаем новый эксперимент, то: \n", - "#experiment_id = mlflow.create_experiment(EXPERIMENT_NAME)\n", "\n", - "# Впоследствии. чтобы добавлять запуски в этот же эксепримент мы должны получить его id:\n", "experiment_id = mlflow.get_experiment_by_name(EXPERIMENT_NAME).experiment_id\n", "\n", "with mlflow.start_run(run_name=RUN_NAME, experiment_id=experiment_id) as run:\n", @@ -1814,8 +3806,8 @@ "name": "stderr", "output_type": "stream", "text": [ - "2024/10/03 19:02:51 INFO mlflow.tracking._tracking_service.client: 🏃 View run no_model at: http://127.0.0.1:5000/#/experiments/1/runs/6f6fe970eb74485d866e918b733f8f61.\n", - "2024/10/03 19:02:51 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" + "2024/10/10 13:36:50 INFO mlflow.tracking._tracking_service.client: 🏃 View run no_model at: http://127.0.0.1:5000/#/experiments/1/runs/3b80343fbdc2434ba18b42e049677a28.\n", + "2024/10/10 13:36:50 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" ] } ], @@ -1850,14 +3842,14 @@ "output_type": "stream", "text": [ "Registered model 'estate_model_rf' already exists. Creating a new version of this model...\n", - "2024/10/03 19:03:14 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: estate_model_rf, version 1\n", + "2024/10/10 13:37:19 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: estate_model_rf, version 1\n", "Created version '1' of model 'estate_model_rf'.\n" ] }, { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 29, @@ -1866,7 +3858,7 @@ } ], "source": [ - "run_id = '' # Указываем run id\n", + "run_id = '06fa7ec1f1b74aedb3509c88dc4ee1c0' # Указываем run id\n", "mlflow.register_model(f\"runs:/{run_id}/models\", REGISTRY_MODEL_NAME)" ] }, @@ -1880,10 +3872,10 @@ "output_type": "stream", "text": [ "Registered model 'estate_model_rf' already exists. Creating a new version of this model...\n", - "2024/10/03 19:03:14 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: estate_model_rf, version 2\n", + "2024/10/10 13:37:21 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: estate_model_rf, version 2\n", "Created version '2' of model 'estate_model_rf'.\n", - "2024/10/03 19:03:14 INFO mlflow.tracking._tracking_service.client: 🏃 View run register_at_run at: http://127.0.0.1:5000/#/experiments/1/runs/ed64a91759ed43c99329810d066ea95a.\n", - "2024/10/03 19:03:14 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" + "2024/10/10 13:37:21 INFO mlflow.tracking._tracking_service.client: 🏃 View run register_at_run at: http://127.0.0.1:5000/#/experiments/1/runs/a34ac1da523a4687846c26f32d44c051.\n", + "2024/10/10 13:37:21 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" ] } ], @@ -1918,7 +3910,7 @@ { "data": { "text/plain": [ - "], name='estate_model_rf', tags={}>" + "], name='estate_model_rf', tags={}>" ] }, "execution_count": 31, @@ -1953,7 +3945,7 @@ { "data": { "text/plain": [ - "array([3438055.97819847])" + "array([3083461.0078044])" ] }, "execution_count": 33, @@ -1973,7 +3965,7 @@ { "data": { "text/plain": [ - "np.int64(3062900)" + "2400000" ] }, "execution_count": 34, @@ -1984,6 +3976,4980 @@ "source": [ "y_test.iloc[0]" ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Feature engineering" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Sklearn" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "metadata": {}, + "outputs": [], + "source": [ + "from sklearn.preprocessing import QuantileTransformer, SplineTransformer, PolynomialFeatures, MinMaxScaler" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "metadata": {}, + "outputs": [], + "source": [ + "X_train_sklearn = X_train.copy()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### PolynomialFeatures\n", + "Создает полином степени `degree` из указанных признаков\n" + ] + }, + { + "cell_type": "code", + "execution_count": 37, + "metadata": {}, + "outputs": [], + "source": [ + "pf = PolynomialFeatures(degree=2)" + ] + }, + { + "cell_type": "code", + "execution_count": 39, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
geo_latgeo_lonregionbuilding_typelevellevelsroomsareakitchen_areaobject_type
87948756.32768643.92806228711810256.00008.5000001
25420854.85931082.9756249654159250.18758.5000001
226056456.07216354.2312702722349139.00009.00000011
335925146.70432738.2736362843155248.00009.0000001
386189560.93348376.59309424842715274.000010.5000001
.................................
338940956.77133960.61251861712918251.31259.5703121
284882755.14861361.3937805282335243.00006.0000001
267200945.09812938.97121828430817139.00009.00000011
422705050.58396936.58186759523316298.500027.7968751
433774759.84457030.40907726611612372.00009.0000001
\n", + "

410775 rows × 10 columns

\n", + "
" + ], + "text/plain": [ + " geo_lat geo_lon region building_type level levels rooms \\\n", + "879487 56.327686 43.928062 2871 1 8 10 2 \n", + "254208 54.859310 82.975624 9654 1 5 9 2 \n", + "2260564 56.072163 54.231270 2722 3 4 9 1 \n", + "3359251 46.704327 38.273636 2843 1 5 5 2 \n", + "3861895 60.933483 76.593094 2484 2 7 15 2 \n", + "... ... ... ... ... ... ... ... \n", + "3389409 56.771339 60.612518 6171 2 9 18 2 \n", + "2848827 55.148613 61.393780 5282 3 3 5 2 \n", + "2672009 45.098129 38.971218 2843 0 8 17 1 \n", + "4227050 50.583969 36.581867 5952 3 3 16 2 \n", + "4337747 59.844570 30.409077 2661 1 6 12 3 \n", + "\n", + " area kitchen_area object_type \n", + "879487 56.0000 8.500000 1 \n", + "254208 50.1875 8.500000 1 \n", + "2260564 39.0000 9.000000 11 \n", + "3359251 48.0000 9.000000 1 \n", + "3861895 74.0000 10.500000 1 \n", + "... ... ... ... \n", + "3389409 51.3125 9.570312 1 \n", + "2848827 43.0000 6.000000 1 \n", + "2672009 39.0000 9.000000 11 \n", + "4227050 98.5000 27.796875 1 \n", + "4337747 72.0000 9.000000 1 \n", + "\n", + "[410775 rows x 10 columns]" + ] + }, + "execution_count": 39, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "X_train_sklearn" + ] + }, + { + "cell_type": "code", + "execution_count": 42, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/andrey/work/institute/MLE/assets/mlflow/.venv_ml2/lib/python3.10/site-packages/sklearn/preprocessing/_polynomial.py:555: RuntimeWarning: overflow encountered in multiply\n", + " np.multiply(\n" + ] + }, + { + "data": { + "text/plain": [ + "array([[1.000e+00, 5.600e+01, 8.500e+00, 3.136e+03, 4.760e+02, 7.225e+01],\n", + " [1.000e+00, 5.019e+01, 8.500e+00, 2.518e+03, 4.265e+02, 7.225e+01],\n", + " [1.000e+00, 3.900e+01, 9.000e+00, 1.521e+03, 3.510e+02, 8.100e+01],\n", + " ...,\n", + " [1.000e+00, 3.900e+01, 9.000e+00, 1.521e+03, 3.510e+02, 8.100e+01],\n", + " [1.000e+00, 9.850e+01, 2.780e+01, 9.704e+03, 2.738e+03, 7.725e+02],\n", + " [1.000e+00, 7.200e+01, 9.000e+00, 5.184e+03, 6.480e+02, 8.100e+01]],\n", + " dtype=float16)" + ] + }, + "execution_count": 42, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "pf.fit_transform(X_train_sklearn[['area','kitchen_area']])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### SplineTransformer\n", + "Cоздаёт новую матрицу признаков, состоящую из сплайнов порядка degree. Количество сгенерированных сплайнов равно `n_splines=n_knots + degree - 1` для каждого признака, где\n", + "\n", + "`n_knots` определяет количество узлов (точек, в которых сопрягаются сплайны) для каждого признака. \n", + "\n", + "`degree` определяет порядок полинома, используемого для построения сплайнов. " + ] + }, + { + "cell_type": "code", + "execution_count": 43, + "metadata": {}, + "outputs": [], + "source": [ + "sp = SplineTransformer(n_knots=3, degree=3)" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[1.558e-01, 6.660e-01, 1.780e-01, 1.848e-06, 0.000e+00],\n", + " [1.569e-01, 6.665e-01, 1.768e-01, 1.311e-06, 0.000e+00],\n", + " [1.591e-01, 6.665e-01, 1.744e-01, 5.960e-07, 0.000e+00],\n", + " ...,\n", + " [1.591e-01, 6.665e-01, 1.744e-01, 5.960e-07, 0.000e+00],\n", + " [1.478e-01, 6.650e-01, 1.870e-01, 1.007e-05, 0.000e+00],\n", + " [1.527e-01, 6.660e-01, 1.814e-01, 3.874e-06, 0.000e+00]],\n", + " dtype=float16)" + ] + }, + "execution_count": 45, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "sp.fit_transform(X_train_sklearn[['area']])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "#### QuantileTransformer\n", + "Этот метод преобразует признаки, чтобы они распределялись равномерно или нормально — так данные меньше подвергаются влиянию выбросов. Преобразование применяется к каждому признаку независимо. Идея метода такова: оценить функцию распределения признака, чтобы преобразовать исходные значения в равномерное или нормальное распределение. \n", + "\n", + "`output_distribution='uniform'` или\n", + "`output_distribution='normal'` соответственно\n", + "\n", + "\n", + "Пример использования: если у вас есть данные о доходах с широким диапазоном значений, квантильное преобразование сделает их более сопоставимыми и устойчивыми к выбросам." + ] + }, + { + "cell_type": "code", + "execution_count": 47, + "metadata": {}, + "outputs": [], + "source": [ + "qt = QuantileTransformer()" + ] + }, + { + "cell_type": "code", + "execution_count": 48, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[0.6226],\n", + " [0.544 ],\n", + " [0.2708],\n", + " ...,\n", + " [0.2708],\n", + " [0.958 ],\n", + " [0.8433]], dtype=float16)" + ] + }, + "execution_count": 48, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "qt.fit_transform(X_train_sklearn[['area']])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Объединяем в ColumnTransformer и создаем Pipeline " + ] + }, + { + "cell_type": "code", + "execution_count": 49, + "metadata": {}, + "outputs": [], + "source": [ + "pf = PolynomialFeatures(degree=2)\n", + "qt = QuantileTransformer()\n", + "sp = SplineTransformer(n_knots=3, degree=3)" + ] + }, + { + "cell_type": "code", + "execution_count": 50, + "metadata": {}, + "outputs": [], + "source": [ + "# Значения преобразованных признаков нужно отскейлить, поэтому создаем pipeline из двух шагов - преобразование и скейлинг\n", + "pf_pipeline = Pipeline(steps=[\n", + " ('poly', pf),\n", + " ('scale', StandardScaler())\n", + "])" + ] + }, + { + "cell_type": "code", + "execution_count": 51, + "metadata": {}, + "outputs": [], + "source": [ + "preprocessor_sklearn = ColumnTransformer(\n", + " transformers=[\n", + " ('num', s_scaler, num_features), # преобразования для числовых признаков\n", + " ('cat', l_encoder, cat_features), # преобразования для категориальных признаков\n", + " ('quantile', qt,num_features),\n", + " ('poly', pf_pipeline, ['area', 'kitchen_area']), # В преобразования добавляем созданный ранее pipeline\n", + " ('spline', sp, ['area'])\n", + " ],\n", + " remainder='drop',\n", + " ) # Удаляем столбцы, которые не затронуты преобразования" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Посмотрим что из себя теперь представляет датафрейм" + ] + }, + { + "cell_type": "code", + "execution_count": 56, + "metadata": {}, + "outputs": [], + "source": [ + "## не влезаем в float64 в полиномальном преобразовании. Использовать его нужно с умом!\n", + "X_train_sklearn[['area', 'kitchen_area']] = X_train_sklearn[['area', 'kitchen_area']].astype('float128')\n", + "X_train_sklearn[['area', 'kitchen_area']] = X_train_sklearn[['area', 'kitchen_area']].astype('float128')" + ] + }, + { + "cell_type": "code", + "execution_count": 58, + "metadata": {}, + "outputs": [], + "source": [ + "X_train_sklearn_raw = preprocessor_sklearn.fit_transform(X_train_sklearn)\n", + "X_train_sklearn = pd.DataFrame(X_train_sklearn_raw, columns=preprocessor_sklearn.get_feature_names_out())" + ] + }, + { + "cell_type": "code", + "execution_count": 77, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
num__geo_latnum__geo_lonnum__levelnum__levelsnum__roomsnum__areanum__kitchen_areacat__regioncat__building_typecat__object_typequantile__geo_latquantile__geo_lonquantile__levelquantile__levelsquantile__roomsquantile__areaquantile__kitchen_areapoly__1poly__areapoly__kitchen_areapoly__area^2poly__area kitchen_areapoly__kitchen_area^2spline__area_sp_0spline__area_sp_1spline__area_sp_2spline__area_sp_3spline__area_sp_4
00.495902-0.4497420.359235-0.2147890.2534130.063735-0.18628520.01.00.00.7662570.5110280.7172170.5365370.6006010.6236240.3748750.00.063735-0.186285-0.010002-0.132188-0.0027920.1558060.6661790.1780130.0000020.0
10.1778061.433673-0.246529-0.3677180.253413-0.114293-0.18628570.01.00.00.2971420.8679990.5220220.3868870.6006010.5415420.3748750.0-0.114293-0.186285-0.017375-0.169370-0.0027920.1569210.6662750.1768030.0000010.0
.......................................................................................
410773-0.748366-0.804077-0.6503710.7027880.2534131.3654411.50183352.03.00.00.1931430.1147530.3098100.7417420.6006010.9613670.9845350.01.3654411.5018330.0684381.5701630.0086160.1478200.6651590.1870110.0000100.0
4107741.257769-1.101815-0.0446080.0910701.1759110.553789-0.14254414.01.00.00.9080360.0757250.6046050.6456460.8673670.8418420.4364360.00.553789-0.1425440.014463-0.002742-0.0026490.1527670.6658600.1813700.0000040.0
\n", + "

410775 rows × 28 columns

\n", + "
" + ], + "text/plain": [ + " num__geo_lat num__geo_lon num__level num__levels num__rooms \\\n", + "0 0.495902 -0.449742 0.359235 -0.214789 0.253413 \n", + "1 0.177806 1.433673 -0.246529 -0.367718 0.253413 \n", + "... ... ... ... ... ... \n", + "410773 -0.748366 -0.804077 -0.650371 0.702788 0.253413 \n", + "410774 1.257769 -1.101815 -0.044608 0.091070 1.175911 \n", + "\n", + " num__area num__kitchen_area cat__region cat__building_type \\\n", + "0 0.063735 -0.186285 20.0 1.0 \n", + "1 -0.114293 -0.186285 70.0 1.0 \n", + "... ... ... ... ... \n", + "410773 1.365441 1.501833 52.0 3.0 \n", + "410774 0.553789 -0.142544 14.0 1.0 \n", + "\n", + " cat__object_type quantile__geo_lat quantile__geo_lon \\\n", + "0 0.0 0.766257 0.511028 \n", + "1 0.0 0.297142 0.867999 \n", + "... ... ... ... \n", + "410773 0.0 0.193143 0.114753 \n", + "410774 0.0 0.908036 0.075725 \n", + "\n", + " quantile__level quantile__levels quantile__rooms quantile__area \\\n", + "0 0.717217 0.536537 0.600601 0.623624 \n", + "1 0.522022 0.386887 0.600601 0.541542 \n", + "... ... ... ... ... \n", + "410773 0.309810 0.741742 0.600601 0.961367 \n", + "410774 0.604605 0.645646 0.867367 0.841842 \n", + "\n", + " quantile__kitchen_area poly__1 poly__area poly__kitchen_area \\\n", + "0 0.374875 0.0 0.063735 -0.186285 \n", + "1 0.374875 0.0 -0.114293 -0.186285 \n", + "... ... ... ... ... \n", + "410773 0.984535 0.0 1.365441 1.501833 \n", + "410774 0.436436 0.0 0.553789 -0.142544 \n", + "\n", + " poly__area^2 poly__area kitchen_area poly__kitchen_area^2 \\\n", + "0 -0.010002 -0.132188 -0.002792 \n", + "1 -0.017375 -0.169370 -0.002792 \n", + "... ... ... ... \n", + "410773 0.068438 1.570163 0.008616 \n", + "410774 0.014463 -0.002742 -0.002649 \n", + "\n", + " spline__area_sp_0 spline__area_sp_1 spline__area_sp_2 \\\n", + "0 0.155806 0.666179 0.178013 \n", + "1 0.156921 0.666275 0.176803 \n", + "... ... ... ... \n", + "410773 0.147820 0.665159 0.187011 \n", + "410774 0.152767 0.665860 0.181370 \n", + "\n", + " spline__area_sp_3 spline__area_sp_4 \n", + "0 0.000002 0.0 \n", + "1 0.000001 0.0 \n", + "... ... ... \n", + "410773 0.000010 0.0 \n", + "410774 0.000004 0.0 \n", + "\n", + "[410775 rows x 28 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Удобно использовать для отображения всех строк\\столбцов в DataFrame\n", + "with pd.option_context('display.max_rows', 5, 'display.max_columns', None):\n", + " display (X_train_sklearn)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Создаем пайплайн с препроцессингом и моделью" + ] + }, + { + "cell_type": "code", + "execution_count": 61, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Learning rate set to 0.105957\n", + "0:\tlearn: 22102731.6794979\ttotal: 17.1ms\tremaining: 17.1s\n", + "1:\tlearn: 21993522.6168201\ttotal: 34.1ms\tremaining: 17s\n", + "2:\tlearn: 21906449.8053083\ttotal: 52.6ms\tremaining: 17.5s\n", + "3:\tlearn: 21832729.3406948\ttotal: 69.3ms\tremaining: 17.2s\n", + "4:\tlearn: 21770057.1251014\ttotal: 86.9ms\tremaining: 17.3s\n", + "5:\tlearn: 21717915.6748554\ttotal: 104ms\tremaining: 17.3s\n", + "6:\tlearn: 21673891.4625967\ttotal: 119ms\tremaining: 16.9s\n", + "7:\tlearn: 21635252.2084506\ttotal: 136ms\tremaining: 16.8s\n", + "8:\tlearn: 21599880.1735130\ttotal: 154ms\tremaining: 17s\n", + "9:\tlearn: 21571896.6598855\ttotal: 174ms\tremaining: 17.2s\n", + "10:\tlearn: 21548918.7920905\ttotal: 191ms\tremaining: 17.1s\n", + "11:\tlearn: 21469169.3085785\ttotal: 210ms\tremaining: 17.3s\n", + "12:\tlearn: 20917151.7105915\ttotal: 227ms\tremaining: 17.2s\n", + "13:\tlearn: 20898090.1442249\ttotal: 245ms\tremaining: 17.2s\n", + "14:\tlearn: 20880604.5205543\ttotal: 269ms\tremaining: 17.7s\n", + "15:\tlearn: 20866509.3261171\ttotal: 292ms\tremaining: 18s\n", + "16:\tlearn: 20853897.8112220\ttotal: 316ms\tremaining: 18.3s\n", + "17:\tlearn: 20843816.1230098\ttotal: 338ms\tremaining: 18.5s\n", + "18:\tlearn: 20373459.6724846\ttotal: 357ms\tremaining: 18.4s\n", + "19:\tlearn: 20363758.0877283\ttotal: 377ms\tremaining: 18.5s\n", + "20:\tlearn: 20356328.7643766\ttotal: 395ms\tremaining: 18.4s\n", + "21:\tlearn: 20348176.2337057\ttotal: 418ms\tremaining: 18.6s\n", + "22:\tlearn: 20334645.4973208\ttotal: 437ms\tremaining: 18.6s\n", + "23:\tlearn: 20311618.5886958\ttotal: 453ms\tremaining: 18.4s\n", + "24:\tlearn: 20248723.8214635\ttotal: 474ms\tremaining: 18.5s\n", + "25:\tlearn: 20244454.1815028\ttotal: 495ms\tremaining: 18.5s\n", + "26:\tlearn: 20232541.1732198\ttotal: 512ms\tremaining: 18.5s\n", + "27:\tlearn: 20227699.3837877\ttotal: 532ms\tremaining: 18.5s\n", + "28:\tlearn: 20222366.2734304\ttotal: 550ms\tremaining: 18.4s\n", + "29:\tlearn: 20216976.8323394\ttotal: 569ms\tremaining: 18.4s\n", + "30:\tlearn: 20213554.7196297\ttotal: 588ms\tremaining: 18.4s\n", + "31:\tlearn: 20208505.6535641\ttotal: 608ms\tremaining: 18.4s\n", + "32:\tlearn: 20203647.1730367\ttotal: 629ms\tremaining: 18.4s\n", + "33:\tlearn: 20176285.8516750\ttotal: 651ms\tremaining: 18.5s\n", + "34:\tlearn: 20023706.8473414\ttotal: 672ms\tremaining: 18.5s\n", + "35:\tlearn: 20019782.0125353\ttotal: 693ms\tremaining: 18.5s\n", + "36:\tlearn: 20016194.9834484\ttotal: 711ms\tremaining: 18.5s\n", + "37:\tlearn: 20010918.9789974\ttotal: 728ms\tremaining: 18.4s\n", + "38:\tlearn: 20005122.2496024\ttotal: 744ms\tremaining: 18.3s\n", + "39:\tlearn: 19826427.5860800\ttotal: 765ms\tremaining: 18.4s\n", + "40:\tlearn: 19821027.0706518\ttotal: 783ms\tremaining: 18.3s\n", + "41:\tlearn: 19816871.6294044\ttotal: 804ms\tremaining: 18.3s\n", + "42:\tlearn: 19810410.3099000\ttotal: 825ms\tremaining: 18.4s\n", + "43:\tlearn: 19808449.0405768\ttotal: 839ms\tremaining: 18.2s\n", + "44:\tlearn: 19807738.5191228\ttotal: 853ms\tremaining: 18.1s\n", + "45:\tlearn: 19787631.7962327\ttotal: 871ms\tremaining: 18.1s\n", + "46:\tlearn: 19782405.6845722\ttotal: 886ms\tremaining: 18s\n", + "47:\tlearn: 19781333.7652417\ttotal: 903ms\tremaining: 17.9s\n", + "48:\tlearn: 19778478.0759503\ttotal: 917ms\tremaining: 17.8s\n", + "49:\tlearn: 19772999.5998384\ttotal: 933ms\tremaining: 17.7s\n", + "50:\tlearn: 19770396.6950975\ttotal: 951ms\tremaining: 17.7s\n", + "51:\tlearn: 19757815.1021254\ttotal: 967ms\tremaining: 17.6s\n", + "52:\tlearn: 19756418.1596308\ttotal: 983ms\tremaining: 17.6s\n", + "53:\tlearn: 19740682.4658575\ttotal: 1s\tremaining: 17.6s\n", + "54:\tlearn: 19731694.5829607\ttotal: 1.02s\tremaining: 17.5s\n", + "55:\tlearn: 19729400.6708806\ttotal: 1.04s\tremaining: 17.5s\n", + "56:\tlearn: 19708097.7439061\ttotal: 1.07s\tremaining: 17.6s\n", + "57:\tlearn: 19706618.2167345\ttotal: 1.09s\tremaining: 17.7s\n", + "58:\tlearn: 19703139.1808417\ttotal: 1.11s\tremaining: 17.7s\n", + "59:\tlearn: 19699441.8400171\ttotal: 1.12s\tremaining: 17.6s\n", + "60:\tlearn: 19697329.4392297\ttotal: 1.14s\tremaining: 17.5s\n", + "61:\tlearn: 19672228.5413298\ttotal: 1.16s\tremaining: 17.5s\n", + "62:\tlearn: 19670302.3131038\ttotal: 1.18s\tremaining: 17.5s\n", + "63:\tlearn: 19636766.6788921\ttotal: 1.2s\tremaining: 17.5s\n", + "64:\tlearn: 19630850.6376277\ttotal: 1.22s\tremaining: 17.5s\n", + "65:\tlearn: 19619131.1000212\ttotal: 1.24s\tremaining: 17.5s\n", + "66:\tlearn: 19615162.5371031\ttotal: 1.26s\tremaining: 17.5s\n", + "67:\tlearn: 19612717.9376186\ttotal: 1.27s\tremaining: 17.5s\n", + "68:\tlearn: 19610093.0375816\ttotal: 1.29s\tremaining: 17.5s\n", + "69:\tlearn: 19604632.9545269\ttotal: 1.31s\tremaining: 17.5s\n", + "70:\tlearn: 19602538.7842824\ttotal: 1.33s\tremaining: 17.5s\n", + "71:\tlearn: 19599538.3152449\ttotal: 1.35s\tremaining: 17.4s\n", + "72:\tlearn: 19597720.9114956\ttotal: 1.37s\tremaining: 17.4s\n", + "73:\tlearn: 19595900.2449834\ttotal: 1.39s\tremaining: 17.3s\n", + "74:\tlearn: 19590119.2198134\ttotal: 1.41s\tremaining: 17.3s\n", + "75:\tlearn: 19563519.7751941\ttotal: 1.42s\tremaining: 17.3s\n", + "76:\tlearn: 19561396.5778182\ttotal: 1.44s\tremaining: 17.3s\n", + "77:\tlearn: 19558409.0529330\ttotal: 1.46s\tremaining: 17.3s\n", + "78:\tlearn: 19556656.6178205\ttotal: 1.48s\tremaining: 17.3s\n", + "79:\tlearn: 19555751.9045509\ttotal: 1.5s\tremaining: 17.3s\n", + "80:\tlearn: 19438777.0032411\ttotal: 1.53s\tremaining: 17.3s\n", + "81:\tlearn: 19433573.5226672\ttotal: 1.54s\tremaining: 17.3s\n", + "82:\tlearn: 19417916.7947902\ttotal: 1.56s\tremaining: 17.2s\n", + "83:\tlearn: 19414603.8189686\ttotal: 1.58s\tremaining: 17.2s\n", + "84:\tlearn: 19412946.2034238\ttotal: 1.6s\tremaining: 17.2s\n", + "85:\tlearn: 19411933.0613585\ttotal: 1.62s\tremaining: 17.2s\n", + "86:\tlearn: 19411233.9838117\ttotal: 1.64s\tremaining: 17.2s\n", + "87:\tlearn: 19393294.2612631\ttotal: 1.66s\tremaining: 17.2s\n", + "88:\tlearn: 19309658.6838140\ttotal: 1.69s\tremaining: 17.3s\n", + "89:\tlearn: 19008541.0625168\ttotal: 1.71s\tremaining: 17.2s\n", + "90:\tlearn: 19008184.7592218\ttotal: 1.72s\tremaining: 17.2s\n", + "91:\tlearn: 18757278.4678381\ttotal: 1.74s\tremaining: 17.2s\n", + "92:\tlearn: 18753814.8365244\ttotal: 1.76s\tremaining: 17.2s\n", + "93:\tlearn: 18743709.2369198\ttotal: 1.78s\tremaining: 17.2s\n", + "94:\tlearn: 18739028.5538371\ttotal: 1.8s\tremaining: 17.2s\n", + "95:\tlearn: 18736686.9989415\ttotal: 1.83s\tremaining: 17.2s\n", + "96:\tlearn: 18689358.8045410\ttotal: 1.85s\tremaining: 17.2s\n", + "97:\tlearn: 18686394.5094769\ttotal: 1.87s\tremaining: 17.2s\n", + "98:\tlearn: 18681331.3190831\ttotal: 1.89s\tremaining: 17.2s\n", + "99:\tlearn: 18679609.4138724\ttotal: 1.91s\tremaining: 17.2s\n", + "100:\tlearn: 18677402.7868206\ttotal: 1.93s\tremaining: 17.2s\n", + "101:\tlearn: 18675795.7763439\ttotal: 1.95s\tremaining: 17.2s\n", + "102:\tlearn: 18655586.7690829\ttotal: 1.97s\tremaining: 17.2s\n", + "103:\tlearn: 18572482.8219956\ttotal: 2s\tremaining: 17.2s\n", + "104:\tlearn: 18562979.6696884\ttotal: 2.02s\tremaining: 17.2s\n", + "105:\tlearn: 18560892.7036464\ttotal: 2.04s\tremaining: 17.2s\n", + "106:\tlearn: 18552826.0741258\ttotal: 2.06s\tremaining: 17.2s\n", + "107:\tlearn: 18542571.4569691\ttotal: 2.08s\tremaining: 17.2s\n", + "108:\tlearn: 18541590.1917815\ttotal: 2.1s\tremaining: 17.2s\n", + "109:\tlearn: 18525641.7974257\ttotal: 2.12s\tremaining: 17.1s\n", + "110:\tlearn: 18525356.6429839\ttotal: 2.14s\tremaining: 17.1s\n", + "111:\tlearn: 18523683.0948846\ttotal: 2.15s\tremaining: 17.1s\n", + "112:\tlearn: 18521935.9363212\ttotal: 2.17s\tremaining: 17s\n", + "113:\tlearn: 18518558.3136450\ttotal: 2.19s\tremaining: 17s\n", + "114:\tlearn: 18516818.0597770\ttotal: 2.2s\tremaining: 16.9s\n", + "115:\tlearn: 18514936.8651584\ttotal: 2.22s\tremaining: 16.9s\n", + "116:\tlearn: 18511592.2096944\ttotal: 2.23s\tremaining: 16.9s\n", + "117:\tlearn: 18509579.5237218\ttotal: 2.25s\tremaining: 16.8s\n", + "118:\tlearn: 18507548.9330612\ttotal: 2.27s\tremaining: 16.8s\n", + "119:\tlearn: 18506183.8309113\ttotal: 2.28s\tremaining: 16.7s\n", + "120:\tlearn: 18503895.7203859\ttotal: 2.3s\tremaining: 16.7s\n", + "121:\tlearn: 18491818.4017809\ttotal: 2.32s\tremaining: 16.7s\n", + "122:\tlearn: 18489744.2684678\ttotal: 2.35s\tremaining: 16.7s\n", + "123:\tlearn: 18415456.8260566\ttotal: 2.37s\tremaining: 16.7s\n", + "124:\tlearn: 18348881.2241651\ttotal: 2.39s\tremaining: 16.8s\n", + "125:\tlearn: 18311367.7530810\ttotal: 2.42s\tremaining: 16.8s\n", + "126:\tlearn: 18309766.0935271\ttotal: 2.45s\tremaining: 16.8s\n", + "127:\tlearn: 18307882.1998270\ttotal: 2.47s\tremaining: 16.8s\n", + "128:\tlearn: 18086194.0185441\ttotal: 2.5s\tremaining: 16.9s\n", + "129:\tlearn: 18054245.4959147\ttotal: 2.52s\tremaining: 16.9s\n", + "130:\tlearn: 18052465.3994692\ttotal: 2.54s\tremaining: 16.9s\n", + "131:\tlearn: 18040453.9990169\ttotal: 2.57s\tremaining: 16.9s\n", + "132:\tlearn: 18038566.6263542\ttotal: 2.58s\tremaining: 16.8s\n", + "133:\tlearn: 18036782.4977680\ttotal: 2.6s\tremaining: 16.8s\n", + "134:\tlearn: 18035468.3341409\ttotal: 2.62s\tremaining: 16.8s\n", + "135:\tlearn: 17848583.6017626\ttotal: 2.64s\tremaining: 16.8s\n", + "136:\tlearn: 17845654.5327536\ttotal: 2.66s\tremaining: 16.7s\n", + "137:\tlearn: 17842697.7991284\ttotal: 2.67s\tremaining: 16.7s\n", + "138:\tlearn: 17841628.6965791\ttotal: 2.69s\tremaining: 16.7s\n", + "139:\tlearn: 17834584.6170110\ttotal: 2.71s\tremaining: 16.6s\n", + "140:\tlearn: 17807268.5157860\ttotal: 2.72s\tremaining: 16.6s\n", + "141:\tlearn: 17805339.1253482\ttotal: 2.74s\tremaining: 16.6s\n", + "142:\tlearn: 17803353.3235615\ttotal: 2.76s\tremaining: 16.5s\n", + "143:\tlearn: 17801850.6143386\ttotal: 2.77s\tremaining: 16.5s\n", + "144:\tlearn: 17642955.2771286\ttotal: 2.79s\tremaining: 16.5s\n", + "145:\tlearn: 17641294.1196675\ttotal: 2.81s\tremaining: 16.4s\n", + "146:\tlearn: 17630140.3074217\ttotal: 2.82s\tremaining: 16.4s\n", + "147:\tlearn: 17627624.1255641\ttotal: 2.84s\tremaining: 16.4s\n", + "148:\tlearn: 17603041.8192761\ttotal: 2.86s\tremaining: 16.3s\n", + "149:\tlearn: 17601706.1883032\ttotal: 2.88s\tremaining: 16.3s\n", + "150:\tlearn: 17598105.5599714\ttotal: 2.89s\tremaining: 16.3s\n", + "151:\tlearn: 17596130.1775217\ttotal: 2.91s\tremaining: 16.2s\n", + "152:\tlearn: 17566158.2893671\ttotal: 2.93s\tremaining: 16.2s\n", + "153:\tlearn: 17561600.3479789\ttotal: 2.95s\tremaining: 16.2s\n", + "154:\tlearn: 17561066.8829644\ttotal: 2.97s\tremaining: 16.2s\n", + "155:\tlearn: 17559497.2458160\ttotal: 2.98s\tremaining: 16.1s\n", + "156:\tlearn: 17549103.1492444\ttotal: 3s\tremaining: 16.1s\n", + "157:\tlearn: 17417460.8175816\ttotal: 3.02s\tremaining: 16.1s\n", + "158:\tlearn: 17354233.0707389\ttotal: 3.03s\tremaining: 16s\n", + "159:\tlearn: 17353510.4138067\ttotal: 3.05s\tremaining: 16s\n", + "160:\tlearn: 17350918.2211301\ttotal: 3.06s\tremaining: 16s\n", + "161:\tlearn: 17340803.3815998\ttotal: 3.08s\tremaining: 16s\n", + "162:\tlearn: 17339833.2389756\ttotal: 3.1s\tremaining: 15.9s\n", + "163:\tlearn: 17337448.1962437\ttotal: 3.12s\tremaining: 15.9s\n", + "164:\tlearn: 17280514.8942759\ttotal: 3.15s\tremaining: 15.9s\n", + "165:\tlearn: 17279967.6717146\ttotal: 3.16s\tremaining: 15.9s\n", + "166:\tlearn: 17259575.6925611\ttotal: 3.18s\tremaining: 15.9s\n", + "167:\tlearn: 17256092.5283615\ttotal: 3.2s\tremaining: 15.9s\n", + "168:\tlearn: 17232370.9072862\ttotal: 3.23s\tremaining: 15.9s\n", + "169:\tlearn: 17222920.1982134\ttotal: 3.25s\tremaining: 15.9s\n", + "170:\tlearn: 17220459.5050885\ttotal: 3.27s\tremaining: 15.8s\n", + "171:\tlearn: 17108933.5754492\ttotal: 3.29s\tremaining: 15.8s\n", + "172:\tlearn: 17107858.3964618\ttotal: 3.3s\tremaining: 15.8s\n", + "173:\tlearn: 17087693.6261034\ttotal: 3.33s\tremaining: 15.8s\n", + "174:\tlearn: 17078725.8587178\ttotal: 3.35s\tremaining: 15.8s\n", + "175:\tlearn: 17076605.6995543\ttotal: 3.37s\tremaining: 15.8s\n", + "176:\tlearn: 17065489.4227360\ttotal: 3.39s\tremaining: 15.8s\n", + "177:\tlearn: 17048293.9417726\ttotal: 3.42s\tremaining: 15.8s\n", + "178:\tlearn: 16956375.7520814\ttotal: 3.44s\tremaining: 15.8s\n", + "179:\tlearn: 16951632.5745194\ttotal: 3.46s\tremaining: 15.8s\n", + "180:\tlearn: 16939921.4923755\ttotal: 3.48s\tremaining: 15.8s\n", + "181:\tlearn: 16937521.5248547\ttotal: 3.51s\tremaining: 15.8s\n", + "182:\tlearn: 16930168.2973283\ttotal: 3.52s\tremaining: 15.7s\n", + "183:\tlearn: 16928350.9826028\ttotal: 3.54s\tremaining: 15.7s\n", + "184:\tlearn: 16921041.8847808\ttotal: 3.56s\tremaining: 15.7s\n", + "185:\tlearn: 16904415.9585111\ttotal: 3.58s\tremaining: 15.7s\n", + "186:\tlearn: 16901279.0112120\ttotal: 3.6s\tremaining: 15.7s\n", + "187:\tlearn: 16899751.4709599\ttotal: 3.62s\tremaining: 15.6s\n", + "188:\tlearn: 16898724.6935600\ttotal: 3.64s\tremaining: 15.6s\n", + "189:\tlearn: 16890919.8609484\ttotal: 3.66s\tremaining: 15.6s\n", + "190:\tlearn: 16885832.6963876\ttotal: 3.67s\tremaining: 15.6s\n", + "191:\tlearn: 16884064.1703843\ttotal: 3.69s\tremaining: 15.5s\n", + "192:\tlearn: 16882805.6715898\ttotal: 3.71s\tremaining: 15.5s\n", + "193:\tlearn: 16882377.4311044\ttotal: 3.72s\tremaining: 15.5s\n", + "194:\tlearn: 16868261.4753760\ttotal: 3.74s\tremaining: 15.4s\n", + "195:\tlearn: 16867864.4150752\ttotal: 3.76s\tremaining: 15.4s\n", + "196:\tlearn: 16867144.3128568\ttotal: 3.77s\tremaining: 15.4s\n", + "197:\tlearn: 16866773.9579493\ttotal: 3.79s\tremaining: 15.4s\n", + "198:\tlearn: 16866248.4342985\ttotal: 3.81s\tremaining: 15.3s\n", + "199:\tlearn: 16865215.1244569\ttotal: 3.83s\tremaining: 15.3s\n", + "200:\tlearn: 16826208.1864283\ttotal: 3.85s\tremaining: 15.3s\n", + "201:\tlearn: 16816325.2863139\ttotal: 3.87s\tremaining: 15.3s\n", + "202:\tlearn: 16741257.5127752\ttotal: 3.89s\tremaining: 15.3s\n", + "203:\tlearn: 16738712.1206975\ttotal: 3.91s\tremaining: 15.3s\n", + "204:\tlearn: 16729911.7175693\ttotal: 3.93s\tremaining: 15.3s\n", + "205:\tlearn: 16725949.0938213\ttotal: 3.95s\tremaining: 15.2s\n", + "206:\tlearn: 16724948.4248716\ttotal: 3.97s\tremaining: 15.2s\n", + "207:\tlearn: 16682907.0572452\ttotal: 3.99s\tremaining: 15.2s\n", + "208:\tlearn: 16681759.2175568\ttotal: 4.01s\tremaining: 15.2s\n", + "209:\tlearn: 16681439.2294628\ttotal: 4.03s\tremaining: 15.1s\n", + "210:\tlearn: 16669482.5532922\ttotal: 4.05s\tremaining: 15.1s\n", + "211:\tlearn: 16653097.2592396\ttotal: 4.07s\tremaining: 15.1s\n", + "212:\tlearn: 16646075.4382241\ttotal: 4.08s\tremaining: 15.1s\n", + "213:\tlearn: 16645055.5859017\ttotal: 4.1s\tremaining: 15.1s\n", + "214:\tlearn: 16644757.0447694\ttotal: 4.12s\tremaining: 15s\n", + "215:\tlearn: 16630229.6079666\ttotal: 4.14s\tremaining: 15s\n", + "216:\tlearn: 16629140.3182322\ttotal: 4.15s\tremaining: 15s\n", + "217:\tlearn: 16628131.2044490\ttotal: 4.17s\tremaining: 15s\n", + "218:\tlearn: 16564908.6961966\ttotal: 4.19s\tremaining: 14.9s\n", + "219:\tlearn: 16562975.9514213\ttotal: 4.21s\tremaining: 14.9s\n", + "220:\tlearn: 16561066.6684942\ttotal: 4.23s\tremaining: 14.9s\n", + "221:\tlearn: 16559949.7001093\ttotal: 4.25s\tremaining: 14.9s\n", + "222:\tlearn: 16558227.2571740\ttotal: 4.27s\tremaining: 14.9s\n", + "223:\tlearn: 16557965.4672609\ttotal: 4.28s\tremaining: 14.8s\n", + "224:\tlearn: 16555916.3204459\ttotal: 4.3s\tremaining: 14.8s\n", + "225:\tlearn: 16555158.9901642\ttotal: 4.32s\tremaining: 14.8s\n", + "226:\tlearn: 16554134.5700631\ttotal: 4.34s\tremaining: 14.8s\n", + "227:\tlearn: 16553333.5497729\ttotal: 4.35s\tremaining: 14.7s\n", + "228:\tlearn: 16550835.3039861\ttotal: 4.37s\tremaining: 14.7s\n", + "229:\tlearn: 16540704.2999595\ttotal: 4.39s\tremaining: 14.7s\n", + "230:\tlearn: 16539430.0502239\ttotal: 4.4s\tremaining: 14.7s\n", + "231:\tlearn: 16538656.7542080\ttotal: 4.42s\tremaining: 14.6s\n", + "232:\tlearn: 16538215.3432778\ttotal: 4.44s\tremaining: 14.6s\n", + "233:\tlearn: 16523491.6516828\ttotal: 4.46s\tremaining: 14.6s\n", + "234:\tlearn: 16522811.4726726\ttotal: 4.48s\tremaining: 14.6s\n", + "235:\tlearn: 16522570.4993597\ttotal: 4.49s\tremaining: 14.6s\n", + "236:\tlearn: 16515326.2207752\ttotal: 4.52s\tremaining: 14.5s\n", + "237:\tlearn: 16514433.7188151\ttotal: 4.54s\tremaining: 14.5s\n", + "238:\tlearn: 16511393.9885983\ttotal: 4.56s\tremaining: 14.5s\n", + "239:\tlearn: 16509426.6581912\ttotal: 4.58s\tremaining: 14.5s\n", + "240:\tlearn: 16507868.7317227\ttotal: 4.6s\tremaining: 14.5s\n", + "241:\tlearn: 16505964.1613172\ttotal: 4.63s\tremaining: 14.5s\n", + "242:\tlearn: 16505412.5915435\ttotal: 4.64s\tremaining: 14.5s\n", + "243:\tlearn: 16502400.4263770\ttotal: 4.66s\tremaining: 14.4s\n", + "244:\tlearn: 16501653.9086014\ttotal: 4.68s\tremaining: 14.4s\n", + "245:\tlearn: 16500906.4506835\ttotal: 4.69s\tremaining: 14.4s\n", + "246:\tlearn: 16487930.8086712\ttotal: 4.71s\tremaining: 14.4s\n", + "247:\tlearn: 16480877.6383523\ttotal: 4.73s\tremaining: 14.3s\n", + "248:\tlearn: 16471564.6718598\ttotal: 4.75s\tremaining: 14.3s\n", + "249:\tlearn: 16470912.2244580\ttotal: 4.76s\tremaining: 14.3s\n", + "250:\tlearn: 16439481.0195614\ttotal: 4.78s\tremaining: 14.3s\n", + "251:\tlearn: 16438722.2145898\ttotal: 4.8s\tremaining: 14.2s\n", + "252:\tlearn: 16436581.1033330\ttotal: 4.82s\tremaining: 14.2s\n", + "253:\tlearn: 16435709.3753331\ttotal: 4.83s\tremaining: 14.2s\n", + "254:\tlearn: 16435393.3040452\ttotal: 4.85s\tremaining: 14.2s\n", + "255:\tlearn: 16427833.6475921\ttotal: 4.87s\tremaining: 14.1s\n", + "256:\tlearn: 16427538.9619444\ttotal: 4.88s\tremaining: 14.1s\n", + "257:\tlearn: 16426754.6014133\ttotal: 4.9s\tremaining: 14.1s\n", + "258:\tlearn: 16425325.4787898\ttotal: 4.91s\tremaining: 14.1s\n", + "259:\tlearn: 16424830.1071685\ttotal: 4.93s\tremaining: 14s\n", + "260:\tlearn: 16424367.4771794\ttotal: 4.95s\tremaining: 14s\n", + "261:\tlearn: 16423620.3866744\ttotal: 4.96s\tremaining: 14s\n", + "262:\tlearn: 16418460.9950502\ttotal: 4.98s\tremaining: 14s\n", + "263:\tlearn: 16418032.0692749\ttotal: 5s\tremaining: 13.9s\n", + "264:\tlearn: 16413318.0802500\ttotal: 5.02s\tremaining: 13.9s\n", + "265:\tlearn: 16412919.4639762\ttotal: 5.04s\tremaining: 13.9s\n", + "266:\tlearn: 16404758.7470010\ttotal: 5.06s\tremaining: 13.9s\n", + "267:\tlearn: 16351627.7878422\ttotal: 5.08s\tremaining: 13.9s\n", + "268:\tlearn: 16344295.2421375\ttotal: 5.1s\tremaining: 13.9s\n", + "269:\tlearn: 16331247.7157844\ttotal: 5.12s\tremaining: 13.8s\n", + "270:\tlearn: 16326915.6148344\ttotal: 5.13s\tremaining: 13.8s\n", + "271:\tlearn: 16326191.8172448\ttotal: 5.15s\tremaining: 13.8s\n", + "272:\tlearn: 16319637.7277947\ttotal: 5.17s\tremaining: 13.8s\n", + "273:\tlearn: 16318923.5839147\ttotal: 5.18s\tremaining: 13.7s\n", + "274:\tlearn: 16317169.0024984\ttotal: 5.21s\tremaining: 13.7s\n", + "275:\tlearn: 16316795.3991608\ttotal: 5.22s\tremaining: 13.7s\n", + "276:\tlearn: 16316109.7416048\ttotal: 5.24s\tremaining: 13.7s\n", + "277:\tlearn: 16310202.8104501\ttotal: 5.26s\tremaining: 13.7s\n", + "278:\tlearn: 16298459.8554384\ttotal: 5.28s\tremaining: 13.6s\n", + "279:\tlearn: 16291155.7372562\ttotal: 5.3s\tremaining: 13.6s\n", + "280:\tlearn: 16290567.5040426\ttotal: 5.32s\tremaining: 13.6s\n", + "281:\tlearn: 16276490.1501273\ttotal: 5.33s\tremaining: 13.6s\n", + "282:\tlearn: 16271927.7428385\ttotal: 5.36s\tremaining: 13.6s\n", + "283:\tlearn: 16269614.0932734\ttotal: 5.38s\tremaining: 13.6s\n", + "284:\tlearn: 16264177.6687309\ttotal: 5.4s\tremaining: 13.5s\n", + "285:\tlearn: 16260208.5522827\ttotal: 5.42s\tremaining: 13.5s\n", + "286:\tlearn: 16256576.4502796\ttotal: 5.43s\tremaining: 13.5s\n", + "287:\tlearn: 16254058.9454826\ttotal: 5.45s\tremaining: 13.5s\n", + "288:\tlearn: 16253471.6159361\ttotal: 5.47s\tremaining: 13.5s\n", + "289:\tlearn: 16242124.2487904\ttotal: 5.49s\tremaining: 13.4s\n", + "290:\tlearn: 16230782.7457944\ttotal: 5.51s\tremaining: 13.4s\n", + "291:\tlearn: 16191161.7445135\ttotal: 5.53s\tremaining: 13.4s\n", + "292:\tlearn: 16167339.9071325\ttotal: 5.55s\tremaining: 13.4s\n", + "293:\tlearn: 16159980.4064988\ttotal: 5.57s\tremaining: 13.4s\n", + "294:\tlearn: 16159399.0252395\ttotal: 5.58s\tremaining: 13.3s\n", + "295:\tlearn: 16150457.3272648\ttotal: 5.6s\tremaining: 13.3s\n", + "296:\tlearn: 16144925.9225207\ttotal: 5.61s\tremaining: 13.3s\n", + "297:\tlearn: 16140131.4634348\ttotal: 5.63s\tremaining: 13.3s\n", + "298:\tlearn: 16127231.1253291\ttotal: 5.66s\tremaining: 13.3s\n", + "299:\tlearn: 16126440.8665551\ttotal: 5.69s\tremaining: 13.3s\n", + "300:\tlearn: 16121670.0530282\ttotal: 5.72s\tremaining: 13.3s\n", + "301:\tlearn: 16118523.8621931\ttotal: 5.74s\tremaining: 13.3s\n", + "302:\tlearn: 16117237.3364004\ttotal: 5.76s\tremaining: 13.3s\n", + "303:\tlearn: 16100956.3152589\ttotal: 5.79s\tremaining: 13.3s\n", + "304:\tlearn: 16100095.8360116\ttotal: 5.81s\tremaining: 13.2s\n", + "305:\tlearn: 16054781.9357689\ttotal: 5.82s\tremaining: 13.2s\n", + "306:\tlearn: 16050463.3724197\ttotal: 5.84s\tremaining: 13.2s\n", + "307:\tlearn: 16048555.0092161\ttotal: 5.86s\tremaining: 13.2s\n", + "308:\tlearn: 16015172.9239833\ttotal: 5.88s\tremaining: 13.2s\n", + "309:\tlearn: 16009335.5533753\ttotal: 5.91s\tremaining: 13.1s\n", + "310:\tlearn: 15998648.8127455\ttotal: 5.93s\tremaining: 13.1s\n", + "311:\tlearn: 15980765.2802978\ttotal: 5.95s\tremaining: 13.1s\n", + "312:\tlearn: 15979245.1236180\ttotal: 5.96s\tremaining: 13.1s\n", + "313:\tlearn: 15978068.6903782\ttotal: 5.98s\tremaining: 13.1s\n", + "314:\tlearn: 15977679.9868077\ttotal: 5.99s\tremaining: 13s\n", + "315:\tlearn: 15973038.9221634\ttotal: 6.01s\tremaining: 13s\n", + "316:\tlearn: 15963196.8958982\ttotal: 6.03s\tremaining: 13s\n", + "317:\tlearn: 15962448.6450769\ttotal: 6.04s\tremaining: 13s\n", + "318:\tlearn: 15950478.3599818\ttotal: 6.06s\tremaining: 12.9s\n", + "319:\tlearn: 15944527.8146609\ttotal: 6.08s\tremaining: 12.9s\n", + "320:\tlearn: 15939031.3101246\ttotal: 6.1s\tremaining: 12.9s\n", + "321:\tlearn: 15924774.0481590\ttotal: 6.11s\tremaining: 12.9s\n", + "322:\tlearn: 15918158.2603557\ttotal: 6.13s\tremaining: 12.8s\n", + "323:\tlearn: 15917976.5103597\ttotal: 6.15s\tremaining: 12.8s\n", + "324:\tlearn: 15917308.9613640\ttotal: 6.17s\tremaining: 12.8s\n", + "325:\tlearn: 15916629.6017292\ttotal: 6.18s\tremaining: 12.8s\n", + "326:\tlearn: 15916123.7456763\ttotal: 6.2s\tremaining: 12.8s\n", + "327:\tlearn: 15904319.7137130\ttotal: 6.22s\tremaining: 12.8s\n", + "328:\tlearn: 15903227.2886123\ttotal: 6.25s\tremaining: 12.7s\n", + "329:\tlearn: 15902710.7326004\ttotal: 6.26s\tremaining: 12.7s\n", + "330:\tlearn: 15901051.2047678\ttotal: 6.28s\tremaining: 12.7s\n", + "331:\tlearn: 15900692.0685465\ttotal: 6.3s\tremaining: 12.7s\n", + "332:\tlearn: 15900158.3561260\ttotal: 6.33s\tremaining: 12.7s\n", + "333:\tlearn: 15899484.8247248\ttotal: 6.35s\tremaining: 12.7s\n", + "334:\tlearn: 15895616.6483881\ttotal: 6.37s\tremaining: 12.6s\n", + "335:\tlearn: 15890182.6803330\ttotal: 6.39s\tremaining: 12.6s\n", + "336:\tlearn: 15886874.9399864\ttotal: 6.41s\tremaining: 12.6s\n", + "337:\tlearn: 15875868.1894422\ttotal: 6.43s\tremaining: 12.6s\n", + "338:\tlearn: 15873032.6717763\ttotal: 6.46s\tremaining: 12.6s\n", + "339:\tlearn: 15872131.4797120\ttotal: 6.48s\tremaining: 12.6s\n", + "340:\tlearn: 15870607.2643030\ttotal: 6.49s\tremaining: 12.6s\n", + "341:\tlearn: 15870336.7168652\ttotal: 6.51s\tremaining: 12.5s\n", + "342:\tlearn: 15858142.8926846\ttotal: 6.54s\tremaining: 12.5s\n", + "343:\tlearn: 15852036.9987456\ttotal: 6.57s\tremaining: 12.5s\n", + "344:\tlearn: 15851211.3433301\ttotal: 6.58s\tremaining: 12.5s\n", + "345:\tlearn: 15847806.5747006\ttotal: 6.6s\tremaining: 12.5s\n", + "346:\tlearn: 15844645.9411026\ttotal: 6.63s\tremaining: 12.5s\n", + "347:\tlearn: 15828961.8759232\ttotal: 6.65s\tremaining: 12.5s\n", + "348:\tlearn: 15828277.1286653\ttotal: 6.67s\tremaining: 12.4s\n", + "349:\tlearn: 15806247.6086130\ttotal: 6.69s\tremaining: 12.4s\n", + "350:\tlearn: 15805767.2260881\ttotal: 6.72s\tremaining: 12.4s\n", + "351:\tlearn: 15802172.9751853\ttotal: 6.74s\tremaining: 12.4s\n", + "352:\tlearn: 15798132.2195380\ttotal: 6.76s\tremaining: 12.4s\n", + "353:\tlearn: 15774125.0526984\ttotal: 6.79s\tremaining: 12.4s\n", + "354:\tlearn: 15767494.8834016\ttotal: 6.82s\tremaining: 12.4s\n", + "355:\tlearn: 15731151.5366199\ttotal: 6.84s\tremaining: 12.4s\n", + "356:\tlearn: 15726875.2608364\ttotal: 6.87s\tremaining: 12.4s\n", + "357:\tlearn: 15724472.5494185\ttotal: 6.88s\tremaining: 12.3s\n", + "358:\tlearn: 15723297.7594400\ttotal: 6.91s\tremaining: 12.3s\n", + "359:\tlearn: 15722321.1192803\ttotal: 6.93s\tremaining: 12.3s\n", + "360:\tlearn: 15721857.6626695\ttotal: 6.95s\tremaining: 12.3s\n", + "361:\tlearn: 15710326.4707038\ttotal: 6.96s\tremaining: 12.3s\n", + "362:\tlearn: 15707760.6100833\ttotal: 6.98s\tremaining: 12.3s\n", + "363:\tlearn: 15707158.0261594\ttotal: 7s\tremaining: 12.2s\n", + "364:\tlearn: 15706003.0865566\ttotal: 7.02s\tremaining: 12.2s\n", + "365:\tlearn: 15705333.4509082\ttotal: 7.04s\tremaining: 12.2s\n", + "366:\tlearn: 15704909.3717924\ttotal: 7.05s\tremaining: 12.2s\n", + "367:\tlearn: 15704572.6361770\ttotal: 7.07s\tremaining: 12.1s\n", + "368:\tlearn: 15704091.9297076\ttotal: 7.09s\tremaining: 12.1s\n", + "369:\tlearn: 15703690.1015585\ttotal: 7.1s\tremaining: 12.1s\n", + "370:\tlearn: 15700680.6466945\ttotal: 7.12s\tremaining: 12.1s\n", + "371:\tlearn: 15700315.6448947\ttotal: 7.14s\tremaining: 12.1s\n", + "372:\tlearn: 15697040.1692281\ttotal: 7.16s\tremaining: 12s\n", + "373:\tlearn: 15696796.2204033\ttotal: 7.18s\tremaining: 12s\n", + "374:\tlearn: 15696264.4344640\ttotal: 7.19s\tremaining: 12s\n", + "375:\tlearn: 15693401.9079029\ttotal: 7.21s\tremaining: 12s\n", + "376:\tlearn: 15689250.4266617\ttotal: 7.23s\tremaining: 12s\n", + "377:\tlearn: 15672430.6718126\ttotal: 7.25s\tremaining: 11.9s\n", + "378:\tlearn: 15672223.6811755\ttotal: 7.27s\tremaining: 11.9s\n", + "379:\tlearn: 15671696.0210353\ttotal: 7.29s\tremaining: 11.9s\n", + "380:\tlearn: 15668194.2779147\ttotal: 7.31s\tremaining: 11.9s\n", + "381:\tlearn: 15657969.9944029\ttotal: 7.33s\tremaining: 11.9s\n", + "382:\tlearn: 15657528.6748760\ttotal: 7.34s\tremaining: 11.8s\n", + "383:\tlearn: 15648138.7441656\ttotal: 7.36s\tremaining: 11.8s\n", + "384:\tlearn: 15638402.0348663\ttotal: 7.38s\tremaining: 11.8s\n", + "385:\tlearn: 15638060.7715624\ttotal: 7.39s\tremaining: 11.8s\n", + "386:\tlearn: 15624483.9251685\ttotal: 7.41s\tremaining: 11.7s\n", + "387:\tlearn: 15622845.2961097\ttotal: 7.43s\tremaining: 11.7s\n", + "388:\tlearn: 15612041.0236596\ttotal: 7.45s\tremaining: 11.7s\n", + "389:\tlearn: 15603146.9662897\ttotal: 7.47s\tremaining: 11.7s\n", + "390:\tlearn: 15601990.1450766\ttotal: 7.48s\tremaining: 11.7s\n", + "391:\tlearn: 15601363.3416613\ttotal: 7.5s\tremaining: 11.6s\n", + "392:\tlearn: 15599191.2003307\ttotal: 7.52s\tremaining: 11.6s\n", + "393:\tlearn: 15598570.0990730\ttotal: 7.54s\tremaining: 11.6s\n", + "394:\tlearn: 15596230.5863681\ttotal: 7.55s\tremaining: 11.6s\n", + "395:\tlearn: 15594189.1350767\ttotal: 7.57s\tremaining: 11.5s\n", + "396:\tlearn: 15593961.6649448\ttotal: 7.58s\tremaining: 11.5s\n", + "397:\tlearn: 15593760.1873024\ttotal: 7.6s\tremaining: 11.5s\n", + "398:\tlearn: 15592809.7875300\ttotal: 7.62s\tremaining: 11.5s\n", + "399:\tlearn: 15589855.2174745\ttotal: 7.63s\tremaining: 11.4s\n", + "400:\tlearn: 15581318.3623033\ttotal: 7.65s\tremaining: 11.4s\n", + "401:\tlearn: 15577270.4245693\ttotal: 7.67s\tremaining: 11.4s\n", + "402:\tlearn: 15576952.7417484\ttotal: 7.68s\tremaining: 11.4s\n", + "403:\tlearn: 15576586.3492924\ttotal: 7.7s\tremaining: 11.4s\n", + "404:\tlearn: 15571167.9875037\ttotal: 7.72s\tremaining: 11.3s\n", + "405:\tlearn: 15569357.4405457\ttotal: 7.74s\tremaining: 11.3s\n", + "406:\tlearn: 15569128.3887365\ttotal: 7.76s\tremaining: 11.3s\n", + "407:\tlearn: 15568355.8228671\ttotal: 7.77s\tremaining: 11.3s\n", + "408:\tlearn: 15552620.3053293\ttotal: 7.79s\tremaining: 11.3s\n", + "409:\tlearn: 15544601.9052501\ttotal: 7.81s\tremaining: 11.2s\n", + "410:\tlearn: 15536694.2284926\ttotal: 7.83s\tremaining: 11.2s\n", + "411:\tlearn: 15536245.0468606\ttotal: 7.85s\tremaining: 11.2s\n", + "412:\tlearn: 15532278.4875558\ttotal: 7.87s\tremaining: 11.2s\n", + "413:\tlearn: 15511482.8547340\ttotal: 7.88s\tremaining: 11.2s\n", + "414:\tlearn: 15511021.6526090\ttotal: 7.9s\tremaining: 11.1s\n", + "415:\tlearn: 15508419.8137283\ttotal: 7.92s\tremaining: 11.1s\n", + "416:\tlearn: 15507384.4861211\ttotal: 7.93s\tremaining: 11.1s\n", + "417:\tlearn: 15507194.7228751\ttotal: 7.95s\tremaining: 11.1s\n", + "418:\tlearn: 15506862.7422276\ttotal: 7.97s\tremaining: 11s\n", + "419:\tlearn: 15506153.8325314\ttotal: 7.99s\tremaining: 11s\n", + "420:\tlearn: 15496691.3479538\ttotal: 8s\tremaining: 11s\n", + "421:\tlearn: 15481503.6612742\ttotal: 8.02s\tremaining: 11s\n", + "422:\tlearn: 15480916.8373585\ttotal: 8.04s\tremaining: 11s\n", + "423:\tlearn: 15477965.4399634\ttotal: 8.05s\tremaining: 10.9s\n", + "424:\tlearn: 15476411.7695084\ttotal: 8.07s\tremaining: 10.9s\n", + "425:\tlearn: 15468806.7207207\ttotal: 8.09s\tremaining: 10.9s\n", + "426:\tlearn: 15468191.0361515\ttotal: 8.11s\tremaining: 10.9s\n", + "427:\tlearn: 15462223.8900134\ttotal: 8.13s\tremaining: 10.9s\n", + "428:\tlearn: 15461803.2638937\ttotal: 8.14s\tremaining: 10.8s\n", + "429:\tlearn: 15461674.6665670\ttotal: 8.16s\tremaining: 10.8s\n", + "430:\tlearn: 15461479.5870748\ttotal: 8.18s\tremaining: 10.8s\n", + "431:\tlearn: 15453195.7741336\ttotal: 8.19s\tremaining: 10.8s\n", + "432:\tlearn: 15453014.5225347\ttotal: 8.21s\tremaining: 10.7s\n", + "433:\tlearn: 15443545.6603354\ttotal: 8.23s\tremaining: 10.7s\n", + "434:\tlearn: 15441941.3301205\ttotal: 8.24s\tremaining: 10.7s\n", + "435:\tlearn: 15427097.1810393\ttotal: 8.26s\tremaining: 10.7s\n", + "436:\tlearn: 15426444.8850360\ttotal: 8.28s\tremaining: 10.7s\n", + "437:\tlearn: 15425931.8135958\ttotal: 8.3s\tremaining: 10.6s\n", + "438:\tlearn: 15425636.6510092\ttotal: 8.31s\tremaining: 10.6s\n", + "439:\tlearn: 15408527.3130183\ttotal: 8.33s\tremaining: 10.6s\n", + "440:\tlearn: 15401292.0399180\ttotal: 8.35s\tremaining: 10.6s\n", + "441:\tlearn: 15400300.0007984\ttotal: 8.37s\tremaining: 10.6s\n", + "442:\tlearn: 15385821.7387360\ttotal: 8.38s\tremaining: 10.5s\n", + "443:\tlearn: 15372274.6725325\ttotal: 8.4s\tremaining: 10.5s\n", + "444:\tlearn: 15358531.6955122\ttotal: 8.42s\tremaining: 10.5s\n", + "445:\tlearn: 15356630.4066797\ttotal: 8.44s\tremaining: 10.5s\n", + "446:\tlearn: 15356309.2575797\ttotal: 8.46s\tremaining: 10.5s\n", + "447:\tlearn: 15356011.0701583\ttotal: 8.47s\tremaining: 10.4s\n", + "448:\tlearn: 15347229.8183089\ttotal: 8.49s\tremaining: 10.4s\n", + "449:\tlearn: 15347025.3608803\ttotal: 8.51s\tremaining: 10.4s\n", + "450:\tlearn: 15345286.7109960\ttotal: 8.53s\tremaining: 10.4s\n", + "451:\tlearn: 15344124.7316516\ttotal: 8.55s\tremaining: 10.4s\n", + "452:\tlearn: 15343817.2840735\ttotal: 8.56s\tremaining: 10.3s\n", + "453:\tlearn: 15343647.5665711\ttotal: 8.58s\tremaining: 10.3s\n", + "454:\tlearn: 15336925.2172151\ttotal: 8.6s\tremaining: 10.3s\n", + "455:\tlearn: 15336698.2993745\ttotal: 8.62s\tremaining: 10.3s\n", + "456:\tlearn: 15329979.7120713\ttotal: 8.63s\tremaining: 10.3s\n", + "457:\tlearn: 15327261.0806467\ttotal: 8.65s\tremaining: 10.2s\n", + "458:\tlearn: 15323010.2123614\ttotal: 8.67s\tremaining: 10.2s\n", + "459:\tlearn: 15311019.0130373\ttotal: 8.69s\tremaining: 10.2s\n", + "460:\tlearn: 15310926.3865304\ttotal: 8.71s\tremaining: 10.2s\n", + "461:\tlearn: 15310336.2245889\ttotal: 8.72s\tremaining: 10.2s\n", + "462:\tlearn: 15310110.4541169\ttotal: 8.73s\tremaining: 10.1s\n", + "463:\tlearn: 15303072.1276272\ttotal: 8.75s\tremaining: 10.1s\n", + "464:\tlearn: 15302242.5369794\ttotal: 8.77s\tremaining: 10.1s\n", + "465:\tlearn: 15297608.7119635\ttotal: 8.79s\tremaining: 10.1s\n", + "466:\tlearn: 15296797.0467934\ttotal: 8.81s\tremaining: 10.1s\n", + "467:\tlearn: 15296300.0517609\ttotal: 8.83s\tremaining: 10s\n", + "468:\tlearn: 15296063.8959076\ttotal: 8.85s\tremaining: 10s\n", + "469:\tlearn: 15295732.1346916\ttotal: 8.86s\tremaining: 9.99s\n", + "470:\tlearn: 15292857.3473687\ttotal: 8.88s\tremaining: 9.97s\n", + "471:\tlearn: 15286918.7601534\ttotal: 8.9s\tremaining: 9.95s\n", + "472:\tlearn: 15286315.3203651\ttotal: 8.92s\tremaining: 9.93s\n", + "473:\tlearn: 15285879.4191647\ttotal: 8.93s\tremaining: 9.91s\n", + "474:\tlearn: 15285274.3372733\ttotal: 8.95s\tremaining: 9.89s\n", + "475:\tlearn: 15256344.7627420\ttotal: 8.97s\tremaining: 9.88s\n", + "476:\tlearn: 15253934.9737803\ttotal: 8.99s\tremaining: 9.86s\n", + "477:\tlearn: 15246843.4949940\ttotal: 9.01s\tremaining: 9.84s\n", + "478:\tlearn: 15243343.3462813\ttotal: 9.03s\tremaining: 9.82s\n", + "479:\tlearn: 15242572.9694361\ttotal: 9.05s\tremaining: 9.8s\n", + "480:\tlearn: 15241326.4672520\ttotal: 9.06s\tremaining: 9.78s\n", + "481:\tlearn: 15239166.7953532\ttotal: 9.08s\tremaining: 9.76s\n", + "482:\tlearn: 15238928.3561927\ttotal: 9.1s\tremaining: 9.74s\n", + "483:\tlearn: 15236052.2752831\ttotal: 9.12s\tremaining: 9.72s\n", + "484:\tlearn: 15235679.3228742\ttotal: 9.13s\tremaining: 9.7s\n", + "485:\tlearn: 15232651.3693052\ttotal: 9.15s\tremaining: 9.68s\n", + "486:\tlearn: 15229975.6083089\ttotal: 9.17s\tremaining: 9.66s\n", + "487:\tlearn: 15229313.9998434\ttotal: 9.19s\tremaining: 9.64s\n", + "488:\tlearn: 15226438.0742381\ttotal: 9.21s\tremaining: 9.62s\n", + "489:\tlearn: 15226207.1844110\ttotal: 9.22s\tremaining: 9.6s\n", + "490:\tlearn: 15225891.0814311\ttotal: 9.24s\tremaining: 9.58s\n", + "491:\tlearn: 15225564.2658996\ttotal: 9.26s\tremaining: 9.56s\n", + "492:\tlearn: 15225253.9290857\ttotal: 9.27s\tremaining: 9.54s\n", + "493:\tlearn: 15225101.8010381\ttotal: 9.29s\tremaining: 9.52s\n", + "494:\tlearn: 15224892.3278437\ttotal: 9.31s\tremaining: 9.5s\n", + "495:\tlearn: 15224266.9506867\ttotal: 9.33s\tremaining: 9.48s\n", + "496:\tlearn: 15223549.6289167\ttotal: 9.35s\tremaining: 9.46s\n", + "497:\tlearn: 15217058.9143160\ttotal: 9.37s\tremaining: 9.44s\n", + "498:\tlearn: 15215188.1396205\ttotal: 9.39s\tremaining: 9.42s\n", + "499:\tlearn: 15213406.6164570\ttotal: 9.4s\tremaining: 9.4s\n", + "500:\tlearn: 15207254.2720114\ttotal: 9.42s\tremaining: 9.39s\n", + "501:\tlearn: 15205153.9646281\ttotal: 9.45s\tremaining: 9.37s\n", + "502:\tlearn: 15200806.1942231\ttotal: 9.46s\tremaining: 9.35s\n", + "503:\tlearn: 15199634.6987740\ttotal: 9.48s\tremaining: 9.33s\n", + "504:\tlearn: 15199380.4930022\ttotal: 9.5s\tremaining: 9.31s\n", + "505:\tlearn: 15198823.7154717\ttotal: 9.51s\tremaining: 9.29s\n", + "506:\tlearn: 15198383.1739967\ttotal: 9.53s\tremaining: 9.27s\n", + "507:\tlearn: 15196517.8231824\ttotal: 9.55s\tremaining: 9.25s\n", + "508:\tlearn: 15196123.7863901\ttotal: 9.57s\tremaining: 9.23s\n", + "509:\tlearn: 15193553.5984846\ttotal: 9.59s\tremaining: 9.21s\n", + "510:\tlearn: 15192915.8305172\ttotal: 9.61s\tremaining: 9.19s\n", + "511:\tlearn: 15192433.8180747\ttotal: 9.62s\tremaining: 9.17s\n", + "512:\tlearn: 15192232.7761378\ttotal: 9.64s\tremaining: 9.15s\n", + "513:\tlearn: 15191287.7774708\ttotal: 9.66s\tremaining: 9.13s\n", + "514:\tlearn: 15185944.6629542\ttotal: 9.67s\tremaining: 9.11s\n", + "515:\tlearn: 15185636.9427557\ttotal: 9.69s\tremaining: 9.09s\n", + "516:\tlearn: 15179818.6448808\ttotal: 9.71s\tremaining: 9.07s\n", + "517:\tlearn: 15166631.2810761\ttotal: 9.73s\tremaining: 9.05s\n", + "518:\tlearn: 15162871.1512677\ttotal: 9.75s\tremaining: 9.04s\n", + "519:\tlearn: 15162079.3860934\ttotal: 9.77s\tremaining: 9.02s\n", + "520:\tlearn: 15159062.9503455\ttotal: 9.78s\tremaining: 8.99s\n", + "521:\tlearn: 15158581.5138320\ttotal: 9.8s\tremaining: 8.98s\n", + "522:\tlearn: 15158391.2959379\ttotal: 9.82s\tremaining: 8.95s\n", + "523:\tlearn: 15152869.4477779\ttotal: 9.84s\tremaining: 8.93s\n", + "524:\tlearn: 15139481.6776876\ttotal: 9.86s\tremaining: 8.92s\n", + "525:\tlearn: 15139265.7200114\ttotal: 9.87s\tremaining: 8.89s\n", + "526:\tlearn: 15138627.9396250\ttotal: 9.89s\tremaining: 8.88s\n", + "527:\tlearn: 15138327.3985020\ttotal: 9.9s\tremaining: 8.86s\n", + "528:\tlearn: 15137232.1130610\ttotal: 9.92s\tremaining: 8.84s\n", + "529:\tlearn: 15133609.2868548\ttotal: 9.94s\tremaining: 8.82s\n", + "530:\tlearn: 15133460.8763140\ttotal: 9.96s\tremaining: 8.8s\n", + "531:\tlearn: 15109552.7299063\ttotal: 9.98s\tremaining: 8.78s\n", + "532:\tlearn: 15107658.2624187\ttotal: 9.99s\tremaining: 8.76s\n", + "533:\tlearn: 15107455.0365402\ttotal: 10s\tremaining: 8.74s\n", + "534:\tlearn: 15102205.8936957\ttotal: 10s\tremaining: 8.72s\n", + "535:\tlearn: 15101649.0029598\ttotal: 10s\tremaining: 8.7s\n", + "536:\tlearn: 15098173.9534708\ttotal: 10.1s\tremaining: 8.68s\n", + "537:\tlearn: 15085621.2661752\ttotal: 10.1s\tremaining: 8.66s\n", + "538:\tlearn: 15082526.6318871\ttotal: 10.1s\tremaining: 8.64s\n", + "539:\tlearn: 15074383.1362937\ttotal: 10.1s\tremaining: 8.62s\n", + "540:\tlearn: 15073773.2640630\ttotal: 10.1s\tremaining: 8.6s\n", + "541:\tlearn: 15063446.7982745\ttotal: 10.2s\tremaining: 8.58s\n", + "542:\tlearn: 15055973.0283975\ttotal: 10.2s\tremaining: 8.56s\n", + "543:\tlearn: 15053040.2567599\ttotal: 10.2s\tremaining: 8.54s\n", + "544:\tlearn: 15041098.8477743\ttotal: 10.2s\tremaining: 8.52s\n", + "545:\tlearn: 15040727.8174604\ttotal: 10.2s\tremaining: 8.5s\n", + "546:\tlearn: 15034658.7375625\ttotal: 10.2s\tremaining: 8.48s\n", + "547:\tlearn: 15033907.3523765\ttotal: 10.3s\tremaining: 8.47s\n", + "548:\tlearn: 15021509.6398422\ttotal: 10.3s\tremaining: 8.45s\n", + "549:\tlearn: 15021114.6655523\ttotal: 10.3s\tremaining: 8.43s\n", + "550:\tlearn: 15020965.6552495\ttotal: 10.3s\tremaining: 8.41s\n", + "551:\tlearn: 15020305.1478324\ttotal: 10.3s\tremaining: 8.4s\n", + "552:\tlearn: 15020011.6987184\ttotal: 10.4s\tremaining: 8.38s\n", + "553:\tlearn: 15008656.9019111\ttotal: 10.4s\tremaining: 8.36s\n", + "554:\tlearn: 15008257.1960286\ttotal: 10.4s\tremaining: 8.35s\n", + "555:\tlearn: 15008047.0288160\ttotal: 10.4s\tremaining: 8.32s\n", + "556:\tlearn: 15007888.1230327\ttotal: 10.4s\tremaining: 8.3s\n", + "557:\tlearn: 15007709.6828693\ttotal: 10.5s\tremaining: 8.29s\n", + "558:\tlearn: 15007209.9707405\ttotal: 10.5s\tremaining: 8.27s\n", + "559:\tlearn: 15007071.5552244\ttotal: 10.5s\tremaining: 8.25s\n", + "560:\tlearn: 15006945.7227292\ttotal: 10.5s\tremaining: 8.22s\n", + "561:\tlearn: 15006659.1182026\ttotal: 10.5s\tremaining: 8.21s\n", + "562:\tlearn: 14990956.2777262\ttotal: 10.6s\tremaining: 8.19s\n", + "563:\tlearn: 14990396.4483383\ttotal: 10.6s\tremaining: 8.18s\n", + "564:\tlearn: 14987445.9684421\ttotal: 10.6s\tremaining: 8.16s\n", + "565:\tlearn: 14987242.4937056\ttotal: 10.6s\tremaining: 8.14s\n", + "566:\tlearn: 14986914.3913112\ttotal: 10.6s\tremaining: 8.13s\n", + "567:\tlearn: 14985397.3680809\ttotal: 10.7s\tremaining: 8.11s\n", + "568:\tlearn: 14983620.9809238\ttotal: 10.7s\tremaining: 8.09s\n", + "569:\tlearn: 14983502.4657892\ttotal: 10.7s\tremaining: 8.07s\n", + "570:\tlearn: 14981044.0324593\ttotal: 10.7s\tremaining: 8.05s\n", + "571:\tlearn: 14980723.2716378\ttotal: 10.7s\tremaining: 8.04s\n", + "572:\tlearn: 14980572.1170260\ttotal: 10.8s\tremaining: 8.02s\n", + "573:\tlearn: 14973680.7983651\ttotal: 10.8s\tremaining: 8s\n", + "574:\tlearn: 14973048.9211978\ttotal: 10.8s\tremaining: 7.98s\n", + "575:\tlearn: 14972915.1126009\ttotal: 10.8s\tremaining: 7.96s\n", + "576:\tlearn: 14972199.0905777\ttotal: 10.8s\tremaining: 7.95s\n", + "577:\tlearn: 14971983.7721189\ttotal: 10.9s\tremaining: 7.93s\n", + "578:\tlearn: 14957898.9717928\ttotal: 10.9s\tremaining: 7.91s\n", + "579:\tlearn: 14957705.1186241\ttotal: 10.9s\tremaining: 7.89s\n", + "580:\tlearn: 14956864.6464907\ttotal: 10.9s\tremaining: 7.88s\n", + "581:\tlearn: 14956793.5985069\ttotal: 10.9s\tremaining: 7.86s\n", + "582:\tlearn: 14936615.4020320\ttotal: 11s\tremaining: 7.84s\n", + "583:\tlearn: 14927341.5935344\ttotal: 11s\tremaining: 7.82s\n", + "584:\tlearn: 14927173.0473101\ttotal: 11s\tremaining: 7.81s\n", + "585:\tlearn: 14925197.4629218\ttotal: 11s\tremaining: 7.79s\n", + "586:\tlearn: 14925098.4418921\ttotal: 11s\tremaining: 7.77s\n", + "587:\tlearn: 14924720.4012434\ttotal: 11.1s\tremaining: 7.75s\n", + "588:\tlearn: 14924548.0850986\ttotal: 11.1s\tremaining: 7.73s\n", + "589:\tlearn: 14924429.2484482\ttotal: 11.1s\tremaining: 7.71s\n", + "590:\tlearn: 14924249.9492593\ttotal: 11.1s\tremaining: 7.7s\n", + "591:\tlearn: 14922760.7637420\ttotal: 11.1s\tremaining: 7.68s\n", + "592:\tlearn: 14922668.6039019\ttotal: 11.2s\tremaining: 7.66s\n", + "593:\tlearn: 14921904.1544328\ttotal: 11.2s\tremaining: 7.64s\n", + "594:\tlearn: 14921616.5712965\ttotal: 11.2s\tremaining: 7.63s\n", + "595:\tlearn: 14921504.4340608\ttotal: 11.2s\tremaining: 7.61s\n", + "596:\tlearn: 14916470.2050042\ttotal: 11.2s\tremaining: 7.59s\n", + "597:\tlearn: 14910126.6958298\ttotal: 11.3s\tremaining: 7.58s\n", + "598:\tlearn: 14910008.9874822\ttotal: 11.3s\tremaining: 7.56s\n", + "599:\tlearn: 14897073.0426329\ttotal: 11.3s\tremaining: 7.54s\n", + "600:\tlearn: 14895441.4673869\ttotal: 11.3s\tremaining: 7.52s\n", + "601:\tlearn: 14886302.0907964\ttotal: 11.3s\tremaining: 7.5s\n", + "602:\tlearn: 14885801.5005932\ttotal: 11.4s\tremaining: 7.48s\n", + "603:\tlearn: 14882503.0646614\ttotal: 11.4s\tremaining: 7.46s\n", + "604:\tlearn: 14874019.3400341\ttotal: 11.4s\tremaining: 7.45s\n", + "605:\tlearn: 14873500.2430527\ttotal: 11.4s\tremaining: 7.43s\n", + "606:\tlearn: 14865462.6336328\ttotal: 11.5s\tremaining: 7.42s\n", + "607:\tlearn: 14862999.7669828\ttotal: 11.5s\tremaining: 7.4s\n", + "608:\tlearn: 14861224.3133684\ttotal: 11.5s\tremaining: 7.39s\n", + "609:\tlearn: 14861075.6580339\ttotal: 11.5s\tremaining: 7.37s\n", + "610:\tlearn: 14860568.8273555\ttotal: 11.6s\tremaining: 7.36s\n", + "611:\tlearn: 14859416.0190514\ttotal: 11.6s\tremaining: 7.34s\n", + "612:\tlearn: 14858007.8996720\ttotal: 11.6s\tremaining: 7.32s\n", + "613:\tlearn: 14841059.8536694\ttotal: 11.6s\tremaining: 7.31s\n", + "614:\tlearn: 14825984.3023723\ttotal: 11.6s\tremaining: 7.29s\n", + "615:\tlearn: 14825713.0129450\ttotal: 11.7s\tremaining: 7.27s\n", + "616:\tlearn: 14825155.2269332\ttotal: 11.7s\tremaining: 7.26s\n", + "617:\tlearn: 14825042.0195200\ttotal: 11.7s\tremaining: 7.24s\n", + "618:\tlearn: 14822382.7492341\ttotal: 11.7s\tremaining: 7.23s\n", + "619:\tlearn: 14822270.4983850\ttotal: 11.8s\tremaining: 7.21s\n", + "620:\tlearn: 14809251.3212298\ttotal: 11.8s\tremaining: 7.19s\n", + "621:\tlearn: 14809144.4174146\ttotal: 11.8s\tremaining: 7.18s\n", + "622:\tlearn: 14808206.9650871\ttotal: 11.8s\tremaining: 7.17s\n", + "623:\tlearn: 14807763.1153500\ttotal: 11.9s\tremaining: 7.16s\n", + "624:\tlearn: 14807694.0482449\ttotal: 11.9s\tremaining: 7.14s\n", + "625:\tlearn: 14803545.1572635\ttotal: 11.9s\tremaining: 7.13s\n", + "626:\tlearn: 14789036.7759627\ttotal: 11.9s\tremaining: 7.11s\n", + "627:\tlearn: 14788794.9986179\ttotal: 12s\tremaining: 7.09s\n", + "628:\tlearn: 14788688.3809499\ttotal: 12s\tremaining: 7.07s\n", + "629:\tlearn: 14788541.8835870\ttotal: 12s\tremaining: 7.05s\n", + "630:\tlearn: 14778230.8477320\ttotal: 12s\tremaining: 7.04s\n", + "631:\tlearn: 14766980.2464707\ttotal: 12.1s\tremaining: 7.02s\n", + "632:\tlearn: 14762227.9098513\ttotal: 12.1s\tremaining: 7s\n", + "633:\tlearn: 14757495.8103489\ttotal: 12.1s\tremaining: 6.98s\n", + "634:\tlearn: 14757376.1815109\ttotal: 12.1s\tremaining: 6.96s\n", + "635:\tlearn: 14756083.0412951\ttotal: 12.1s\tremaining: 6.95s\n", + "636:\tlearn: 14753767.4530116\ttotal: 12.2s\tremaining: 6.93s\n", + "637:\tlearn: 14753290.3447435\ttotal: 12.2s\tremaining: 6.91s\n", + "638:\tlearn: 14753022.7557700\ttotal: 12.2s\tremaining: 6.89s\n", + "639:\tlearn: 14740249.2596392\ttotal: 12.2s\tremaining: 6.87s\n", + "640:\tlearn: 14736397.7056314\ttotal: 12.2s\tremaining: 6.85s\n", + "641:\tlearn: 14736172.9141301\ttotal: 12.3s\tremaining: 6.83s\n", + "642:\tlearn: 14731660.9563354\ttotal: 12.3s\tremaining: 6.81s\n", + "643:\tlearn: 14729836.0515586\ttotal: 12.3s\tremaining: 6.79s\n", + "644:\tlearn: 14720035.6906060\ttotal: 12.3s\tremaining: 6.78s\n", + "645:\tlearn: 14719242.6927172\ttotal: 12.3s\tremaining: 6.76s\n", + "646:\tlearn: 14716028.6349038\ttotal: 12.3s\tremaining: 6.74s\n", + "647:\tlearn: 14715855.5810228\ttotal: 12.4s\tremaining: 6.71s\n", + "648:\tlearn: 14711356.1065186\ttotal: 12.4s\tremaining: 6.7s\n", + "649:\tlearn: 14703879.5938587\ttotal: 12.4s\tremaining: 6.67s\n", + "650:\tlearn: 14701959.0037503\ttotal: 12.4s\tremaining: 6.65s\n", + "651:\tlearn: 14694579.1428590\ttotal: 12.4s\tremaining: 6.63s\n", + "652:\tlearn: 14694447.7757472\ttotal: 12.4s\tremaining: 6.61s\n", + "653:\tlearn: 14694346.4527135\ttotal: 12.5s\tremaining: 6.59s\n", + "654:\tlearn: 14694282.0699231\ttotal: 12.5s\tremaining: 6.57s\n", + "655:\tlearn: 14692672.9854013\ttotal: 12.5s\tremaining: 6.55s\n", + "656:\tlearn: 14691720.9415672\ttotal: 12.5s\tremaining: 6.53s\n", + "657:\tlearn: 14691539.7097721\ttotal: 12.5s\tremaining: 6.51s\n", + "658:\tlearn: 14691425.5541377\ttotal: 12.5s\tremaining: 6.49s\n", + "659:\tlearn: 14690590.5168329\ttotal: 12.6s\tremaining: 6.47s\n", + "660:\tlearn: 14690157.7456528\ttotal: 12.6s\tremaining: 6.45s\n", + "661:\tlearn: 14688578.7084883\ttotal: 12.6s\tremaining: 6.43s\n", + "662:\tlearn: 14680972.4839535\ttotal: 12.6s\tremaining: 6.41s\n", + "663:\tlearn: 14671047.3675193\ttotal: 12.6s\tremaining: 6.39s\n", + "664:\tlearn: 14670912.0116672\ttotal: 12.6s\tremaining: 6.37s\n", + "665:\tlearn: 14670742.4164308\ttotal: 12.7s\tremaining: 6.35s\n", + "666:\tlearn: 14670612.3448712\ttotal: 12.7s\tremaining: 6.33s\n", + "667:\tlearn: 14667180.9906929\ttotal: 12.7s\tremaining: 6.31s\n", + "668:\tlearn: 14657593.1840317\ttotal: 12.7s\tremaining: 6.29s\n", + "669:\tlearn: 14657416.5658364\ttotal: 12.7s\tremaining: 6.27s\n", + "670:\tlearn: 14649340.4331746\ttotal: 12.7s\tremaining: 6.25s\n", + "671:\tlearn: 14649117.3383210\ttotal: 12.8s\tremaining: 6.23s\n", + "672:\tlearn: 14646999.1447110\ttotal: 12.8s\tremaining: 6.21s\n", + "673:\tlearn: 14646883.2358912\ttotal: 12.8s\tremaining: 6.19s\n", + "674:\tlearn: 14646693.4691223\ttotal: 12.8s\tremaining: 6.17s\n", + "675:\tlearn: 14646427.3184101\ttotal: 12.8s\tremaining: 6.15s\n", + "676:\tlearn: 14638983.7297169\ttotal: 12.8s\tremaining: 6.13s\n", + "677:\tlearn: 14632759.5485617\ttotal: 12.9s\tremaining: 6.11s\n", + "678:\tlearn: 14630933.2619290\ttotal: 12.9s\tremaining: 6.09s\n", + "679:\tlearn: 14630504.9824220\ttotal: 12.9s\tremaining: 6.07s\n", + "680:\tlearn: 14630187.6471958\ttotal: 12.9s\tremaining: 6.05s\n", + "681:\tlearn: 14621229.4404689\ttotal: 12.9s\tremaining: 6.03s\n", + "682:\tlearn: 14619264.7885720\ttotal: 13s\tremaining: 6.01s\n", + "683:\tlearn: 14618533.6254987\ttotal: 13s\tremaining: 5.99s\n", + "684:\tlearn: 14618199.1681064\ttotal: 13s\tremaining: 5.97s\n", + "685:\tlearn: 14617552.3724416\ttotal: 13s\tremaining: 5.95s\n", + "686:\tlearn: 14616075.9835293\ttotal: 13s\tremaining: 5.93s\n", + "687:\tlearn: 14615027.0844831\ttotal: 13s\tremaining: 5.91s\n", + "688:\tlearn: 14603166.3928840\ttotal: 13.1s\tremaining: 5.89s\n", + "689:\tlearn: 14602919.4938364\ttotal: 13.1s\tremaining: 5.87s\n", + "690:\tlearn: 14601251.8111274\ttotal: 13.1s\tremaining: 5.85s\n", + "691:\tlearn: 14596937.5458267\ttotal: 13.1s\tremaining: 5.83s\n", + "692:\tlearn: 14588431.1296637\ttotal: 13.1s\tremaining: 5.82s\n", + "693:\tlearn: 14581396.6142492\ttotal: 13.1s\tremaining: 5.79s\n", + "694:\tlearn: 14576640.2415778\ttotal: 13.2s\tremaining: 5.78s\n", + "695:\tlearn: 14572975.8845101\ttotal: 13.2s\tremaining: 5.76s\n", + "696:\tlearn: 14566867.0286308\ttotal: 13.2s\tremaining: 5.74s\n", + "697:\tlearn: 14560679.2228769\ttotal: 13.2s\tremaining: 5.72s\n", + "698:\tlearn: 14559934.4479083\ttotal: 13.2s\tremaining: 5.7s\n", + "699:\tlearn: 14559445.0309658\ttotal: 13.3s\tremaining: 5.68s\n", + "700:\tlearn: 14559340.3958885\ttotal: 13.3s\tremaining: 5.66s\n", + "701:\tlearn: 14556887.9166841\ttotal: 13.3s\tremaining: 5.64s\n", + "702:\tlearn: 14556769.9275109\ttotal: 13.3s\tremaining: 5.62s\n", + "703:\tlearn: 14556665.0100120\ttotal: 13.3s\tremaining: 5.6s\n", + "704:\tlearn: 14556497.0075167\ttotal: 13.3s\tremaining: 5.58s\n", + "705:\tlearn: 14555063.1170817\ttotal: 13.4s\tremaining: 5.56s\n", + "706:\tlearn: 14553042.7361420\ttotal: 13.4s\tremaining: 5.54s\n", + "707:\tlearn: 14552708.2241446\ttotal: 13.4s\tremaining: 5.52s\n", + "708:\tlearn: 14552599.2682841\ttotal: 13.4s\tremaining: 5.5s\n", + "709:\tlearn: 14552382.0908993\ttotal: 13.4s\tremaining: 5.48s\n", + "710:\tlearn: 14552142.4536412\ttotal: 13.4s\tremaining: 5.46s\n", + "711:\tlearn: 14552000.4502329\ttotal: 13.5s\tremaining: 5.44s\n", + "712:\tlearn: 14536742.2382155\ttotal: 13.5s\tremaining: 5.42s\n", + "713:\tlearn: 14536667.3161668\ttotal: 13.5s\tremaining: 5.4s\n", + "714:\tlearn: 14536304.3680604\ttotal: 13.5s\tremaining: 5.38s\n", + "715:\tlearn: 14536182.7158109\ttotal: 13.5s\tremaining: 5.36s\n", + "716:\tlearn: 14535967.7392353\ttotal: 13.5s\tremaining: 5.34s\n", + "717:\tlearn: 14534508.6142506\ttotal: 13.6s\tremaining: 5.33s\n", + "718:\tlearn: 14528638.1263730\ttotal: 13.6s\tremaining: 5.31s\n", + "719:\tlearn: 14517357.7804254\ttotal: 13.6s\tremaining: 5.29s\n", + "720:\tlearn: 14516851.6147617\ttotal: 13.6s\tremaining: 5.27s\n", + "721:\tlearn: 14511638.9369384\ttotal: 13.6s\tremaining: 5.25s\n", + "722:\tlearn: 14503521.8166802\ttotal: 13.7s\tremaining: 5.23s\n", + "723:\tlearn: 14502188.4019520\ttotal: 13.7s\tremaining: 5.21s\n", + "724:\tlearn: 14500440.0215795\ttotal: 13.7s\tremaining: 5.19s\n", + "725:\tlearn: 14499768.4571353\ttotal: 13.7s\tremaining: 5.17s\n", + "726:\tlearn: 14499524.8931128\ttotal: 13.7s\tremaining: 5.15s\n", + "727:\tlearn: 14499176.3274553\ttotal: 13.7s\tremaining: 5.13s\n", + "728:\tlearn: 14498980.3385646\ttotal: 13.8s\tremaining: 5.11s\n", + "729:\tlearn: 14498855.9496290\ttotal: 13.8s\tremaining: 5.09s\n", + "730:\tlearn: 14498596.7397089\ttotal: 13.8s\tremaining: 5.07s\n", + "731:\tlearn: 14492352.8619711\ttotal: 13.8s\tremaining: 5.05s\n", + "732:\tlearn: 14492020.3137913\ttotal: 13.8s\tremaining: 5.03s\n", + "733:\tlearn: 14491666.2108348\ttotal: 13.8s\tremaining: 5.01s\n", + "734:\tlearn: 14480978.4686218\ttotal: 13.9s\tremaining: 5s\n", + "735:\tlearn: 14479030.1594746\ttotal: 13.9s\tremaining: 4.98s\n", + "736:\tlearn: 14478950.2204465\ttotal: 13.9s\tremaining: 4.96s\n", + "737:\tlearn: 14473787.5987907\ttotal: 13.9s\tremaining: 4.94s\n", + "738:\tlearn: 14472472.5216995\ttotal: 13.9s\tremaining: 4.92s\n", + "739:\tlearn: 14468027.3906271\ttotal: 13.9s\tremaining: 4.9s\n", + "740:\tlearn: 14463046.5031942\ttotal: 14s\tremaining: 4.88s\n", + "741:\tlearn: 14462567.8451495\ttotal: 14s\tremaining: 4.86s\n", + "742:\tlearn: 14461662.1640772\ttotal: 14s\tremaining: 4.84s\n", + "743:\tlearn: 14461412.7787956\ttotal: 14s\tremaining: 4.82s\n", + "744:\tlearn: 14457080.3104863\ttotal: 14s\tremaining: 4.8s\n", + "745:\tlearn: 14455816.8580813\ttotal: 14s\tremaining: 4.78s\n", + "746:\tlearn: 14451274.2659032\ttotal: 14.1s\tremaining: 4.76s\n", + "747:\tlearn: 14450022.1655873\ttotal: 14.1s\tremaining: 4.74s\n", + "748:\tlearn: 14449711.3365770\ttotal: 14.1s\tremaining: 4.72s\n", + "749:\tlearn: 14449156.8766032\ttotal: 14.1s\tremaining: 4.7s\n", + "750:\tlearn: 14448991.1128152\ttotal: 14.1s\tremaining: 4.68s\n", + "751:\tlearn: 14444878.7492399\ttotal: 14.1s\tremaining: 4.66s\n", + "752:\tlearn: 14444785.6319899\ttotal: 14.1s\tremaining: 4.64s\n", + "753:\tlearn: 14444523.1195537\ttotal: 14.2s\tremaining: 4.62s\n", + "754:\tlearn: 14443980.9051959\ttotal: 14.2s\tremaining: 4.6s\n", + "755:\tlearn: 14442449.8568196\ttotal: 14.2s\tremaining: 4.58s\n", + "756:\tlearn: 14442022.7067717\ttotal: 14.2s\tremaining: 4.56s\n", + "757:\tlearn: 14441891.8251215\ttotal: 14.2s\tremaining: 4.54s\n", + "758:\tlearn: 14434173.5480419\ttotal: 14.2s\tremaining: 4.52s\n", + "759:\tlearn: 14431966.3321856\ttotal: 14.3s\tremaining: 4.5s\n", + "760:\tlearn: 14421158.4055385\ttotal: 14.3s\tremaining: 4.49s\n", + "761:\tlearn: 14420380.5526392\ttotal: 14.3s\tremaining: 4.47s\n", + "762:\tlearn: 14418464.4429589\ttotal: 14.3s\tremaining: 4.45s\n", + "763:\tlearn: 14414445.0255361\ttotal: 14.3s\tremaining: 4.43s\n", + "764:\tlearn: 14413829.3708310\ttotal: 14.3s\tremaining: 4.41s\n", + "765:\tlearn: 14413694.8356085\ttotal: 14.4s\tremaining: 4.39s\n", + "766:\tlearn: 14413259.2614315\ttotal: 14.4s\tremaining: 4.37s\n", + "767:\tlearn: 14413210.9305813\ttotal: 14.4s\tremaining: 4.35s\n", + "768:\tlearn: 14412451.5241734\ttotal: 14.4s\tremaining: 4.33s\n", + "769:\tlearn: 14411887.7754672\ttotal: 14.4s\tremaining: 4.31s\n", + "770:\tlearn: 14407291.5589952\ttotal: 14.5s\tremaining: 4.29s\n", + "771:\tlearn: 14401850.3194431\ttotal: 14.5s\tremaining: 4.27s\n", + "772:\tlearn: 14401685.0712261\ttotal: 14.5s\tremaining: 4.25s\n", + "773:\tlearn: 14399916.3391059\ttotal: 14.5s\tremaining: 4.23s\n", + "774:\tlearn: 14395953.2355612\ttotal: 14.5s\tremaining: 4.21s\n", + "775:\tlearn: 14395882.8636725\ttotal: 14.5s\tremaining: 4.2s\n", + "776:\tlearn: 14388546.8521669\ttotal: 14.6s\tremaining: 4.18s\n", + "777:\tlearn: 14386149.3990942\ttotal: 14.6s\tremaining: 4.16s\n", + "778:\tlearn: 14385003.5881008\ttotal: 14.6s\tremaining: 4.14s\n", + "779:\tlearn: 14383835.8605113\ttotal: 14.6s\tremaining: 4.12s\n", + "780:\tlearn: 14380021.4648146\ttotal: 14.6s\tremaining: 4.1s\n", + "781:\tlearn: 14379157.4604392\ttotal: 14.6s\tremaining: 4.08s\n", + "782:\tlearn: 14375259.3113908\ttotal: 14.7s\tremaining: 4.06s\n", + "783:\tlearn: 14373440.9548670\ttotal: 14.7s\tremaining: 4.04s\n", + "784:\tlearn: 14369819.6596951\ttotal: 14.7s\tremaining: 4.02s\n", + "785:\tlearn: 14369482.1310883\ttotal: 14.7s\tremaining: 4s\n", + "786:\tlearn: 14369186.9789937\ttotal: 14.7s\tremaining: 3.98s\n", + "787:\tlearn: 14368930.0521956\ttotal: 14.7s\tremaining: 3.96s\n", + "788:\tlearn: 14360771.3978411\ttotal: 14.8s\tremaining: 3.95s\n", + "789:\tlearn: 14360698.7609296\ttotal: 14.8s\tremaining: 3.93s\n", + "790:\tlearn: 14360526.7964010\ttotal: 14.8s\tremaining: 3.91s\n", + "791:\tlearn: 14356907.0861542\ttotal: 14.8s\tremaining: 3.89s\n", + "792:\tlearn: 14355859.4628108\ttotal: 14.8s\tremaining: 3.87s\n", + "793:\tlearn: 14352436.1140971\ttotal: 14.8s\tremaining: 3.85s\n", + "794:\tlearn: 14352269.6156056\ttotal: 14.9s\tremaining: 3.83s\n", + "795:\tlearn: 14352158.3097017\ttotal: 14.9s\tremaining: 3.81s\n", + "796:\tlearn: 14351870.9927012\ttotal: 14.9s\tremaining: 3.79s\n", + "797:\tlearn: 14351656.8250020\ttotal: 14.9s\tremaining: 3.77s\n", + "798:\tlearn: 14351543.1131320\ttotal: 14.9s\tremaining: 3.75s\n", + "799:\tlearn: 14348289.1928420\ttotal: 14.9s\tremaining: 3.73s\n", + "800:\tlearn: 14347506.1242205\ttotal: 15s\tremaining: 3.71s\n", + "801:\tlearn: 14347325.5675423\ttotal: 15s\tremaining: 3.7s\n", + "802:\tlearn: 14346999.1868577\ttotal: 15s\tremaining: 3.68s\n", + "803:\tlearn: 14337132.6989066\ttotal: 15s\tremaining: 3.66s\n", + "804:\tlearn: 14334237.2898098\ttotal: 15s\tremaining: 3.64s\n", + "805:\tlearn: 14327707.1452830\ttotal: 15s\tremaining: 3.62s\n", + "806:\tlearn: 14326745.9149928\ttotal: 15.1s\tremaining: 3.6s\n", + "807:\tlearn: 14326077.7205286\ttotal: 15.1s\tremaining: 3.58s\n", + "808:\tlearn: 14325968.2777152\ttotal: 15.1s\tremaining: 3.56s\n", + "809:\tlearn: 14325282.3589449\ttotal: 15.1s\tremaining: 3.54s\n", + "810:\tlearn: 14325153.8711618\ttotal: 15.1s\tremaining: 3.52s\n", + "811:\tlearn: 14322236.6861694\ttotal: 15.1s\tremaining: 3.51s\n", + "812:\tlearn: 14322002.4573230\ttotal: 15.2s\tremaining: 3.49s\n", + "813:\tlearn: 14321635.4752667\ttotal: 15.2s\tremaining: 3.47s\n", + "814:\tlearn: 14318290.7066198\ttotal: 15.2s\tremaining: 3.45s\n", + "815:\tlearn: 14318236.9649185\ttotal: 15.2s\tremaining: 3.43s\n", + "816:\tlearn: 14317989.8840358\ttotal: 15.2s\tremaining: 3.41s\n", + "817:\tlearn: 14315936.4882069\ttotal: 15.2s\tremaining: 3.39s\n", + "818:\tlearn: 14315843.3014233\ttotal: 15.3s\tremaining: 3.37s\n", + "819:\tlearn: 14315758.7406470\ttotal: 15.3s\tremaining: 3.35s\n", + "820:\tlearn: 14315554.5036181\ttotal: 15.3s\tremaining: 3.33s\n", + "821:\tlearn: 14315435.5805586\ttotal: 15.3s\tremaining: 3.31s\n", + "822:\tlearn: 14315329.6866876\ttotal: 15.3s\tremaining: 3.29s\n", + "823:\tlearn: 14308498.9925178\ttotal: 15.3s\tremaining: 3.28s\n", + "824:\tlearn: 14296422.6750795\ttotal: 15.4s\tremaining: 3.26s\n", + "825:\tlearn: 14294363.8932914\ttotal: 15.4s\tremaining: 3.24s\n", + "826:\tlearn: 14280774.5940710\ttotal: 15.4s\tremaining: 3.22s\n", + "827:\tlearn: 14280567.1863461\ttotal: 15.4s\tremaining: 3.2s\n", + "828:\tlearn: 14276167.6246580\ttotal: 15.4s\tremaining: 3.18s\n", + "829:\tlearn: 14273133.7845702\ttotal: 15.4s\tremaining: 3.16s\n", + "830:\tlearn: 14272060.1648094\ttotal: 15.5s\tremaining: 3.15s\n", + "831:\tlearn: 14257262.4559377\ttotal: 15.5s\tremaining: 3.13s\n", + "832:\tlearn: 14256257.7015837\ttotal: 15.5s\tremaining: 3.11s\n", + "833:\tlearn: 14249758.7552648\ttotal: 15.5s\tremaining: 3.09s\n", + "834:\tlearn: 14245611.9143671\ttotal: 15.5s\tremaining: 3.07s\n", + "835:\tlearn: 14245098.2241487\ttotal: 15.6s\tremaining: 3.05s\n", + "836:\tlearn: 14240734.3798939\ttotal: 15.6s\tremaining: 3.03s\n", + "837:\tlearn: 14238120.3063837\ttotal: 15.6s\tremaining: 3.02s\n", + "838:\tlearn: 14237924.0213227\ttotal: 15.6s\tremaining: 3s\n", + "839:\tlearn: 14237790.8302519\ttotal: 15.6s\tremaining: 2.98s\n", + "840:\tlearn: 14235942.8856731\ttotal: 15.6s\tremaining: 2.96s\n", + "841:\tlearn: 14226526.4945080\ttotal: 15.7s\tremaining: 2.94s\n", + "842:\tlearn: 14222342.2375670\ttotal: 15.7s\tremaining: 2.92s\n", + "843:\tlearn: 14221989.7829792\ttotal: 15.7s\tremaining: 2.9s\n", + "844:\tlearn: 14221776.2390137\ttotal: 15.7s\tremaining: 2.88s\n", + "845:\tlearn: 14217665.9042702\ttotal: 15.7s\tremaining: 2.86s\n", + "846:\tlearn: 14216177.5775088\ttotal: 15.8s\tremaining: 2.85s\n", + "847:\tlearn: 14216118.2488163\ttotal: 15.8s\tremaining: 2.83s\n", + "848:\tlearn: 14215942.4545638\ttotal: 15.8s\tremaining: 2.81s\n", + "849:\tlearn: 14215596.9497439\ttotal: 15.8s\tremaining: 2.79s\n", + "850:\tlearn: 14215079.5388703\ttotal: 15.8s\tremaining: 2.77s\n", + "851:\tlearn: 14214985.8639517\ttotal: 15.8s\tremaining: 2.75s\n", + "852:\tlearn: 14211015.9062503\ttotal: 15.9s\tremaining: 2.73s\n", + "853:\tlearn: 14208109.7579081\ttotal: 15.9s\tremaining: 2.71s\n", + "854:\tlearn: 14207069.5540754\ttotal: 15.9s\tremaining: 2.69s\n", + "855:\tlearn: 14204056.3353921\ttotal: 15.9s\tremaining: 2.67s\n", + "856:\tlearn: 14199737.3760866\ttotal: 15.9s\tremaining: 2.66s\n", + "857:\tlearn: 14193953.9975060\ttotal: 15.9s\tremaining: 2.64s\n", + "858:\tlearn: 14193531.3710520\ttotal: 16s\tremaining: 2.62s\n", + "859:\tlearn: 14184582.7017658\ttotal: 16s\tremaining: 2.6s\n", + "860:\tlearn: 14181360.5909759\ttotal: 16s\tremaining: 2.58s\n", + "861:\tlearn: 14178407.2830794\ttotal: 16s\tremaining: 2.56s\n", + "862:\tlearn: 14169917.4406672\ttotal: 16s\tremaining: 2.54s\n", + "863:\tlearn: 14161867.3578748\ttotal: 16s\tremaining: 2.53s\n", + "864:\tlearn: 14161611.6299398\ttotal: 16.1s\tremaining: 2.51s\n", + "865:\tlearn: 14149098.2506612\ttotal: 16.1s\tremaining: 2.49s\n", + "866:\tlearn: 14149008.2568752\ttotal: 16.1s\tremaining: 2.47s\n", + "867:\tlearn: 14147575.3003422\ttotal: 16.1s\tremaining: 2.45s\n", + "868:\tlearn: 14141370.5683044\ttotal: 16.1s\tremaining: 2.43s\n", + "869:\tlearn: 14141139.9686371\ttotal: 16.2s\tremaining: 2.41s\n", + "870:\tlearn: 14140906.6108388\ttotal: 16.2s\tremaining: 2.39s\n", + "871:\tlearn: 14136972.2399677\ttotal: 16.2s\tremaining: 2.38s\n", + "872:\tlearn: 14132863.3886077\ttotal: 16.2s\tremaining: 2.36s\n", + "873:\tlearn: 14130176.4823179\ttotal: 16.2s\tremaining: 2.34s\n", + "874:\tlearn: 14130009.1218439\ttotal: 16.2s\tremaining: 2.32s\n", + "875:\tlearn: 14129938.0340251\ttotal: 16.3s\tremaining: 2.3s\n", + "876:\tlearn: 14129730.1355073\ttotal: 16.3s\tremaining: 2.28s\n", + "877:\tlearn: 14129144.1514023\ttotal: 16.3s\tremaining: 2.26s\n", + "878:\tlearn: 14128257.5078373\ttotal: 16.3s\tremaining: 2.24s\n", + "879:\tlearn: 14123974.5623694\ttotal: 16.3s\tremaining: 2.23s\n", + "880:\tlearn: 14117573.9417940\ttotal: 16.3s\tremaining: 2.21s\n", + "881:\tlearn: 14117199.4807352\ttotal: 16.4s\tremaining: 2.19s\n", + "882:\tlearn: 14116866.5963846\ttotal: 16.4s\tremaining: 2.17s\n", + "883:\tlearn: 14114235.9424007\ttotal: 16.4s\tremaining: 2.15s\n", + "884:\tlearn: 14114037.4869654\ttotal: 16.4s\tremaining: 2.13s\n", + "885:\tlearn: 14113795.5264062\ttotal: 16.4s\tremaining: 2.11s\n", + "886:\tlearn: 14103701.3697817\ttotal: 16.4s\tremaining: 2.09s\n", + "887:\tlearn: 14102545.5321288\ttotal: 16.5s\tremaining: 2.08s\n", + "888:\tlearn: 14096643.9605945\ttotal: 16.5s\tremaining: 2.06s\n", + "889:\tlearn: 14093637.9560237\ttotal: 16.5s\tremaining: 2.04s\n", + "890:\tlearn: 14093398.8314629\ttotal: 16.5s\tremaining: 2.02s\n", + "891:\tlearn: 14091893.0920055\ttotal: 16.5s\tremaining: 2s\n", + "892:\tlearn: 14089051.2497645\ttotal: 16.5s\tremaining: 1.98s\n", + "893:\tlearn: 14088275.7091360\ttotal: 16.6s\tremaining: 1.96s\n", + "894:\tlearn: 14088065.0293484\ttotal: 16.6s\tremaining: 1.94s\n", + "895:\tlearn: 14085363.8111205\ttotal: 16.6s\tremaining: 1.93s\n", + "896:\tlearn: 14070847.0790061\ttotal: 16.6s\tremaining: 1.91s\n", + "897:\tlearn: 14066938.0257631\ttotal: 16.6s\tremaining: 1.89s\n", + "898:\tlearn: 14061332.0592375\ttotal: 16.6s\tremaining: 1.87s\n", + "899:\tlearn: 14061275.8140325\ttotal: 16.7s\tremaining: 1.85s\n", + "900:\tlearn: 14049395.4083470\ttotal: 16.7s\tremaining: 1.83s\n", + "901:\tlearn: 14049223.6858779\ttotal: 16.7s\tremaining: 1.81s\n", + "902:\tlearn: 14049090.7140090\ttotal: 16.7s\tremaining: 1.79s\n", + "903:\tlearn: 14038509.6975127\ttotal: 16.7s\tremaining: 1.78s\n", + "904:\tlearn: 14037966.9340313\ttotal: 16.7s\tremaining: 1.76s\n", + "905:\tlearn: 14037929.7816334\ttotal: 16.8s\tremaining: 1.74s\n", + "906:\tlearn: 14035361.6403404\ttotal: 16.8s\tremaining: 1.72s\n", + "907:\tlearn: 14025931.6101841\ttotal: 16.8s\tremaining: 1.7s\n", + "908:\tlearn: 14020604.5578164\ttotal: 16.8s\tremaining: 1.68s\n", + "909:\tlearn: 14018039.0515475\ttotal: 16.8s\tremaining: 1.67s\n", + "910:\tlearn: 14016909.4019073\ttotal: 16.9s\tremaining: 1.65s\n", + "911:\tlearn: 14012709.8837074\ttotal: 16.9s\tremaining: 1.63s\n", + "912:\tlearn: 14012638.0469528\ttotal: 16.9s\tremaining: 1.61s\n", + "913:\tlearn: 14008600.0892461\ttotal: 16.9s\tremaining: 1.59s\n", + "914:\tlearn: 14007920.3022894\ttotal: 17s\tremaining: 1.57s\n", + "915:\tlearn: 14007639.0218815\ttotal: 17s\tremaining: 1.56s\n", + "916:\tlearn: 14000153.4424950\ttotal: 17s\tremaining: 1.54s\n", + "917:\tlearn: 13993056.1891854\ttotal: 17s\tremaining: 1.52s\n", + "918:\tlearn: 13989377.5685025\ttotal: 17.1s\tremaining: 1.5s\n", + "919:\tlearn: 13989042.4889836\ttotal: 17.1s\tremaining: 1.49s\n", + "920:\tlearn: 13988995.2411999\ttotal: 17.1s\tremaining: 1.47s\n", + "921:\tlearn: 13979426.4500026\ttotal: 17.1s\tremaining: 1.45s\n", + "922:\tlearn: 13979364.5169235\ttotal: 17.1s\tremaining: 1.43s\n", + "923:\tlearn: 13978421.6270684\ttotal: 17.2s\tremaining: 1.41s\n", + "924:\tlearn: 13978232.1417556\ttotal: 17.2s\tremaining: 1.39s\n", + "925:\tlearn: 13977413.1833242\ttotal: 17.2s\tremaining: 1.38s\n", + "926:\tlearn: 13974984.8789985\ttotal: 17.2s\tremaining: 1.36s\n", + "927:\tlearn: 13973117.9034490\ttotal: 17.3s\tremaining: 1.34s\n", + "928:\tlearn: 13973040.8894481\ttotal: 17.3s\tremaining: 1.32s\n", + "929:\tlearn: 13969309.5963784\ttotal: 17.3s\tremaining: 1.3s\n", + "930:\tlearn: 13964934.5197786\ttotal: 17.3s\tremaining: 1.28s\n", + "931:\tlearn: 13964075.8787535\ttotal: 17.3s\tremaining: 1.26s\n", + "932:\tlearn: 13962384.7500669\ttotal: 17.4s\tremaining: 1.25s\n", + "933:\tlearn: 13962116.7297324\ttotal: 17.4s\tremaining: 1.23s\n", + "934:\tlearn: 13961984.2654879\ttotal: 17.4s\tremaining: 1.21s\n", + "935:\tlearn: 13961283.4113454\ttotal: 17.4s\tremaining: 1.19s\n", + "936:\tlearn: 13961119.7725702\ttotal: 17.4s\tremaining: 1.17s\n", + "937:\tlearn: 13957581.9271590\ttotal: 17.5s\tremaining: 1.15s\n", + "938:\tlearn: 13956711.5699546\ttotal: 17.5s\tremaining: 1.14s\n", + "939:\tlearn: 13956195.7669965\ttotal: 17.5s\tremaining: 1.12s\n", + "940:\tlearn: 13956019.7725634\ttotal: 17.5s\tremaining: 1.1s\n", + "941:\tlearn: 13953817.7128654\ttotal: 17.5s\tremaining: 1.08s\n", + "942:\tlearn: 13951728.3708942\ttotal: 17.5s\tremaining: 1.06s\n", + "943:\tlearn: 13948374.3380936\ttotal: 17.6s\tremaining: 1.04s\n", + "944:\tlearn: 13946396.1172757\ttotal: 17.6s\tremaining: 1.02s\n", + "945:\tlearn: 13946318.2181768\ttotal: 17.6s\tremaining: 1s\n", + "946:\tlearn: 13946223.0768320\ttotal: 17.6s\tremaining: 986ms\n", + "947:\tlearn: 13943040.9225619\ttotal: 17.6s\tremaining: 968ms\n", + "948:\tlearn: 13942723.6463439\ttotal: 17.7s\tremaining: 949ms\n", + "949:\tlearn: 13940873.9346942\ttotal: 17.7s\tremaining: 930ms\n", + "950:\tlearn: 13940657.5960414\ttotal: 17.7s\tremaining: 912ms\n", + "951:\tlearn: 13937643.8070567\ttotal: 17.7s\tremaining: 893ms\n", + "952:\tlearn: 13927861.6944498\ttotal: 17.7s\tremaining: 874ms\n", + "953:\tlearn: 13927713.1158209\ttotal: 17.7s\tremaining: 856ms\n", + "954:\tlearn: 13919785.6047620\ttotal: 17.8s\tremaining: 837ms\n", + "955:\tlearn: 13919738.2222830\ttotal: 17.8s\tremaining: 818ms\n", + "956:\tlearn: 13914879.1616648\ttotal: 17.8s\tremaining: 800ms\n", + "957:\tlearn: 13913904.0967614\ttotal: 17.8s\tremaining: 782ms\n", + "958:\tlearn: 13913369.8497656\ttotal: 17.9s\tremaining: 763ms\n", + "959:\tlearn: 13913210.3609993\ttotal: 17.9s\tremaining: 745ms\n", + "960:\tlearn: 13912875.9812733\ttotal: 17.9s\tremaining: 726ms\n", + "961:\tlearn: 13910617.8439350\ttotal: 17.9s\tremaining: 708ms\n", + "962:\tlearn: 13908740.7255579\ttotal: 17.9s\tremaining: 690ms\n", + "963:\tlearn: 13908173.0945609\ttotal: 18s\tremaining: 672ms\n", + "964:\tlearn: 13906394.6235472\ttotal: 18s\tremaining: 653ms\n", + "965:\tlearn: 13904710.4079211\ttotal: 18s\tremaining: 634ms\n", + "966:\tlearn: 13901845.4980302\ttotal: 18s\tremaining: 616ms\n", + "967:\tlearn: 13901304.7542491\ttotal: 18.1s\tremaining: 597ms\n", + "968:\tlearn: 13901227.5036081\ttotal: 18.1s\tremaining: 578ms\n", + "969:\tlearn: 13900564.3429757\ttotal: 18.1s\tremaining: 560ms\n", + "970:\tlearn: 13900431.9676622\ttotal: 18.1s\tremaining: 541ms\n", + "971:\tlearn: 13897718.5750538\ttotal: 18.1s\tremaining: 522ms\n", + "972:\tlearn: 13892434.2051176\ttotal: 18.2s\tremaining: 504ms\n", + "973:\tlearn: 13892171.4126968\ttotal: 18.2s\tremaining: 485ms\n", + "974:\tlearn: 13891466.0625935\ttotal: 18.2s\tremaining: 467ms\n", + "975:\tlearn: 13886354.8726715\ttotal: 18.2s\tremaining: 448ms\n", + "976:\tlearn: 13872366.2643593\ttotal: 18.2s\tremaining: 430ms\n", + "977:\tlearn: 13871047.0266210\ttotal: 18.3s\tremaining: 411ms\n", + "978:\tlearn: 13870327.9449845\ttotal: 18.3s\tremaining: 392ms\n", + "979:\tlearn: 13869912.2574551\ttotal: 18.3s\tremaining: 374ms\n", + "980:\tlearn: 13865882.9385922\ttotal: 18.3s\tremaining: 355ms\n", + "981:\tlearn: 13864637.2130234\ttotal: 18.4s\tremaining: 336ms\n", + "982:\tlearn: 13863950.1729492\ttotal: 18.4s\tremaining: 318ms\n", + "983:\tlearn: 13860344.3288697\ttotal: 18.4s\tremaining: 299ms\n", + "984:\tlearn: 13860136.8254682\ttotal: 18.4s\tremaining: 281ms\n", + "985:\tlearn: 13859522.8057507\ttotal: 18.4s\tremaining: 262ms\n", + "986:\tlearn: 13858306.6249102\ttotal: 18.5s\tremaining: 243ms\n", + "987:\tlearn: 13856860.1094265\ttotal: 18.5s\tremaining: 224ms\n", + "988:\tlearn: 13855323.9194082\ttotal: 18.5s\tremaining: 206ms\n", + "989:\tlearn: 13848524.5771249\ttotal: 18.5s\tremaining: 187ms\n", + "990:\tlearn: 13848229.3075576\ttotal: 18.5s\tremaining: 168ms\n", + "991:\tlearn: 13844412.8445986\ttotal: 18.6s\tremaining: 150ms\n", + "992:\tlearn: 13844333.0927820\ttotal: 18.6s\tremaining: 131ms\n", + "993:\tlearn: 13840812.1167177\ttotal: 18.6s\tremaining: 112ms\n", + "994:\tlearn: 13839796.5731119\ttotal: 18.6s\tremaining: 93.5ms\n", + "995:\tlearn: 13839035.4643377\ttotal: 18.6s\tremaining: 74.8ms\n", + "996:\tlearn: 13838769.7099457\ttotal: 18.6s\tremaining: 56.1ms\n", + "997:\tlearn: 13838205.6345566\ttotal: 18.7s\tremaining: 37.4ms\n", + "998:\tlearn: 13837777.8741797\ttotal: 18.7s\tremaining: 18.7ms\n", + "999:\tlearn: 13837569.4194450\ttotal: 18.7s\tremaining: 0us\n" + ] + } + ], + "source": [ + "pipeline_sklearn = Pipeline(steps=[\n", + " ('transform', preprocessor_sklearn),\n", + " ('model', regressor)\n", + "])\n", + "\n", + "model_sklearn = pipeline_sklearn.fit(X_train, y_train)" + ] + }, + { + "cell_type": "code", + "execution_count": 62, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
Pipeline(steps=[('transform',\n",
+       "                 ColumnTransformer(transformers=[('num', StandardScaler(),\n",
+       "                                                  ['geo_lat', 'geo_lon',\n",
+       "                                                   'level', 'levels', 'rooms',\n",
+       "                                                   'area', 'kitchen_area']),\n",
+       "                                                 ('cat',\n",
+       "                                                  OrdinalEncoder(handle_unknown='use_encoded_value',\n",
+       "                                                                 unknown_value=99999999),\n",
+       "                                                  ['region', 'building_type',\n",
+       "                                                   'object_type']),\n",
+       "                                                 ('quantile',\n",
+       "                                                  QuantileTransformer(),\n",
+       "                                                  ['geo_lat', 'geo_lon',\n",
+       "                                                   'level', 'levels', 'rooms',\n",
+       "                                                   'area', 'kitchen_area']),\n",
+       "                                                 ('poly',\n",
+       "                                                  Pipeline(steps=[('poly',\n",
+       "                                                                   PolynomialFeatures()),\n",
+       "                                                                  ('scale',\n",
+       "                                                                   StandardScaler())]),\n",
+       "                                                  ['area', 'kitchen_area']),\n",
+       "                                                 ('spline',\n",
+       "                                                  SplineTransformer(n_knots=3),\n",
+       "                                                  ['area'])])),\n",
+       "                ('model',\n",
+       "                 <catboost.core.CatBoostRegressor object at 0x7448bd575f60>)])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ], + "text/plain": [ + "Pipeline(steps=[('transform',\n", + " ColumnTransformer(transformers=[('num', StandardScaler(),\n", + " ['geo_lat', 'geo_lon',\n", + " 'level', 'levels', 'rooms',\n", + " 'area', 'kitchen_area']),\n", + " ('cat',\n", + " OrdinalEncoder(handle_unknown='use_encoded_value',\n", + " unknown_value=99999999),\n", + " ['region', 'building_type',\n", + " 'object_type']),\n", + " ('quantile',\n", + " QuantileTransformer(),\n", + " ['geo_lat', 'geo_lon',\n", + " 'level', 'levels', 'rooms',\n", + " 'area', 'kitchen_area']),\n", + " ('poly',\n", + " Pipeline(steps=[('poly',\n", + " PolynomialFeatures()),\n", + " ('scale',\n", + " StandardScaler())]),\n", + " ['area', 'kitchen_area']),\n", + " ('spline',\n", + " SplineTransformer(n_knots=3),\n", + " ['area'])])),\n", + " ('model',\n", + " )])" + ] + }, + "execution_count": 62, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "model_sklearn" + ] + }, + { + "cell_type": "code", + "execution_count": 63, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "{'mae': 1371773.3280469999,\n", + " 'mape': 1.651780914989046e+18,\n", + " 'mse': 271999053666354.03}" + ] + }, + "execution_count": 63, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "predictions = model_sklearn.predict(X_test) \n", + "metrics = {}\n", + "metrics[\"mae\"] = mean_absolute_error(y_test, predictions) \n", + "metrics[\"mape\"] = mean_absolute_percentage_error(y_test, predictions)\n", + "metrics[\"mse\"] = mean_squared_error(y_test, predictions)\n", + "\n", + "metrics" + ] + }, + { + "cell_type": "code", + "execution_count": 64, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024/10/10 14:29:09 INFO mlflow.tracking._tracking_service.client: 🏃 View run fe_sklearn at: http://127.0.0.1:5000/#/experiments/1/runs/aa74ed5ed8aa48458ae26929bc237338.\n", + "2024/10/10 14:29:09 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" + ] + } + ], + "source": [ + "experiment_id = mlflow.get_experiment_by_name(EXPERIMENT_NAME).experiment_id\n", + "RUN_NAME = 'fe_sklearn'\n", + "\n", + "with mlflow.start_run(run_name=RUN_NAME, experiment_id=experiment_id) as run:\n", + " # получаем уникальный идентификатор запуска эксперимента\n", + " run_id = run.info.run_id \n", + " mlflow.sklearn.log_model(model_sklearn, \n", + " artifact_path=\"models\",\n", + " signature=signature,\n", + " input_example=input_example,\n", + " pip_requirements=req_file\n", + " )\n", + " mlflow.log_metrics(metrics)\n", + " mlflow.log_artifact(art)\n", + " mlflow.log_params(model_sklearn.get_params())\n", + "\n", + "run = mlflow.get_run(run_id) \n", + "assert (run.info.status =='FINISHED')" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Autofeat" + ] + }, + { + "cell_type": "code", + "execution_count": 65, + "metadata": {}, + "outputs": [], + "source": [ + "from autofeat import AutoFeatRegressor\n", + "transformations = [\"1/\", \"exp\", \"log\", \"abs\", \"sqrt\", \"^2\", \"^3\", \"1+\", \"1-\", \"sin\", \"cos\", \"exp-\", \"2^\"] " + ] + }, + { + "cell_type": "code", + "execution_count": 66, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-10-10 14:34:43,949 INFO: [AutoFeat] The 2 step feature engineering process could generate up to 105 features.\n", + "2024-10-10 14:34:43,950 INFO: [AutoFeat] With 410775 data points this new feature matrix would use about 0.17 gb of space.\n", + "2024-10-10 14:34:43,986 INFO: [feateng] Step 1: transformation of original features\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[feateng] 0/ 7 features transformed\r" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-10-10 14:34:45,030 INFO: [feateng] Generated 12 transformed features from 7 original features - done.\n", + "2024-10-10 14:34:45,061 INFO: [feateng] Step 2: first combination of features\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[feateng] 100/ 171 feature tuples combined\r" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-10-10 14:34:47,541 INFO: [feateng] Generated 171 feature combinations from 171 original feature tuples - done.\n", + "2024-10-10 14:34:48,231 INFO: [feateng] Generated altogether 183 new features in 2 steps\n", + "2024-10-10 14:34:48,231 INFO: [feateng] Removing correlated features, as well as additions at the highest level\n", + "2024-10-10 14:34:49,298 INFO: [feateng] Generated a total of 98 additional features\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[featsel] Scaling data..." + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-10-10 14:34:49,789 INFO: [featsel] Feature selection run 1/5\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "done.\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-10-10 14:35:35,384 INFO: [featsel] Feature selection run 2/5\n", + "2024-10-10 14:36:01,168 INFO: [featsel] Feature selection run 3/5\n", + "2024-10-10 14:36:22,873 INFO: [featsel] Feature selection run 4/5\n", + "2024-10-10 14:36:47,567 INFO: [featsel] Feature selection run 5/5\n", + "2024-10-10 14:37:16,308 INFO: [featsel] 53 features after 5 feature selection runs\n", + "/home/andrey/work/institute/MLE/assets/mlflow/.venv_ml2/lib/python3.10/site-packages/autofeat/featsel.py:270: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " if np.max(np.abs(correlations[c].ravel()[:i])) < 0.9:\n", + "2024-10-10 14:37:21,842 INFO: [featsel] 35 features after correlation filtering\n", + "2024-10-10 14:37:28,155 INFO: [featsel] 24 features after noise filtering\n", + "2024-10-10 14:37:28,214 INFO: [AutoFeat] Computing 18 new features.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[AutoFeat] 17/ 18 new features\r" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-10-10 14:37:31,290 INFO: [AutoFeat] 18/ 18 new features ...done.\n", + "2024-10-10 14:37:31,430 INFO: [AutoFeat] Final dataframe with 28 feature columns (18 new).\n", + "2024-10-10 14:37:31,431 INFO: [AutoFeat] Training final regression model.\n", + "2024-10-10 14:37:32,884 INFO: [AutoFeat] Trained model: largest coefficients:\n", + "2024-10-10 14:37:32,885 INFO: 1397647.486074062\n", + "2024-10-10 14:37:32,885 INFO: -444567.439631 * sqrt(area)*log(geo_lon)\n", + "2024-10-10 14:37:32,886 INFO: -229561.254963 * kitchen_area\n", + "2024-10-10 14:37:32,886 INFO: 228699.320633 * geo_lon\n", + "2024-10-10 14:37:32,887 INFO: -77449.633774 * sqrt(geo_lon)*sqrt(kitchen_area)\n", + "2024-10-10 14:37:32,887 INFO: 68865.207107 * sqrt(area)*log(levels)\n", + "2024-10-10 14:37:32,887 INFO: -58234.749991 * geo_lat*log(geo_lon)\n", + "2024-10-10 14:37:32,888 INFO: 53857.792425 * sqrt(area)*kitchen_area\n", + "2024-10-10 14:37:32,888 INFO: 48541.235065 * sqrt(area)*geo_lat\n", + "2024-10-10 14:37:32,888 INFO: -27253.343264 * geo_lon*rooms\n", + "2024-10-10 14:37:32,888 INFO: 27008.036497 * area*rooms\n", + "2024-10-10 14:37:32,889 INFO: -18811.054781 * object_type\n", + "2024-10-10 14:37:32,889 INFO: 15811.284677 * sqrt(area)*log(level)\n", + "2024-10-10 14:37:32,890 INFO: 5197.560812 * geo_lon*log(levels)\n", + "2024-10-10 14:37:32,890 INFO: 3144.707961 * geo_lat*log(kitchen_area)\n", + "2024-10-10 14:37:32,891 INFO: -1348.579984 * area*geo_lon\n", + "2024-10-10 14:37:32,891 INFO: -674.637306 * area*kitchen_area\n", + "2024-10-10 14:37:32,891 INFO: -445.508078 * region\n", + "2024-10-10 14:37:32,892 INFO: 63.854761 * area**(3/2)\n", + "2024-10-10 14:37:32,953 INFO: [AutoFeat] Final score: 0.0394\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
geo_latgeo_lonregionbuilding_typelevellevelsroomsareakitchen_areaobject_type...geo_lat*log(kitchen_area)sqrt(geo_lon)*sqrt(level)sqrt(geo_lon)*sqrt(kitchen_area)sqrt(area)*log(levels)sqrt(area)*log(geo_lon)sqrt(area)*kitchen_areaarea**(3/2)geo_lon*roomsgeo_lon*log(levels)geo_lat*log(geo_lon)
056.32768643.9280622871.01.08.010.02.056.00008.5000001.0...120.54497618.74632019.32326417.23096928.30603763.608176419.06562787.856125101.148102213.062479
154.85931082.9756249654.01.05.09.02.050.18758.5000001.0...117.40255320.36855726.55734915.56582831.30237260.216666355.543992165.951248182.316081242.398434
256.07216354.2312702722.03.04.09.01.039.00009.00000011.0...123.20313414.72837722.09256513.72166324.93788656.204982243.55492254.231270119.158279223.910594
346.70432738.2736362843.01.05.05.02.048.00009.0000001.0...102.61989413.83358918.55970711.15051325.25164762.353829332.55375576.54727261.599041170.226122
460.93348376.5930942484.02.07.015.02.074.000010.5000001.0...143.27748523.15494928.35890523.29552937.32124890.324415636.572070153.186188207.417943264.360338
..................................................................
41077056.77133960.6125186171.02.09.018.02.051.31259.5703121.0...128.22748623.35621324.08486520.70452629.40167068.554774367.565517121.225037175.192711233.018045
41077155.14861361.3937805282.03.03.05.02.043.00006.0000001.0...98.81305013.57134319.19277710.55379026.99899839.344631281.969857122.78756098.809477227.063854
41077245.09812938.9712182843.00.08.017.01.039.00009.00000011.0...99.09071817.65700318.72808017.69341222.87432556.204982243.55492238.971218110.413775165.186482
41077350.58396936.5818675952.03.03.016.02.098.500027.7968751.0...168.18783310.47595331.88826727.51715735.724540275.876107977.58458773.163734101.426472182.079662
41077459.84457030.4090772661.01.06.012.03.072.00009.0000001.0...131.49196013.50757016.54332821.08513228.97503976.367532610.94025991.22723075.563717204.353716
\n", + "

410775 rows × 28 columns

\n", + "
" + ], + "text/plain": [ + " geo_lat geo_lon region building_type level levels rooms \\\n", + "0 56.327686 43.928062 2871.0 1.0 8.0 10.0 2.0 \n", + "1 54.859310 82.975624 9654.0 1.0 5.0 9.0 2.0 \n", + "2 56.072163 54.231270 2722.0 3.0 4.0 9.0 1.0 \n", + "3 46.704327 38.273636 2843.0 1.0 5.0 5.0 2.0 \n", + "4 60.933483 76.593094 2484.0 2.0 7.0 15.0 2.0 \n", + "... ... ... ... ... ... ... ... \n", + "410770 56.771339 60.612518 6171.0 2.0 9.0 18.0 2.0 \n", + "410771 55.148613 61.393780 5282.0 3.0 3.0 5.0 2.0 \n", + "410772 45.098129 38.971218 2843.0 0.0 8.0 17.0 1.0 \n", + "410773 50.583969 36.581867 5952.0 3.0 3.0 16.0 2.0 \n", + "410774 59.844570 30.409077 2661.0 1.0 6.0 12.0 3.0 \n", + "\n", + " area kitchen_area object_type ... geo_lat*log(kitchen_area) \\\n", + "0 56.0000 8.500000 1.0 ... 120.544976 \n", + "1 50.1875 8.500000 1.0 ... 117.402553 \n", + "2 39.0000 9.000000 11.0 ... 123.203134 \n", + "3 48.0000 9.000000 1.0 ... 102.619894 \n", + "4 74.0000 10.500000 1.0 ... 143.277485 \n", + "... ... ... ... ... ... \n", + "410770 51.3125 9.570312 1.0 ... 128.227486 \n", + "410771 43.0000 6.000000 1.0 ... 98.813050 \n", + "410772 39.0000 9.000000 11.0 ... 99.090718 \n", + "410773 98.5000 27.796875 1.0 ... 168.187833 \n", + "410774 72.0000 9.000000 1.0 ... 131.491960 \n", + "\n", + " sqrt(geo_lon)*sqrt(level) sqrt(geo_lon)*sqrt(kitchen_area) \\\n", + "0 18.746320 19.323264 \n", + "1 20.368557 26.557349 \n", + "2 14.728377 22.092565 \n", + "3 13.833589 18.559707 \n", + "4 23.154949 28.358905 \n", + "... ... ... \n", + "410770 23.356213 24.084865 \n", + "410771 13.571343 19.192777 \n", + "410772 17.657003 18.728080 \n", + "410773 10.475953 31.888267 \n", + "410774 13.507570 16.543328 \n", + "\n", + " sqrt(area)*log(levels) sqrt(area)*log(geo_lon) \\\n", + "0 17.230969 28.306037 \n", + "1 15.565828 31.302372 \n", + "2 13.721663 24.937886 \n", + "3 11.150513 25.251647 \n", + "4 23.295529 37.321248 \n", + "... ... ... \n", + "410770 20.704526 29.401670 \n", + "410771 10.553790 26.998998 \n", + "410772 17.693412 22.874325 \n", + "410773 27.517157 35.724540 \n", + "410774 21.085132 28.975039 \n", + "\n", + " sqrt(area)*kitchen_area area**(3/2) geo_lon*rooms \\\n", + "0 63.608176 419.065627 87.856125 \n", + "1 60.216666 355.543992 165.951248 \n", + "2 56.204982 243.554922 54.231270 \n", + "3 62.353829 332.553755 76.547272 \n", + "4 90.324415 636.572070 153.186188 \n", + "... ... ... ... \n", + "410770 68.554774 367.565517 121.225037 \n", + "410771 39.344631 281.969857 122.787560 \n", + "410772 56.204982 243.554922 38.971218 \n", + "410773 275.876107 977.584587 73.163734 \n", + "410774 76.367532 610.940259 91.227230 \n", + "\n", + " geo_lon*log(levels) geo_lat*log(geo_lon) \n", + "0 101.148102 213.062479 \n", + "1 182.316081 242.398434 \n", + "2 119.158279 223.910594 \n", + "3 61.599041 170.226122 \n", + "4 207.417943 264.360338 \n", + "... ... ... \n", + "410770 175.192711 233.018045 \n", + "410771 98.809477 227.063854 \n", + "410772 110.413775 165.186482 \n", + "410773 101.426472 182.079662 \n", + "410774 75.563717 204.353716 \n", + "\n", + "[410775 rows x 28 columns]" + ] + }, + "execution_count": 66, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "afreg = AutoFeatRegressor(verbose=1, feateng_steps=2, max_gb=8, transformations=[\"log\", \"sqrt\"],feateng_cols=num_features)\n", + "X_train_arf = afreg.fit_transform(X_train,y_train)\n", + "X_train_arf" + ] + }, + { + "cell_type": "code", + "execution_count": 67, + "metadata": {}, + "outputs": [], + "source": [ + "# Создаем обертку, в которой добавляем метод get_feature_names_out() для получения названий признаков\n", + "import numpy as np\n", + "\n", + "class AutoFeatWrapper():\n", + " def __init__(self, feateng_cols, feateng_steps=1, max_gb=16, transformations=[\"1/\", \"exp\", \"log\"], n_jobs=-1, verbose=1):\n", + " self.feateng_cols = feateng_cols\n", + " self.feateng_steps = feateng_steps\n", + " self.max_gb = max_gb\n", + " self.transformations = transformations\n", + " self.n_jobs = n_jobs\n", + " self.afreg = AutoFeatRegressor(feateng_cols=self.feateng_cols,\n", + " feateng_steps=self.feateng_steps,\n", + " max_gb=self.max_gb,\n", + " transformations=self.transformations,\n", + " n_jobs=self.n_jobs)\n", + " \n", + " def fit(self, X, y=None):\n", + " self.afreg.fit(X, y)\n", + " return self\n", + " \n", + " def transform(self, X):\n", + " return self.afreg.transform(X)\n", + " \n", + " def get_feature_names_out(self, input_features=None):\n", + " # Преобразуем данные и возвращаем имена фичей из DataFrame\n", + " transformed_X = self.afreg.transform(pd.DataFrame(np.zeros((1, len(self.feateng_cols))), columns=self.feateng_cols))\n", + " return transformed_X.columns.tolist()" + ] + }, + { + "cell_type": "code", + "execution_count": 69, + "metadata": {}, + "outputs": [], + "source": [ + "afreg_pipeline = Pipeline(steps=[\n", + " ('autofeat', AutoFeatWrapper( feateng_steps=2, max_gb=16, transformations=[\"log\", \"sqrt\"],feateng_cols=num_features)),\n", + " ('scaler', StandardScaler()),\n", + "])" + ] + }, + { + "cell_type": "code", + "execution_count": 70, + "metadata": {}, + "outputs": [], + "source": [ + "preprocessor_afr = ColumnTransformer(\n", + " transformers=[\n", + " ('num', s_scaler, num_features), # преобразования для числовых признаков\n", + " ('cat', l_encoder, cat_features), # преобразования для категориальных признаков\n", + " ('afr', afreg_pipeline, num_features), # преобразования autofeat\n", + " ],\n", + " remainder='drop', # Удаляем столбцы, которые не затронуты преобразованиями\n", + " ) " + ] + }, + { + "cell_type": "code", + "execution_count": 71, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/andrey/work/institute/MLE/assets/mlflow/.venv_ml2/lib/python3.10/site-packages/autofeat/featsel.py:270: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " if np.max(np.abs(correlations[c].ravel()[:i])) < 0.9:\n" + ] + } + ], + "source": [ + "X_train_afr_raw = preprocessor_afr.fit_transform(X_train,y_train)\n", + "X_train_afr = pd.DataFrame(X_train_afr_raw, columns=preprocessor_afr.get_feature_names_out())" + ] + }, + { + "cell_type": "code", + "execution_count": 72, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
num__geo_latnum__geo_lonnum__levelnum__levelsnum__roomsnum__areanum__kitchen_areacat__regioncat__building_typecat__object_typeafr__geo_latafr__geo_lonafr__levelafr__levelsafr__roomsafr__areaafr__kitchen_areaafr__area*roomsafr__area*geo_lonafr__levels*roomsafr__area*kitchen_areaafr__sqrt(area)*geo_latafr__sqrt(area)*log(level)afr__kitchen_area*log(level)afr__sqrt(area)*kitchen_areaafr__geo_lon*log(kitchen_area)afr__sqrt(area)*sqrt(kitchen_area)afr__sqrt(geo_lon)*sqrt(kitchen_area)afr__log(area)afr__rooms*log(level)afr__kitchen_area*roomsafr__kitchen_area*levelsafr__sqrt(geo_lon)*sqrt(level)afr__area**(3/2)afr__geo_lat*log(kitchen_area)afr__geo_lat*log(geo_lon)
00.495902-0.4497420.359235-0.2147890.2534130.063735-0.18628520.01.00.00.495902-0.4497420.359235-0.2147890.2534130.063735-0.1862850.006208-0.1951290.060916-0.1321880.3731510.6880760.044178-0.211335-0.481294-0.153548-0.4908050.3078350.690329-0.132529-0.3528340.323880-0.008748-0.0315290.068167
10.1778061.433673-0.246529-0.3677180.253413-0.114293-0.18628570.01.00.00.1778061.433673-0.246529-0.3677180.253413-0.114293-0.186285-0.0834020.655053-0.054279-0.1693700.0051140.071369-0.173647-0.2527751.191304-0.2672680.6157980.0319070.282625-0.132529-0.4186430.552794-0.056540-0.1438291.129118
...............................................................................................................
410773-0.748366-0.804077-0.6503710.7027880.2534131.3654411.50183352.03.00.0-0.748366-0.804077-0.6503710.7027880.2534131.3654411.5018330.6614270.3751990.7520881.5701631.274445-0.0025210.7455072.3822580.0715992.8288901.4312721.729715-0.1604911.5814362.432437-0.8431500.4114751.671069-1.052343
4107741.257769-1.101815-0.0446080.0910701.1759110.553789-0.14254414.01.00.01.257769-1.101815-0.0446080.0910701.1759110.553789-0.1425440.807887-0.3300700.982478-0.0027421.3389960.635065-0.040302-0.055435-1.0255880.202136-0.9160540.9406241.2179100.311575-0.174762-0.4153590.1356170.359680-0.246790
\n", + "

410775 rows × 36 columns

\n", + "
" + ], + "text/plain": [ + " num__geo_lat num__geo_lon num__level num__levels num__rooms \\\n", + "0 0.495902 -0.449742 0.359235 -0.214789 0.253413 \n", + "1 0.177806 1.433673 -0.246529 -0.367718 0.253413 \n", + "... ... ... ... ... ... \n", + "410773 -0.748366 -0.804077 -0.650371 0.702788 0.253413 \n", + "410774 1.257769 -1.101815 -0.044608 0.091070 1.175911 \n", + "\n", + " num__area num__kitchen_area cat__region cat__building_type \\\n", + "0 0.063735 -0.186285 20.0 1.0 \n", + "1 -0.114293 -0.186285 70.0 1.0 \n", + "... ... ... ... ... \n", + "410773 1.365441 1.501833 52.0 3.0 \n", + "410774 0.553789 -0.142544 14.0 1.0 \n", + "\n", + " cat__object_type afr__geo_lat afr__geo_lon afr__level afr__levels \\\n", + "0 0.0 0.495902 -0.449742 0.359235 -0.214789 \n", + "1 0.0 0.177806 1.433673 -0.246529 -0.367718 \n", + "... ... ... ... ... ... \n", + "410773 0.0 -0.748366 -0.804077 -0.650371 0.702788 \n", + "410774 0.0 1.257769 -1.101815 -0.044608 0.091070 \n", + "\n", + " afr__rooms afr__area afr__kitchen_area afr__area*rooms \\\n", + "0 0.253413 0.063735 -0.186285 0.006208 \n", + "1 0.253413 -0.114293 -0.186285 -0.083402 \n", + "... ... ... ... ... \n", + "410773 0.253413 1.365441 1.501833 0.661427 \n", + "410774 1.175911 0.553789 -0.142544 0.807887 \n", + "\n", + " afr__area*geo_lon afr__levels*rooms afr__area*kitchen_area \\\n", + "0 -0.195129 0.060916 -0.132188 \n", + "1 0.655053 -0.054279 -0.169370 \n", + "... ... ... ... \n", + "410773 0.375199 0.752088 1.570163 \n", + "410774 -0.330070 0.982478 -0.002742 \n", + "\n", + " afr__sqrt(area)*geo_lat afr__sqrt(area)*log(level) \\\n", + "0 0.373151 0.688076 \n", + "1 0.005114 0.071369 \n", + "... ... ... \n", + "410773 1.274445 -0.002521 \n", + "410774 1.338996 0.635065 \n", + "\n", + " afr__kitchen_area*log(level) afr__sqrt(area)*kitchen_area \\\n", + "0 0.044178 -0.211335 \n", + "1 -0.173647 -0.252775 \n", + "... ... ... \n", + "410773 0.745507 2.382258 \n", + "410774 -0.040302 -0.055435 \n", + "\n", + " afr__geo_lon*log(kitchen_area) afr__sqrt(area)*sqrt(kitchen_area) \\\n", + "0 -0.481294 -0.153548 \n", + "1 1.191304 -0.267268 \n", + "... ... ... \n", + "410773 0.071599 2.828890 \n", + "410774 -1.025588 0.202136 \n", + "\n", + " afr__sqrt(geo_lon)*sqrt(kitchen_area) afr__log(area) \\\n", + "0 -0.490805 0.307835 \n", + "1 0.615798 0.031907 \n", + "... ... ... \n", + "410773 1.431272 1.729715 \n", + "410774 -0.916054 0.940624 \n", + "\n", + " afr__rooms*log(level) afr__kitchen_area*rooms \\\n", + "0 0.690329 -0.132529 \n", + "1 0.282625 -0.132529 \n", + "... ... ... \n", + "410773 -0.160491 1.581436 \n", + "410774 1.217910 0.311575 \n", + "\n", + " afr__kitchen_area*levels afr__sqrt(geo_lon)*sqrt(level) \\\n", + "0 -0.352834 0.323880 \n", + "1 -0.418643 0.552794 \n", + "... ... ... \n", + "410773 2.432437 -0.843150 \n", + "410774 -0.174762 -0.415359 \n", + "\n", + " afr__area**(3/2) afr__geo_lat*log(kitchen_area) \\\n", + "0 -0.008748 -0.031529 \n", + "1 -0.056540 -0.143829 \n", + "... ... ... \n", + "410773 0.411475 1.671069 \n", + "410774 0.135617 0.359680 \n", + "\n", + " afr__geo_lat*log(geo_lon) \n", + "0 0.068167 \n", + "1 1.129118 \n", + "... ... \n", + "410773 -1.052343 \n", + "410774 -0.246790 \n", + "\n", + "[410775 rows x 36 columns]" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "with pd.option_context('display.max_rows', 5, 'display.max_columns', None):\n", + " display (X_train_afr)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 73, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/andrey/work/institute/MLE/assets/mlflow/.venv_ml2/lib/python3.10/site-packages/autofeat/featsel.py:270: FutureWarning: Series.ravel is deprecated. The underlying array is already 1D, so ravel is not necessary. Use `to_numpy()` for conversion to a numpy array instead.\n", + " if np.max(np.abs(correlations[c].ravel()[:i])) < 0.9:\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Learning rate set to 0.105957\n", + "0:\tlearn: 22098446.5317890\ttotal: 26.8ms\tremaining: 26.8s\n", + "1:\tlearn: 21992091.7064548\ttotal: 48.4ms\tremaining: 24.2s\n", + "2:\tlearn: 21903440.5994842\ttotal: 68.8ms\tremaining: 22.9s\n", + "3:\tlearn: 21830593.5742740\ttotal: 91.3ms\tremaining: 22.7s\n", + "4:\tlearn: 21748722.8033271\ttotal: 133ms\tremaining: 26.4s\n", + "5:\tlearn: 21688987.8050450\ttotal: 165ms\tremaining: 27.3s\n", + "6:\tlearn: 21646302.4819252\ttotal: 184ms\tremaining: 26.1s\n", + "7:\tlearn: 21610896.0334767\ttotal: 212ms\tremaining: 26.3s\n", + "8:\tlearn: 21575613.5629762\ttotal: 231ms\tremaining: 25.4s\n", + "9:\tlearn: 21532487.6625332\ttotal: 250ms\tremaining: 24.7s\n", + "10:\tlearn: 21508450.4917176\ttotal: 268ms\tremaining: 24.1s\n", + "11:\tlearn: 21488319.9869637\ttotal: 286ms\tremaining: 23.6s\n", + "12:\tlearn: 21469291.7749161\ttotal: 306ms\tremaining: 23.2s\n", + "13:\tlearn: 21451889.3409628\ttotal: 322ms\tremaining: 22.7s\n", + "14:\tlearn: 20959897.0899417\ttotal: 342ms\tremaining: 22.4s\n", + "15:\tlearn: 20915005.0974918\ttotal: 361ms\tremaining: 22.2s\n", + "16:\tlearn: 20902468.9186388\ttotal: 381ms\tremaining: 22s\n", + "17:\tlearn: 20876275.9865972\ttotal: 401ms\tremaining: 21.9s\n", + "18:\tlearn: 20863745.4030443\ttotal: 419ms\tremaining: 21.6s\n", + "19:\tlearn: 20844197.8334799\ttotal: 435ms\tremaining: 21.3s\n", + "20:\tlearn: 20835022.9165113\ttotal: 453ms\tremaining: 21.1s\n", + "21:\tlearn: 20819361.3350514\ttotal: 470ms\tremaining: 20.9s\n", + "22:\tlearn: 20806038.1056633\ttotal: 490ms\tremaining: 20.8s\n", + "23:\tlearn: 20791066.0897815\ttotal: 509ms\tremaining: 20.7s\n", + "24:\tlearn: 20785138.4355827\ttotal: 530ms\tremaining: 20.7s\n", + "25:\tlearn: 20779683.7658929\ttotal: 549ms\tremaining: 20.6s\n", + "26:\tlearn: 20756913.1557035\ttotal: 569ms\tremaining: 20.5s\n", + "27:\tlearn: 20520618.7506031\ttotal: 588ms\tremaining: 20.4s\n", + "28:\tlearn: 20515058.2045385\ttotal: 613ms\tremaining: 20.5s\n", + "29:\tlearn: 20497214.8435344\ttotal: 629ms\tremaining: 20.3s\n", + "30:\tlearn: 20485495.5395393\ttotal: 645ms\tremaining: 20.2s\n", + "31:\tlearn: 20362395.4848014\ttotal: 665ms\tremaining: 20.1s\n", + "32:\tlearn: 20356605.5880092\ttotal: 680ms\tremaining: 19.9s\n", + "33:\tlearn: 20347239.3145337\ttotal: 697ms\tremaining: 19.8s\n", + "34:\tlearn: 20318678.4277386\ttotal: 716ms\tremaining: 19.7s\n", + "35:\tlearn: 20315715.1306063\ttotal: 736ms\tremaining: 19.7s\n", + "36:\tlearn: 20306647.3686296\ttotal: 750ms\tremaining: 19.5s\n", + "37:\tlearn: 20304056.5675365\ttotal: 768ms\tremaining: 19.4s\n", + "38:\tlearn: 20279918.6270216\ttotal: 787ms\tremaining: 19.4s\n", + "39:\tlearn: 20245735.4325516\ttotal: 807ms\tremaining: 19.4s\n", + "40:\tlearn: 20232809.7613533\ttotal: 823ms\tremaining: 19.2s\n", + "41:\tlearn: 20226855.1916398\ttotal: 840ms\tremaining: 19.2s\n", + "42:\tlearn: 20165027.4942330\ttotal: 858ms\tremaining: 19.1s\n", + "43:\tlearn: 20133428.0260678\ttotal: 878ms\tremaining: 19.1s\n", + "44:\tlearn: 20127643.2042711\ttotal: 894ms\tremaining: 19s\n", + "45:\tlearn: 20123741.9957328\ttotal: 909ms\tremaining: 18.9s\n", + "46:\tlearn: 20119779.4220799\ttotal: 923ms\tremaining: 18.7s\n", + "47:\tlearn: 20092957.0547636\ttotal: 943ms\tremaining: 18.7s\n", + "48:\tlearn: 19946079.6554416\ttotal: 960ms\tremaining: 18.6s\n", + "49:\tlearn: 19821808.8946012\ttotal: 977ms\tremaining: 18.6s\n", + "50:\tlearn: 19803278.9061355\ttotal: 996ms\tremaining: 18.5s\n", + "51:\tlearn: 19801590.6142514\ttotal: 1.01s\tremaining: 18.5s\n", + "52:\tlearn: 19789100.0910140\ttotal: 1.03s\tremaining: 18.5s\n", + "53:\tlearn: 19777917.4738028\ttotal: 1.05s\tremaining: 18.4s\n", + "54:\tlearn: 19774511.9959684\ttotal: 1.06s\tremaining: 18.3s\n", + "55:\tlearn: 19770278.0583089\ttotal: 1.08s\tremaining: 18.2s\n", + "56:\tlearn: 19767272.1525211\ttotal: 1.1s\tremaining: 18.1s\n", + "57:\tlearn: 19737586.6619601\ttotal: 1.12s\tremaining: 18.2s\n", + "58:\tlearn: 19735073.7186798\ttotal: 1.14s\tremaining: 18.1s\n", + "59:\tlearn: 19731977.1775776\ttotal: 1.16s\tremaining: 18.1s\n", + "60:\tlearn: 19729360.9863921\ttotal: 1.17s\tremaining: 18s\n", + "61:\tlearn: 19707341.8305192\ttotal: 1.19s\tremaining: 18s\n", + "62:\tlearn: 19705912.9617897\ttotal: 1.21s\tremaining: 17.9s\n", + "63:\tlearn: 19695569.9610055\ttotal: 1.22s\tremaining: 17.9s\n", + "64:\tlearn: 19602043.4428838\ttotal: 1.24s\tremaining: 17.9s\n", + "65:\tlearn: 19595860.8468991\ttotal: 1.26s\tremaining: 17.8s\n", + "66:\tlearn: 19585975.7792716\ttotal: 1.27s\tremaining: 17.7s\n", + "67:\tlearn: 19234362.3548543\ttotal: 1.29s\tremaining: 17.7s\n", + "68:\tlearn: 19222847.8276263\ttotal: 1.31s\tremaining: 17.6s\n", + "69:\tlearn: 19203670.0381568\ttotal: 1.33s\tremaining: 17.6s\n", + "70:\tlearn: 19194731.8572529\ttotal: 1.34s\tremaining: 17.6s\n", + "71:\tlearn: 19189178.2195288\ttotal: 1.36s\tremaining: 17.5s\n", + "72:\tlearn: 19185908.3707087\ttotal: 1.37s\tremaining: 17.4s\n", + "73:\tlearn: 19155559.0100268\ttotal: 1.39s\tremaining: 17.4s\n", + "74:\tlearn: 19152289.8647161\ttotal: 1.41s\tremaining: 17.4s\n", + "75:\tlearn: 19150374.8353251\ttotal: 1.43s\tremaining: 17.3s\n", + "76:\tlearn: 19147819.7625768\ttotal: 1.45s\tremaining: 17.3s\n", + "77:\tlearn: 18905771.5461154\ttotal: 1.47s\tremaining: 17.3s\n", + "78:\tlearn: 18813956.7525671\ttotal: 1.48s\tremaining: 17.3s\n", + "79:\tlearn: 18812152.2844158\ttotal: 1.5s\tremaining: 17.2s\n", + "80:\tlearn: 18810840.1198449\ttotal: 1.52s\tremaining: 17.2s\n", + "81:\tlearn: 18807792.9722698\ttotal: 1.54s\tremaining: 17.2s\n", + "82:\tlearn: 18794957.0370160\ttotal: 1.56s\tremaining: 17.2s\n", + "83:\tlearn: 18789742.4771797\ttotal: 1.57s\tremaining: 17.2s\n", + "84:\tlearn: 18788127.4684607\ttotal: 1.59s\tremaining: 17.2s\n", + "85:\tlearn: 18779174.8446130\ttotal: 1.61s\tremaining: 17.1s\n", + "86:\tlearn: 18776838.6599274\ttotal: 1.63s\tremaining: 17.1s\n", + "87:\tlearn: 18762161.1954900\ttotal: 1.65s\tremaining: 17.1s\n", + "88:\tlearn: 18757830.5865033\ttotal: 1.67s\tremaining: 17.1s\n", + "89:\tlearn: 18755840.4393729\ttotal: 1.68s\tremaining: 17s\n", + "90:\tlearn: 18751801.5819665\ttotal: 1.7s\tremaining: 17s\n", + "91:\tlearn: 18748802.4640957\ttotal: 1.72s\tremaining: 17s\n", + "92:\tlearn: 18745806.6911284\ttotal: 1.75s\tremaining: 17s\n", + "93:\tlearn: 18742900.1222446\ttotal: 1.77s\tremaining: 17s\n", + "94:\tlearn: 18738594.5552785\ttotal: 1.8s\tremaining: 17.1s\n", + "95:\tlearn: 18730694.4604996\ttotal: 1.82s\tremaining: 17.2s\n", + "96:\tlearn: 18716702.4931339\ttotal: 1.85s\tremaining: 17.2s\n", + "97:\tlearn: 18709192.3246265\ttotal: 1.87s\tremaining: 17.2s\n", + "98:\tlearn: 18635095.6806873\ttotal: 1.89s\tremaining: 17.2s\n", + "99:\tlearn: 18632966.0990175\ttotal: 1.91s\tremaining: 17.2s\n", + "100:\tlearn: 18630198.4193049\ttotal: 1.93s\tremaining: 17.2s\n", + "101:\tlearn: 18627320.2940055\ttotal: 1.95s\tremaining: 17.2s\n", + "102:\tlearn: 18624992.5308547\ttotal: 1.97s\tremaining: 17.2s\n", + "103:\tlearn: 18618176.9079508\ttotal: 2s\tremaining: 17.2s\n", + "104:\tlearn: 18616616.2003003\ttotal: 2.02s\tremaining: 17.2s\n", + "105:\tlearn: 18615481.8394393\ttotal: 2.04s\tremaining: 17.2s\n", + "106:\tlearn: 18594311.1106124\ttotal: 2.08s\tremaining: 17.3s\n", + "107:\tlearn: 18586972.8173311\ttotal: 2.11s\tremaining: 17.4s\n", + "108:\tlearn: 18579105.5403110\ttotal: 2.13s\tremaining: 17.5s\n", + "109:\tlearn: 18571608.0568352\ttotal: 2.16s\tremaining: 17.5s\n", + "110:\tlearn: 18564537.7386806\ttotal: 2.18s\tremaining: 17.5s\n", + "111:\tlearn: 18542636.2829769\ttotal: 2.2s\tremaining: 17.5s\n", + "112:\tlearn: 18537851.5356183\ttotal: 2.23s\tremaining: 17.5s\n", + "113:\tlearn: 18522312.5392837\ttotal: 2.26s\tremaining: 17.5s\n", + "114:\tlearn: 18299782.3915596\ttotal: 2.28s\tremaining: 17.6s\n", + "115:\tlearn: 18218935.9622759\ttotal: 2.31s\tremaining: 17.6s\n", + "116:\tlearn: 18146172.4915107\ttotal: 2.33s\tremaining: 17.6s\n", + "117:\tlearn: 18141986.4938230\ttotal: 2.35s\tremaining: 17.6s\n", + "118:\tlearn: 18140036.5077328\ttotal: 2.38s\tremaining: 17.6s\n", + "119:\tlearn: 18139373.2496994\ttotal: 2.4s\tremaining: 17.6s\n", + "120:\tlearn: 18100168.1884696\ttotal: 2.43s\tremaining: 17.6s\n", + "121:\tlearn: 18094370.6540347\ttotal: 2.45s\tremaining: 17.6s\n", + "122:\tlearn: 18076110.8650986\ttotal: 2.47s\tremaining: 17.6s\n", + "123:\tlearn: 18002967.3050945\ttotal: 2.5s\tremaining: 17.6s\n", + "124:\tlearn: 18000869.0550730\ttotal: 2.52s\tremaining: 17.6s\n", + "125:\tlearn: 17998261.8587970\ttotal: 2.54s\tremaining: 17.6s\n", + "126:\tlearn: 17994552.1626533\ttotal: 2.56s\tremaining: 17.6s\n", + "127:\tlearn: 17989433.5442655\ttotal: 2.58s\tremaining: 17.6s\n", + "128:\tlearn: 17987664.3924249\ttotal: 2.6s\tremaining: 17.5s\n", + "129:\tlearn: 17985129.5906005\ttotal: 2.62s\tremaining: 17.5s\n", + "130:\tlearn: 17767790.7079583\ttotal: 2.64s\tremaining: 17.5s\n", + "131:\tlearn: 17764614.5830372\ttotal: 2.67s\tremaining: 17.5s\n", + "132:\tlearn: 17759771.7712261\ttotal: 2.69s\tremaining: 17.5s\n", + "133:\tlearn: 17754155.8700602\ttotal: 2.71s\tremaining: 17.5s\n", + "134:\tlearn: 17693934.4243247\ttotal: 2.73s\tremaining: 17.5s\n", + "135:\tlearn: 17690379.1381722\ttotal: 2.76s\tremaining: 17.5s\n", + "136:\tlearn: 17656262.4869520\ttotal: 2.78s\tremaining: 17.5s\n", + "137:\tlearn: 17654254.9885345\ttotal: 2.8s\tremaining: 17.5s\n", + "138:\tlearn: 17652331.2748711\ttotal: 2.83s\tremaining: 17.5s\n", + "139:\tlearn: 17598030.4596968\ttotal: 2.85s\tremaining: 17.5s\n", + "140:\tlearn: 17578507.1040494\ttotal: 2.88s\tremaining: 17.5s\n", + "141:\tlearn: 17577574.9902621\ttotal: 2.9s\tremaining: 17.5s\n", + "142:\tlearn: 17573676.9930119\ttotal: 2.92s\tremaining: 17.5s\n", + "143:\tlearn: 17569875.4728705\ttotal: 2.94s\tremaining: 17.5s\n", + "144:\tlearn: 17567623.4002267\ttotal: 2.97s\tremaining: 17.5s\n", + "145:\tlearn: 17562562.5781482\ttotal: 2.99s\tremaining: 17.5s\n", + "146:\tlearn: 17557735.6477683\ttotal: 3.01s\tremaining: 17.5s\n", + "147:\tlearn: 17555290.2733648\ttotal: 3.02s\tremaining: 17.4s\n", + "148:\tlearn: 17506148.2650265\ttotal: 3.04s\tremaining: 17.4s\n", + "149:\tlearn: 17501859.3755534\ttotal: 3.06s\tremaining: 17.3s\n", + "150:\tlearn: 17320336.3209735\ttotal: 3.08s\tremaining: 17.3s\n", + "151:\tlearn: 17310638.0817615\ttotal: 3.09s\tremaining: 17.3s\n", + "152:\tlearn: 17309298.5975507\ttotal: 3.11s\tremaining: 17.2s\n", + "153:\tlearn: 17264598.6767781\ttotal: 3.13s\tremaining: 17.2s\n", + "154:\tlearn: 17262893.4148730\ttotal: 3.15s\tremaining: 17.2s\n", + "155:\tlearn: 17257977.2188279\ttotal: 3.17s\tremaining: 17.1s\n", + "156:\tlearn: 17257618.1172949\ttotal: 3.18s\tremaining: 17.1s\n", + "157:\tlearn: 17253459.4416097\ttotal: 3.2s\tremaining: 17s\n", + "158:\tlearn: 17251834.0223366\ttotal: 3.22s\tremaining: 17s\n", + "159:\tlearn: 17250494.0433762\ttotal: 3.23s\tremaining: 17s\n", + "160:\tlearn: 17246530.7193890\ttotal: 3.25s\tremaining: 16.9s\n", + "161:\tlearn: 17244324.5437997\ttotal: 3.27s\tremaining: 16.9s\n", + "162:\tlearn: 17242945.7857344\ttotal: 3.28s\tremaining: 16.9s\n", + "163:\tlearn: 17089182.2500413\ttotal: 3.3s\tremaining: 16.8s\n", + "164:\tlearn: 17088482.3072367\ttotal: 3.32s\tremaining: 16.8s\n", + "165:\tlearn: 17084007.2547317\ttotal: 3.34s\tremaining: 16.8s\n", + "166:\tlearn: 17081285.9812766\ttotal: 3.36s\tremaining: 16.8s\n", + "167:\tlearn: 17077090.3713744\ttotal: 3.38s\tremaining: 16.7s\n", + "168:\tlearn: 17074688.2414973\ttotal: 3.4s\tremaining: 16.7s\n", + "169:\tlearn: 17072284.3050889\ttotal: 3.42s\tremaining: 16.7s\n", + "170:\tlearn: 17068468.9102157\ttotal: 3.44s\tremaining: 16.7s\n", + "171:\tlearn: 17066197.4252855\ttotal: 3.46s\tremaining: 16.6s\n", + "172:\tlearn: 17065070.8976940\ttotal: 3.48s\tremaining: 16.6s\n", + "173:\tlearn: 17058249.7058775\ttotal: 3.51s\tremaining: 16.6s\n", + "174:\tlearn: 17057138.2699917\ttotal: 3.56s\tremaining: 16.8s\n", + "175:\tlearn: 17046622.4909743\ttotal: 3.6s\tremaining: 16.8s\n", + "176:\tlearn: 17039258.9949482\ttotal: 3.62s\tremaining: 16.8s\n", + "177:\tlearn: 17035020.6708538\ttotal: 3.65s\tremaining: 16.8s\n", + "178:\tlearn: 17033414.0467352\ttotal: 3.67s\tremaining: 16.8s\n", + "179:\tlearn: 17033127.0083630\ttotal: 3.69s\tremaining: 16.8s\n", + "180:\tlearn: 17031867.2300183\ttotal: 3.71s\tremaining: 16.8s\n", + "181:\tlearn: 16899607.9791508\ttotal: 3.73s\tremaining: 16.8s\n", + "182:\tlearn: 16898005.1178062\ttotal: 3.75s\tremaining: 16.7s\n", + "183:\tlearn: 16896717.9374833\ttotal: 3.76s\tremaining: 16.7s\n", + "184:\tlearn: 16894382.9799229\ttotal: 3.78s\tremaining: 16.7s\n", + "185:\tlearn: 16890727.0168983\ttotal: 3.8s\tremaining: 16.6s\n", + "186:\tlearn: 16889455.2180397\ttotal: 3.81s\tremaining: 16.6s\n", + "187:\tlearn: 16887062.5404989\ttotal: 3.83s\tremaining: 16.5s\n", + "188:\tlearn: 16886735.9764847\ttotal: 3.85s\tremaining: 16.5s\n", + "189:\tlearn: 16886228.5886393\ttotal: 3.87s\tremaining: 16.5s\n", + "190:\tlearn: 16885081.6996963\ttotal: 3.89s\tremaining: 16.5s\n", + "191:\tlearn: 16882001.5269585\ttotal: 3.9s\tremaining: 16.4s\n", + "192:\tlearn: 16856550.5531312\ttotal: 3.92s\tremaining: 16.4s\n", + "193:\tlearn: 16855256.6233436\ttotal: 3.94s\tremaining: 16.4s\n", + "194:\tlearn: 16854312.8679165\ttotal: 3.95s\tremaining: 16.3s\n", + "195:\tlearn: 16851237.1947603\ttotal: 3.97s\tremaining: 16.3s\n", + "196:\tlearn: 16783006.6207643\ttotal: 3.99s\tremaining: 16.3s\n", + "197:\tlearn: 16781595.9134206\ttotal: 4s\tremaining: 16.2s\n", + "198:\tlearn: 16781234.9503645\ttotal: 4.02s\tremaining: 16.2s\n", + "199:\tlearn: 16779709.3482853\ttotal: 4.04s\tremaining: 16.2s\n", + "200:\tlearn: 16774173.0594017\ttotal: 4.06s\tremaining: 16.1s\n", + "201:\tlearn: 16759421.8485325\ttotal: 4.08s\tremaining: 16.1s\n", + "202:\tlearn: 16758955.3742817\ttotal: 4.1s\tremaining: 16.1s\n", + "203:\tlearn: 16757936.9617034\ttotal: 4.13s\tremaining: 16.1s\n", + "204:\tlearn: 16757664.0812284\ttotal: 4.14s\tremaining: 16.1s\n", + "205:\tlearn: 16740267.4215143\ttotal: 4.17s\tremaining: 16.1s\n", + "206:\tlearn: 16738705.6070140\ttotal: 4.2s\tremaining: 16.1s\n", + "207:\tlearn: 16737706.6477167\ttotal: 4.22s\tremaining: 16.1s\n", + "208:\tlearn: 16736478.8671502\ttotal: 4.25s\tremaining: 16.1s\n", + "209:\tlearn: 16735204.9463557\ttotal: 4.28s\tremaining: 16.1s\n", + "210:\tlearn: 16732141.0391019\ttotal: 4.3s\tremaining: 16.1s\n", + "211:\tlearn: 16637668.6111380\ttotal: 4.34s\tremaining: 16.1s\n", + "212:\tlearn: 16636325.3039320\ttotal: 4.36s\tremaining: 16.1s\n", + "213:\tlearn: 16633425.3422990\ttotal: 4.38s\tremaining: 16.1s\n", + "214:\tlearn: 16633379.4218933\ttotal: 4.4s\tremaining: 16.1s\n", + "215:\tlearn: 16631776.4435300\ttotal: 4.42s\tremaining: 16s\n", + "216:\tlearn: 16628682.2349567\ttotal: 4.44s\tremaining: 16s\n", + "217:\tlearn: 16621402.6020240\ttotal: 4.46s\tremaining: 16s\n", + "218:\tlearn: 16618529.2945134\ttotal: 4.49s\tremaining: 16s\n", + "219:\tlearn: 16618289.4608715\ttotal: 4.51s\tremaining: 16s\n", + "220:\tlearn: 16616246.7258237\ttotal: 4.53s\tremaining: 16s\n", + "221:\tlearn: 16613775.7036046\ttotal: 4.55s\tremaining: 15.9s\n", + "222:\tlearn: 16612492.1305852\ttotal: 4.57s\tremaining: 15.9s\n", + "223:\tlearn: 16531999.7964791\ttotal: 4.6s\tremaining: 15.9s\n", + "224:\tlearn: 16531623.8181282\ttotal: 4.61s\tremaining: 15.9s\n", + "225:\tlearn: 16526053.4271391\ttotal: 4.64s\tremaining: 15.9s\n", + "226:\tlearn: 16513833.4679535\ttotal: 4.66s\tremaining: 15.9s\n", + "227:\tlearn: 16472689.7602332\ttotal: 4.68s\tremaining: 15.9s\n", + "228:\tlearn: 16471447.4428699\ttotal: 4.7s\tremaining: 15.8s\n", + "229:\tlearn: 16464946.2498018\ttotal: 4.73s\tremaining: 15.8s\n", + "230:\tlearn: 16455905.0848516\ttotal: 4.75s\tremaining: 15.8s\n", + "231:\tlearn: 16453392.6114472\ttotal: 4.77s\tremaining: 15.8s\n", + "232:\tlearn: 16449023.0849558\ttotal: 4.79s\tremaining: 15.8s\n", + "233:\tlearn: 16438897.0011580\ttotal: 4.81s\tremaining: 15.8s\n", + "234:\tlearn: 16438736.3967663\ttotal: 4.83s\tremaining: 15.7s\n", + "235:\tlearn: 16437731.8258800\ttotal: 4.86s\tremaining: 15.7s\n", + "236:\tlearn: 16436980.6157656\ttotal: 4.88s\tremaining: 15.7s\n", + "237:\tlearn: 16428114.7269303\ttotal: 4.9s\tremaining: 15.7s\n", + "238:\tlearn: 16428049.7767929\ttotal: 4.92s\tremaining: 15.7s\n", + "239:\tlearn: 16427992.0895827\ttotal: 4.94s\tremaining: 15.7s\n", + "240:\tlearn: 16422121.5472397\ttotal: 4.98s\tremaining: 15.7s\n", + "241:\tlearn: 16417230.7540032\ttotal: 5.01s\tremaining: 15.7s\n", + "242:\tlearn: 16416409.8463259\ttotal: 5.04s\tremaining: 15.7s\n", + "243:\tlearn: 16408285.5767443\ttotal: 5.07s\tremaining: 15.7s\n", + "244:\tlearn: 16401737.8703071\ttotal: 5.09s\tremaining: 15.7s\n", + "245:\tlearn: 16400754.7007342\ttotal: 5.12s\tremaining: 15.7s\n", + "246:\tlearn: 16400112.0638635\ttotal: 5.14s\tremaining: 15.7s\n", + "247:\tlearn: 16398418.5457083\ttotal: 5.17s\tremaining: 15.7s\n", + "248:\tlearn: 16398232.1545492\ttotal: 5.18s\tremaining: 15.6s\n", + "249:\tlearn: 16388945.7310621\ttotal: 5.21s\tremaining: 15.6s\n", + "250:\tlearn: 16375911.0840421\ttotal: 5.23s\tremaining: 15.6s\n", + "251:\tlearn: 16373443.5034802\ttotal: 5.25s\tremaining: 15.6s\n", + "252:\tlearn: 16372753.0693531\ttotal: 5.26s\tremaining: 15.5s\n", + "253:\tlearn: 16367086.2902241\ttotal: 5.28s\tremaining: 15.5s\n", + "254:\tlearn: 16358517.8494899\ttotal: 5.3s\tremaining: 15.5s\n", + "255:\tlearn: 16353748.1922851\ttotal: 5.32s\tremaining: 15.5s\n", + "256:\tlearn: 16353453.6682681\ttotal: 5.34s\tremaining: 15.4s\n", + "257:\tlearn: 16352481.0366512\ttotal: 5.36s\tremaining: 15.4s\n", + "258:\tlearn: 16351858.3396725\ttotal: 5.37s\tremaining: 15.4s\n", + "259:\tlearn: 16351336.2107277\ttotal: 5.39s\tremaining: 15.3s\n", + "260:\tlearn: 16350293.5343411\ttotal: 5.41s\tremaining: 15.3s\n", + "261:\tlearn: 16350084.7826337\ttotal: 5.42s\tremaining: 15.3s\n", + "262:\tlearn: 16345905.3652731\ttotal: 5.44s\tremaining: 15.2s\n", + "263:\tlearn: 16345758.0845827\ttotal: 5.46s\tremaining: 15.2s\n", + "264:\tlearn: 16343268.6433109\ttotal: 5.48s\tremaining: 15.2s\n", + "265:\tlearn: 16342154.0157515\ttotal: 5.5s\tremaining: 15.2s\n", + "266:\tlearn: 16342037.5821118\ttotal: 5.51s\tremaining: 15.1s\n", + "267:\tlearn: 16341927.4242046\ttotal: 5.53s\tremaining: 15.1s\n", + "268:\tlearn: 16338213.3074397\ttotal: 5.55s\tremaining: 15.1s\n", + "269:\tlearn: 16327216.0649996\ttotal: 5.57s\tremaining: 15s\n", + "270:\tlearn: 16326925.2351069\ttotal: 5.58s\tremaining: 15s\n", + "271:\tlearn: 16326820.4998486\ttotal: 5.6s\tremaining: 15s\n", + "272:\tlearn: 16322749.1113577\ttotal: 5.61s\tremaining: 14.9s\n", + "273:\tlearn: 16322549.8844895\ttotal: 5.63s\tremaining: 14.9s\n", + "274:\tlearn: 16322371.4247002\ttotal: 5.65s\tremaining: 14.9s\n", + "275:\tlearn: 16320864.8994995\ttotal: 5.67s\tremaining: 14.9s\n", + "276:\tlearn: 16320396.8778183\ttotal: 5.7s\tremaining: 14.9s\n", + "277:\tlearn: 16314920.6844460\ttotal: 5.72s\tremaining: 14.9s\n", + "278:\tlearn: 16313439.3322462\ttotal: 5.74s\tremaining: 14.8s\n", + "279:\tlearn: 16313236.2051088\ttotal: 5.76s\tremaining: 14.8s\n", + "280:\tlearn: 16311407.0527601\ttotal: 5.79s\tremaining: 14.8s\n", + "281:\tlearn: 16304548.9854054\ttotal: 5.81s\tremaining: 14.8s\n", + "282:\tlearn: 16303849.9772984\ttotal: 5.83s\tremaining: 14.8s\n", + "283:\tlearn: 16303036.5859092\ttotal: 5.85s\tremaining: 14.8s\n", + "284:\tlearn: 16301982.1736398\ttotal: 5.88s\tremaining: 14.7s\n", + "285:\tlearn: 16298399.6511182\ttotal: 5.9s\tremaining: 14.7s\n", + "286:\tlearn: 16296854.5778267\ttotal: 5.92s\tremaining: 14.7s\n", + "287:\tlearn: 16296542.4281682\ttotal: 5.94s\tremaining: 14.7s\n", + "288:\tlearn: 16295967.0032405\ttotal: 5.96s\tremaining: 14.7s\n", + "289:\tlearn: 16293311.6501461\ttotal: 5.99s\tremaining: 14.7s\n", + "290:\tlearn: 16283984.2593662\ttotal: 6.01s\tremaining: 14.6s\n", + "291:\tlearn: 16282886.4463887\ttotal: 6.03s\tremaining: 14.6s\n", + "292:\tlearn: 16281871.7729414\ttotal: 6.05s\tremaining: 14.6s\n", + "293:\tlearn: 16246204.5878946\ttotal: 6.07s\tremaining: 14.6s\n", + "294:\tlearn: 16245494.1797635\ttotal: 6.09s\tremaining: 14.6s\n", + "295:\tlearn: 16245310.9552655\ttotal: 6.11s\tremaining: 14.5s\n", + "296:\tlearn: 16179763.1037634\ttotal: 6.13s\tremaining: 14.5s\n", + "297:\tlearn: 16174244.5922696\ttotal: 6.15s\tremaining: 14.5s\n", + "298:\tlearn: 16173695.6328831\ttotal: 6.17s\tremaining: 14.5s\n", + "299:\tlearn: 16172405.7367805\ttotal: 6.2s\tremaining: 14.5s\n", + "300:\tlearn: 16172246.0701414\ttotal: 6.22s\tremaining: 14.4s\n", + "301:\tlearn: 16171107.5926661\ttotal: 6.24s\tremaining: 14.4s\n", + "302:\tlearn: 16157568.5702286\ttotal: 6.26s\tremaining: 14.4s\n", + "303:\tlearn: 16144286.9945713\ttotal: 6.29s\tremaining: 14.4s\n", + "304:\tlearn: 16144147.1664354\ttotal: 6.32s\tremaining: 14.4s\n", + "305:\tlearn: 16143334.2932193\ttotal: 6.34s\tremaining: 14.4s\n", + "306:\tlearn: 16141949.9492589\ttotal: 6.36s\tremaining: 14.4s\n", + "307:\tlearn: 16141366.3505893\ttotal: 6.39s\tremaining: 14.3s\n", + "308:\tlearn: 16137962.0407039\ttotal: 6.41s\tremaining: 14.3s\n", + "309:\tlearn: 16136765.2464605\ttotal: 6.44s\tremaining: 14.3s\n", + "310:\tlearn: 16136435.2151644\ttotal: 6.46s\tremaining: 14.3s\n", + "311:\tlearn: 16135167.3393975\ttotal: 6.49s\tremaining: 14.3s\n", + "312:\tlearn: 16126588.3099796\ttotal: 6.51s\tremaining: 14.3s\n", + "313:\tlearn: 16090881.5848383\ttotal: 6.54s\tremaining: 14.3s\n", + "314:\tlearn: 16086166.4184594\ttotal: 6.57s\tremaining: 14.3s\n", + "315:\tlearn: 16085587.1075779\ttotal: 6.58s\tremaining: 14.3s\n", + "316:\tlearn: 16080938.2192392\ttotal: 6.6s\tremaining: 14.2s\n", + "317:\tlearn: 16067462.0352494\ttotal: 6.63s\tremaining: 14.2s\n", + "318:\tlearn: 16065922.6279599\ttotal: 6.65s\tremaining: 14.2s\n", + "319:\tlearn: 16064453.8241861\ttotal: 6.67s\tremaining: 14.2s\n", + "320:\tlearn: 16064268.9105041\ttotal: 6.68s\tremaining: 14.1s\n", + "321:\tlearn: 16063736.1872678\ttotal: 6.7s\tremaining: 14.1s\n", + "322:\tlearn: 16059090.5784481\ttotal: 6.72s\tremaining: 14.1s\n", + "323:\tlearn: 16058389.9467196\ttotal: 6.74s\tremaining: 14.1s\n", + "324:\tlearn: 16058214.2923750\ttotal: 6.75s\tremaining: 14s\n", + "325:\tlearn: 16058048.1995323\ttotal: 6.76s\tremaining: 14s\n", + "326:\tlearn: 16057417.1434230\ttotal: 6.78s\tremaining: 14s\n", + "327:\tlearn: 16040701.6305809\ttotal: 6.8s\tremaining: 13.9s\n", + "328:\tlearn: 16039470.8630631\ttotal: 6.82s\tremaining: 13.9s\n", + "329:\tlearn: 16032631.5814722\ttotal: 6.84s\tremaining: 13.9s\n", + "330:\tlearn: 16032070.6826659\ttotal: 6.86s\tremaining: 13.9s\n", + "331:\tlearn: 16031915.8360432\ttotal: 6.87s\tremaining: 13.8s\n", + "332:\tlearn: 16028975.9765947\ttotal: 6.89s\tremaining: 13.8s\n", + "333:\tlearn: 16016824.1323937\ttotal: 6.91s\tremaining: 13.8s\n", + "334:\tlearn: 16016228.4994905\ttotal: 6.93s\tremaining: 13.8s\n", + "335:\tlearn: 16016070.8522111\ttotal: 6.95s\tremaining: 13.7s\n", + "336:\tlearn: 16014982.2574027\ttotal: 6.97s\tremaining: 13.7s\n", + "337:\tlearn: 16012619.6711177\ttotal: 6.98s\tremaining: 13.7s\n", + "338:\tlearn: 16011597.9679398\ttotal: 7s\tremaining: 13.6s\n", + "339:\tlearn: 16011439.6179806\ttotal: 7.01s\tremaining: 13.6s\n", + "340:\tlearn: 16011237.0264874\ttotal: 7.03s\tremaining: 13.6s\n", + "341:\tlearn: 16011100.7602433\ttotal: 7.04s\tremaining: 13.6s\n", + "342:\tlearn: 16010972.5659207\ttotal: 7.06s\tremaining: 13.5s\n", + "343:\tlearn: 16009852.0826445\ttotal: 7.08s\tremaining: 13.5s\n", + "344:\tlearn: 16009168.4228047\ttotal: 7.1s\tremaining: 13.5s\n", + "345:\tlearn: 15976690.8343983\ttotal: 7.12s\tremaining: 13.5s\n", + "346:\tlearn: 15976185.1053865\ttotal: 7.14s\tremaining: 13.4s\n", + "347:\tlearn: 15975217.3778555\ttotal: 7.16s\tremaining: 13.4s\n", + "348:\tlearn: 15946059.6698335\ttotal: 7.18s\tremaining: 13.4s\n", + "349:\tlearn: 15943921.9286677\ttotal: 7.19s\tremaining: 13.4s\n", + "350:\tlearn: 15923347.5578938\ttotal: 7.21s\tremaining: 13.3s\n", + "351:\tlearn: 15909368.9323773\ttotal: 7.23s\tremaining: 13.3s\n", + "352:\tlearn: 15908103.4358003\ttotal: 7.25s\tremaining: 13.3s\n", + "353:\tlearn: 15907697.8671225\ttotal: 7.27s\tremaining: 13.3s\n", + "354:\tlearn: 15906419.4144558\ttotal: 7.28s\tremaining: 13.2s\n", + "355:\tlearn: 15889596.3525450\ttotal: 7.3s\tremaining: 13.2s\n", + "356:\tlearn: 15887061.3042659\ttotal: 7.32s\tremaining: 13.2s\n", + "357:\tlearn: 15875563.8672287\ttotal: 7.34s\tremaining: 13.2s\n", + "358:\tlearn: 15875514.5027199\ttotal: 7.36s\tremaining: 13.1s\n", + "359:\tlearn: 15867206.7753817\ttotal: 7.38s\tremaining: 13.1s\n", + "360:\tlearn: 15859470.9974766\ttotal: 7.4s\tremaining: 13.1s\n", + "361:\tlearn: 15858985.1414787\ttotal: 7.42s\tremaining: 13.1s\n", + "362:\tlearn: 15858517.7532614\ttotal: 7.43s\tremaining: 13s\n", + "363:\tlearn: 15837491.6512974\ttotal: 7.46s\tremaining: 13s\n", + "364:\tlearn: 15831413.2891292\ttotal: 7.47s\tremaining: 13s\n", + "365:\tlearn: 15821910.7944137\ttotal: 7.5s\tremaining: 13s\n", + "366:\tlearn: 15821799.9898810\ttotal: 7.51s\tremaining: 12.9s\n", + "367:\tlearn: 15820518.3355521\ttotal: 7.53s\tremaining: 12.9s\n", + "368:\tlearn: 15819654.1102879\ttotal: 7.55s\tremaining: 12.9s\n", + "369:\tlearn: 15817089.7995770\ttotal: 7.57s\tremaining: 12.9s\n", + "370:\tlearn: 15766391.6002990\ttotal: 7.58s\tremaining: 12.9s\n", + "371:\tlearn: 15765040.0677397\ttotal: 7.6s\tremaining: 12.8s\n", + "372:\tlearn: 15760061.5223284\ttotal: 7.62s\tremaining: 12.8s\n", + "373:\tlearn: 15749082.3035499\ttotal: 7.64s\tremaining: 12.8s\n", + "374:\tlearn: 15747797.5966906\ttotal: 7.66s\tremaining: 12.8s\n", + "375:\tlearn: 15747152.5320861\ttotal: 7.67s\tremaining: 12.7s\n", + "376:\tlearn: 15746747.2372895\ttotal: 7.69s\tremaining: 12.7s\n", + "377:\tlearn: 15745624.1380163\ttotal: 7.71s\tremaining: 12.7s\n", + "378:\tlearn: 15744096.9300078\ttotal: 7.73s\tremaining: 12.7s\n", + "379:\tlearn: 15742874.3850591\ttotal: 7.75s\tremaining: 12.6s\n", + "380:\tlearn: 15733524.5241015\ttotal: 7.77s\tremaining: 12.6s\n", + "381:\tlearn: 15732270.9170805\ttotal: 7.79s\tremaining: 12.6s\n", + "382:\tlearn: 15719725.8632121\ttotal: 7.81s\tremaining: 12.6s\n", + "383:\tlearn: 15719230.3528365\ttotal: 7.83s\tremaining: 12.6s\n", + "384:\tlearn: 15676373.5724698\ttotal: 7.84s\tremaining: 12.5s\n", + "385:\tlearn: 15671887.7149678\ttotal: 7.87s\tremaining: 12.5s\n", + "386:\tlearn: 15670551.0130636\ttotal: 7.88s\tremaining: 12.5s\n", + "387:\tlearn: 15670425.5008512\ttotal: 7.9s\tremaining: 12.5s\n", + "388:\tlearn: 15669002.2342788\ttotal: 7.92s\tremaining: 12.4s\n", + "389:\tlearn: 15658431.6652457\ttotal: 7.94s\tremaining: 12.4s\n", + "390:\tlearn: 15652471.3699308\ttotal: 7.96s\tremaining: 12.4s\n", + "391:\tlearn: 15651498.0813307\ttotal: 7.98s\tremaining: 12.4s\n", + "392:\tlearn: 15651135.3309948\ttotal: 8s\tremaining: 12.4s\n", + "393:\tlearn: 15650319.0160971\ttotal: 8.02s\tremaining: 12.3s\n", + "394:\tlearn: 15623676.9804036\ttotal: 8.04s\tremaining: 12.3s\n", + "395:\tlearn: 15623224.1380220\ttotal: 8.06s\tremaining: 12.3s\n", + "396:\tlearn: 15623121.3090365\ttotal: 8.07s\tremaining: 12.3s\n", + "397:\tlearn: 15623023.2887270\ttotal: 8.09s\tremaining: 12.2s\n", + "398:\tlearn: 15622927.4120694\ttotal: 8.11s\tremaining: 12.2s\n", + "399:\tlearn: 15609221.8376412\ttotal: 8.13s\tremaining: 12.2s\n", + "400:\tlearn: 15608322.7052502\ttotal: 8.15s\tremaining: 12.2s\n", + "401:\tlearn: 15584384.0903136\ttotal: 8.17s\tremaining: 12.1s\n", + "402:\tlearn: 15583221.7488379\ttotal: 8.19s\tremaining: 12.1s\n", + "403:\tlearn: 15582377.4877972\ttotal: 8.2s\tremaining: 12.1s\n", + "404:\tlearn: 15582017.7231803\ttotal: 8.23s\tremaining: 12.1s\n", + "405:\tlearn: 15576732.9317645\ttotal: 8.24s\tremaining: 12.1s\n", + "406:\tlearn: 15574973.9524782\ttotal: 8.26s\tremaining: 12s\n", + "407:\tlearn: 15564448.9029698\ttotal: 8.28s\tremaining: 12s\n", + "408:\tlearn: 15543394.5508652\ttotal: 8.3s\tremaining: 12s\n", + "409:\tlearn: 15534527.8559499\ttotal: 8.32s\tremaining: 12s\n", + "410:\tlearn: 15530876.5817135\ttotal: 8.34s\tremaining: 11.9s\n", + "411:\tlearn: 15530436.2178164\ttotal: 8.35s\tremaining: 11.9s\n", + "412:\tlearn: 15508718.2497187\ttotal: 8.37s\tremaining: 11.9s\n", + "413:\tlearn: 15508312.0517832\ttotal: 8.39s\tremaining: 11.9s\n", + "414:\tlearn: 15497587.2325260\ttotal: 8.41s\tremaining: 11.9s\n", + "415:\tlearn: 15496501.5231033\ttotal: 8.43s\tremaining: 11.8s\n", + "416:\tlearn: 15495748.9679620\ttotal: 8.44s\tremaining: 11.8s\n", + "417:\tlearn: 15495493.0983923\ttotal: 8.46s\tremaining: 11.8s\n", + "418:\tlearn: 15495157.7485844\ttotal: 8.48s\tremaining: 11.8s\n", + "419:\tlearn: 15494761.4867226\ttotal: 8.49s\tremaining: 11.7s\n", + "420:\tlearn: 15476869.5579718\ttotal: 8.52s\tremaining: 11.7s\n", + "421:\tlearn: 15476568.3211614\ttotal: 8.54s\tremaining: 11.7s\n", + "422:\tlearn: 15457053.1662726\ttotal: 8.56s\tremaining: 11.7s\n", + "423:\tlearn: 15454248.0587157\ttotal: 8.58s\tremaining: 11.7s\n", + "424:\tlearn: 15453981.7021927\ttotal: 8.6s\tremaining: 11.6s\n", + "425:\tlearn: 15445132.0573074\ttotal: 8.62s\tremaining: 11.6s\n", + "426:\tlearn: 15444808.9479642\ttotal: 8.64s\tremaining: 11.6s\n", + "427:\tlearn: 15443874.8831344\ttotal: 8.65s\tremaining: 11.6s\n", + "428:\tlearn: 15425719.7215241\ttotal: 8.67s\tremaining: 11.5s\n", + "429:\tlearn: 15407994.7740128\ttotal: 8.69s\tremaining: 11.5s\n", + "430:\tlearn: 15398613.7193006\ttotal: 8.72s\tremaining: 11.5s\n", + "431:\tlearn: 15398242.2656151\ttotal: 8.73s\tremaining: 11.5s\n", + "432:\tlearn: 15396326.0844633\ttotal: 8.75s\tremaining: 11.5s\n", + "433:\tlearn: 15396117.7697572\ttotal: 8.76s\tremaining: 11.4s\n", + "434:\tlearn: 15394399.1314246\ttotal: 8.78s\tremaining: 11.4s\n", + "435:\tlearn: 15394033.3822923\ttotal: 8.8s\tremaining: 11.4s\n", + "436:\tlearn: 15390957.2266817\ttotal: 8.82s\tremaining: 11.4s\n", + "437:\tlearn: 15382395.8217422\ttotal: 8.84s\tremaining: 11.3s\n", + "438:\tlearn: 15365582.8921318\ttotal: 8.87s\tremaining: 11.3s\n", + "439:\tlearn: 15364963.5068288\ttotal: 8.88s\tremaining: 11.3s\n", + "440:\tlearn: 15361146.0696731\ttotal: 8.91s\tremaining: 11.3s\n", + "441:\tlearn: 15360088.6108971\ttotal: 8.93s\tremaining: 11.3s\n", + "442:\tlearn: 15358099.3599881\ttotal: 8.95s\tremaining: 11.2s\n", + "443:\tlearn: 15357486.6853299\ttotal: 8.96s\tremaining: 11.2s\n", + "444:\tlearn: 15357164.2756586\ttotal: 8.98s\tremaining: 11.2s\n", + "445:\tlearn: 15356822.8743038\ttotal: 8.99s\tremaining: 11.2s\n", + "446:\tlearn: 15344640.3766766\ttotal: 9.02s\tremaining: 11.2s\n", + "447:\tlearn: 15334858.3266831\ttotal: 9.04s\tremaining: 11.1s\n", + "448:\tlearn: 15309546.1099028\ttotal: 9.06s\tremaining: 11.1s\n", + "449:\tlearn: 15309027.9299073\ttotal: 9.08s\tremaining: 11.1s\n", + "450:\tlearn: 15307537.2549447\ttotal: 9.1s\tremaining: 11.1s\n", + "451:\tlearn: 15307029.7535768\ttotal: 9.11s\tremaining: 11s\n", + "452:\tlearn: 15306746.7683229\ttotal: 9.13s\tremaining: 11s\n", + "453:\tlearn: 15305123.1900217\ttotal: 9.14s\tremaining: 11s\n", + "454:\tlearn: 15295824.5259009\ttotal: 9.16s\tremaining: 11s\n", + "455:\tlearn: 15290559.5079922\ttotal: 9.18s\tremaining: 11s\n", + "456:\tlearn: 15289236.3362428\ttotal: 9.2s\tremaining: 10.9s\n", + "457:\tlearn: 15285680.8241152\ttotal: 9.23s\tremaining: 10.9s\n", + "458:\tlearn: 15276859.6343040\ttotal: 9.24s\tremaining: 10.9s\n", + "459:\tlearn: 15275894.2453187\ttotal: 9.26s\tremaining: 10.9s\n", + "460:\tlearn: 15271709.8818145\ttotal: 9.28s\tremaining: 10.9s\n", + "461:\tlearn: 15271188.8035764\ttotal: 9.3s\tremaining: 10.8s\n", + "462:\tlearn: 15267899.4071259\ttotal: 9.32s\tremaining: 10.8s\n", + "463:\tlearn: 15267657.0601824\ttotal: 9.33s\tremaining: 10.8s\n", + "464:\tlearn: 15237652.9178159\ttotal: 9.35s\tremaining: 10.8s\n", + "465:\tlearn: 15229269.3743958\ttotal: 9.37s\tremaining: 10.7s\n", + "466:\tlearn: 15228949.2022209\ttotal: 9.38s\tremaining: 10.7s\n", + "467:\tlearn: 15228708.7243314\ttotal: 9.4s\tremaining: 10.7s\n", + "468:\tlearn: 15228104.4976588\ttotal: 9.42s\tremaining: 10.7s\n", + "469:\tlearn: 15220454.4378489\ttotal: 9.44s\tremaining: 10.6s\n", + "470:\tlearn: 15218384.6387816\ttotal: 9.46s\tremaining: 10.6s\n", + "471:\tlearn: 15207740.8458334\ttotal: 9.48s\tremaining: 10.6s\n", + "472:\tlearn: 15188696.0366634\ttotal: 9.5s\tremaining: 10.6s\n", + "473:\tlearn: 15187194.3517643\ttotal: 9.52s\tremaining: 10.6s\n", + "474:\tlearn: 15181425.0998155\ttotal: 9.54s\tremaining: 10.5s\n", + "475:\tlearn: 15175973.9576126\ttotal: 9.56s\tremaining: 10.5s\n", + "476:\tlearn: 15175836.7822360\ttotal: 9.58s\tremaining: 10.5s\n", + "477:\tlearn: 15161401.1824127\ttotal: 9.6s\tremaining: 10.5s\n", + "478:\tlearn: 15153175.3872623\ttotal: 9.62s\tremaining: 10.5s\n", + "479:\tlearn: 15150675.0588374\ttotal: 9.64s\tremaining: 10.4s\n", + "480:\tlearn: 15125191.0312629\ttotal: 9.66s\tremaining: 10.4s\n", + "481:\tlearn: 15124142.4843150\ttotal: 9.68s\tremaining: 10.4s\n", + "482:\tlearn: 15113727.7397332\ttotal: 9.7s\tremaining: 10.4s\n", + "483:\tlearn: 15113631.2944187\ttotal: 9.71s\tremaining: 10.4s\n", + "484:\tlearn: 15112144.2828101\ttotal: 9.73s\tremaining: 10.3s\n", + "485:\tlearn: 15104136.3121881\ttotal: 9.75s\tremaining: 10.3s\n", + "486:\tlearn: 15101260.1405676\ttotal: 9.76s\tremaining: 10.3s\n", + "487:\tlearn: 15100486.1242395\ttotal: 9.78s\tremaining: 10.3s\n", + "488:\tlearn: 15087634.4821946\ttotal: 9.8s\tremaining: 10.2s\n", + "489:\tlearn: 15087440.2245165\ttotal: 9.82s\tremaining: 10.2s\n", + "490:\tlearn: 15086512.4354190\ttotal: 9.84s\tremaining: 10.2s\n", + "491:\tlearn: 15086303.4626980\ttotal: 9.86s\tremaining: 10.2s\n", + "492:\tlearn: 15085311.6786629\ttotal: 9.87s\tremaining: 10.2s\n", + "493:\tlearn: 15084910.4748658\ttotal: 9.89s\tremaining: 10.1s\n", + "494:\tlearn: 15084117.3189372\ttotal: 9.91s\tremaining: 10.1s\n", + "495:\tlearn: 15083173.0721914\ttotal: 9.93s\tremaining: 10.1s\n", + "496:\tlearn: 15083050.9104097\ttotal: 9.95s\tremaining: 10.1s\n", + "497:\tlearn: 15082864.4673084\ttotal: 9.96s\tremaining: 10s\n", + "498:\tlearn: 15080381.3453750\ttotal: 9.98s\tremaining: 10s\n", + "499:\tlearn: 15080168.5694770\ttotal: 9.99s\tremaining: 9.99s\n", + "500:\tlearn: 15078324.0060394\ttotal: 10s\tremaining: 9.97s\n", + "501:\tlearn: 15056755.6274416\ttotal: 10s\tremaining: 9.95s\n", + "502:\tlearn: 15042382.2129683\ttotal: 10.1s\tremaining: 9.94s\n", + "503:\tlearn: 15026820.6795480\ttotal: 10.1s\tremaining: 9.92s\n", + "504:\tlearn: 15024745.0307165\ttotal: 10.1s\tremaining: 9.9s\n", + "505:\tlearn: 15024568.7525820\ttotal: 10.1s\tremaining: 9.87s\n", + "506:\tlearn: 15024336.2234656\ttotal: 10.1s\tremaining: 9.85s\n", + "507:\tlearn: 15023422.4830957\ttotal: 10.1s\tremaining: 9.83s\n", + "508:\tlearn: 15023240.8907292\ttotal: 10.2s\tremaining: 9.8s\n", + "509:\tlearn: 15022616.1932522\ttotal: 10.2s\tremaining: 9.78s\n", + "510:\tlearn: 15022376.1992724\ttotal: 10.2s\tremaining: 9.76s\n", + "511:\tlearn: 15022278.6413562\ttotal: 10.2s\tremaining: 9.74s\n", + "512:\tlearn: 15004327.6956801\ttotal: 10.2s\tremaining: 9.72s\n", + "513:\tlearn: 15003257.9686608\ttotal: 10.3s\tremaining: 9.69s\n", + "514:\tlearn: 14990786.3446388\ttotal: 10.3s\tremaining: 9.68s\n", + "515:\tlearn: 14990443.3344833\ttotal: 10.3s\tremaining: 9.65s\n", + "516:\tlearn: 14990121.8336965\ttotal: 10.3s\tremaining: 9.63s\n", + "517:\tlearn: 14989690.9644545\ttotal: 10.3s\tremaining: 9.61s\n", + "518:\tlearn: 14989542.6281186\ttotal: 10.3s\tremaining: 9.58s\n", + "519:\tlearn: 14989263.7455940\ttotal: 10.4s\tremaining: 9.56s\n", + "520:\tlearn: 14989009.4488058\ttotal: 10.4s\tremaining: 9.54s\n", + "521:\tlearn: 14987311.9040697\ttotal: 10.4s\tremaining: 9.52s\n", + "522:\tlearn: 14986317.7528784\ttotal: 10.4s\tremaining: 9.49s\n", + "523:\tlearn: 14985961.1745765\ttotal: 10.4s\tremaining: 9.47s\n", + "524:\tlearn: 14968328.3342953\ttotal: 10.4s\tremaining: 9.45s\n", + "525:\tlearn: 14962501.6277491\ttotal: 10.5s\tremaining: 9.43s\n", + "526:\tlearn: 14942891.6903938\ttotal: 10.5s\tremaining: 9.41s\n", + "527:\tlearn: 14942347.7961868\ttotal: 10.5s\tremaining: 9.39s\n", + "528:\tlearn: 14934661.6369111\ttotal: 10.5s\tremaining: 9.37s\n", + "529:\tlearn: 14934441.0003035\ttotal: 10.5s\tremaining: 9.35s\n", + "530:\tlearn: 14931641.8082484\ttotal: 10.6s\tremaining: 9.33s\n", + "531:\tlearn: 14930520.1388588\ttotal: 10.6s\tremaining: 9.31s\n", + "532:\tlearn: 14929412.0007072\ttotal: 10.6s\tremaining: 9.29s\n", + "533:\tlearn: 14915426.1594625\ttotal: 10.6s\tremaining: 9.27s\n", + "534:\tlearn: 14904596.5361540\ttotal: 10.6s\tremaining: 9.26s\n", + "535:\tlearn: 14890185.4475504\ttotal: 10.7s\tremaining: 9.24s\n", + "536:\tlearn: 14886779.6625443\ttotal: 10.7s\tremaining: 9.22s\n", + "537:\tlearn: 14876751.8448602\ttotal: 10.7s\tremaining: 9.2s\n", + "538:\tlearn: 14876547.3177551\ttotal: 10.7s\tremaining: 9.18s\n", + "539:\tlearn: 14875843.9123383\ttotal: 10.8s\tremaining: 9.17s\n", + "540:\tlearn: 14875635.6509434\ttotal: 10.8s\tremaining: 9.15s\n", + "541:\tlearn: 14874902.7520310\ttotal: 10.8s\tremaining: 9.13s\n", + "542:\tlearn: 14859983.2972329\ttotal: 10.8s\tremaining: 9.12s\n", + "543:\tlearn: 14859037.2857669\ttotal: 10.8s\tremaining: 9.09s\n", + "544:\tlearn: 14858735.3241348\ttotal: 10.9s\tremaining: 9.07s\n", + "545:\tlearn: 14858447.8254575\ttotal: 10.9s\tremaining: 9.05s\n", + "546:\tlearn: 14858039.6683507\ttotal: 10.9s\tremaining: 9.03s\n", + "547:\tlearn: 14855695.5721103\ttotal: 10.9s\tremaining: 9.01s\n", + "548:\tlearn: 14855414.9960052\ttotal: 10.9s\tremaining: 8.98s\n", + "549:\tlearn: 14842814.4427164\ttotal: 11s\tremaining: 8.96s\n", + "550:\tlearn: 14835564.3391437\ttotal: 11s\tremaining: 8.94s\n", + "551:\tlearn: 14819577.7196652\ttotal: 11s\tremaining: 8.92s\n", + "552:\tlearn: 14819455.2265537\ttotal: 11s\tremaining: 8.9s\n", + "553:\tlearn: 14806899.3496044\ttotal: 11s\tremaining: 8.88s\n", + "554:\tlearn: 14773549.6793954\ttotal: 11.1s\tremaining: 8.86s\n", + "555:\tlearn: 14769089.4691329\ttotal: 11.1s\tremaining: 8.85s\n", + "556:\tlearn: 14759669.5852017\ttotal: 11.1s\tremaining: 8.83s\n", + "557:\tlearn: 14752711.4885984\ttotal: 11.1s\tremaining: 8.81s\n", + "558:\tlearn: 14752461.2596739\ttotal: 11.1s\tremaining: 8.78s\n", + "559:\tlearn: 14745918.7043695\ttotal: 11.2s\tremaining: 8.76s\n", + "560:\tlearn: 14736330.8545338\ttotal: 11.2s\tremaining: 8.74s\n", + "561:\tlearn: 14735421.4083610\ttotal: 11.2s\tremaining: 8.72s\n", + "562:\tlearn: 14734938.2732989\ttotal: 11.2s\tremaining: 8.7s\n", + "563:\tlearn: 14734447.1224459\ttotal: 11.2s\tremaining: 8.67s\n", + "564:\tlearn: 14726118.5713267\ttotal: 11.2s\tremaining: 8.65s\n", + "565:\tlearn: 14719546.1823157\ttotal: 11.3s\tremaining: 8.64s\n", + "566:\tlearn: 14709567.4599738\ttotal: 11.3s\tremaining: 8.62s\n", + "567:\tlearn: 14694027.9597995\ttotal: 11.3s\tremaining: 8.6s\n", + "568:\tlearn: 14693692.9986164\ttotal: 11.3s\tremaining: 8.58s\n", + "569:\tlearn: 14693561.9187439\ttotal: 11.3s\tremaining: 8.55s\n", + "570:\tlearn: 14693442.0160345\ttotal: 11.3s\tremaining: 8.53s\n", + "571:\tlearn: 14692737.3364226\ttotal: 11.4s\tremaining: 8.51s\n", + "572:\tlearn: 14692137.5271357\ttotal: 11.4s\tremaining: 8.49s\n", + "573:\tlearn: 14691530.7388111\ttotal: 11.4s\tremaining: 8.46s\n", + "574:\tlearn: 14691029.8955245\ttotal: 11.4s\tremaining: 8.44s\n", + "575:\tlearn: 14690818.7607646\ttotal: 11.4s\tremaining: 8.42s\n", + "576:\tlearn: 14681703.1018489\ttotal: 11.5s\tremaining: 8.4s\n", + "577:\tlearn: 14680949.4082557\ttotal: 11.5s\tremaining: 8.38s\n", + "578:\tlearn: 14676048.4086374\ttotal: 11.5s\tremaining: 8.36s\n", + "579:\tlearn: 14675589.1077203\ttotal: 11.5s\tremaining: 8.34s\n", + "580:\tlearn: 14671195.7882485\ttotal: 11.5s\tremaining: 8.32s\n", + "581:\tlearn: 14670480.8296800\ttotal: 11.6s\tremaining: 8.3s\n", + "582:\tlearn: 14654727.4539384\ttotal: 11.6s\tremaining: 8.28s\n", + "583:\tlearn: 14639429.1254808\ttotal: 11.6s\tremaining: 8.26s\n", + "584:\tlearn: 14633852.7201094\ttotal: 11.6s\tremaining: 8.24s\n", + "585:\tlearn: 14633703.6960294\ttotal: 11.6s\tremaining: 8.22s\n", + "586:\tlearn: 14624412.9464839\ttotal: 11.7s\tremaining: 8.2s\n", + "587:\tlearn: 14618972.8317531\ttotal: 11.7s\tremaining: 8.18s\n", + "588:\tlearn: 14616486.0209931\ttotal: 11.7s\tremaining: 8.16s\n", + "589:\tlearn: 14615131.0944873\ttotal: 11.7s\tremaining: 8.14s\n", + "590:\tlearn: 14613962.8427238\ttotal: 11.7s\tremaining: 8.12s\n", + "591:\tlearn: 14613842.5042162\ttotal: 11.7s\tremaining: 8.1s\n", + "592:\tlearn: 14607367.4815511\ttotal: 11.8s\tremaining: 8.08s\n", + "593:\tlearn: 14595899.0274001\ttotal: 11.8s\tremaining: 8.06s\n", + "594:\tlearn: 14581493.5554420\ttotal: 11.8s\tremaining: 8.04s\n", + "595:\tlearn: 14569978.9787320\ttotal: 11.8s\tremaining: 8.02s\n", + "596:\tlearn: 14567790.5291056\ttotal: 11.8s\tremaining: 8s\n", + "597:\tlearn: 14565382.7120784\ttotal: 11.9s\tremaining: 7.98s\n", + "598:\tlearn: 14563687.4847174\ttotal: 11.9s\tremaining: 7.96s\n", + "599:\tlearn: 14560814.4487806\ttotal: 11.9s\tremaining: 7.94s\n", + "600:\tlearn: 14560701.1190891\ttotal: 11.9s\tremaining: 7.92s\n", + "601:\tlearn: 14559812.7564715\ttotal: 11.9s\tremaining: 7.89s\n", + "602:\tlearn: 14551099.3396330\ttotal: 12s\tremaining: 7.87s\n", + "603:\tlearn: 14543008.3301129\ttotal: 12s\tremaining: 7.85s\n", + "604:\tlearn: 14538512.1145981\ttotal: 12s\tremaining: 7.83s\n", + "605:\tlearn: 14524779.0889213\ttotal: 12s\tremaining: 7.82s\n", + "606:\tlearn: 14524617.6849148\ttotal: 12s\tremaining: 7.8s\n", + "607:\tlearn: 14511616.4568623\ttotal: 12.1s\tremaining: 7.78s\n", + "608:\tlearn: 14511232.8044844\ttotal: 12.1s\tremaining: 7.76s\n", + "609:\tlearn: 14511060.2609313\ttotal: 12.1s\tremaining: 7.74s\n", + "610:\tlearn: 14479396.7266039\ttotal: 12.1s\tremaining: 7.72s\n", + "611:\tlearn: 14473120.4120408\ttotal: 12.1s\tremaining: 7.7s\n", + "612:\tlearn: 14472716.0504790\ttotal: 12.2s\tremaining: 7.68s\n", + "613:\tlearn: 14471477.4257026\ttotal: 12.2s\tremaining: 7.66s\n", + "614:\tlearn: 14465162.0927641\ttotal: 12.2s\tremaining: 7.64s\n", + "615:\tlearn: 14450476.2512675\ttotal: 12.2s\tremaining: 7.62s\n", + "616:\tlearn: 14441209.0170320\ttotal: 12.2s\tremaining: 7.6s\n", + "617:\tlearn: 14436956.0584123\ttotal: 12.3s\tremaining: 7.58s\n", + "618:\tlearn: 14428404.3865433\ttotal: 12.3s\tremaining: 7.56s\n", + "619:\tlearn: 14428240.6940479\ttotal: 12.3s\tremaining: 7.54s\n", + "620:\tlearn: 14426072.4923693\ttotal: 12.3s\tremaining: 7.52s\n", + "621:\tlearn: 14425962.6965766\ttotal: 12.3s\tremaining: 7.5s\n", + "622:\tlearn: 14425800.7502365\ttotal: 12.4s\tremaining: 7.47s\n", + "623:\tlearn: 14425596.9394628\ttotal: 12.4s\tremaining: 7.45s\n", + "624:\tlearn: 14425425.0487104\ttotal: 12.4s\tremaining: 7.43s\n", + "625:\tlearn: 14424623.9312247\ttotal: 12.4s\tremaining: 7.41s\n", + "626:\tlearn: 14424381.6428818\ttotal: 12.4s\tremaining: 7.39s\n", + "627:\tlearn: 14424300.2618762\ttotal: 12.4s\tremaining: 7.37s\n", + "628:\tlearn: 14423658.2257666\ttotal: 12.5s\tremaining: 7.34s\n", + "629:\tlearn: 14423367.1811322\ttotal: 12.5s\tremaining: 7.32s\n", + "630:\tlearn: 14422983.6379172\ttotal: 12.5s\tremaining: 7.3s\n", + "631:\tlearn: 14419256.4088209\ttotal: 12.5s\tremaining: 7.29s\n", + "632:\tlearn: 14417950.8214231\ttotal: 12.5s\tremaining: 7.27s\n", + "633:\tlearn: 14417585.5655445\ttotal: 12.6s\tremaining: 7.25s\n", + "634:\tlearn: 14417285.4725608\ttotal: 12.6s\tremaining: 7.23s\n", + "635:\tlearn: 14411450.5385767\ttotal: 12.6s\tremaining: 7.21s\n", + "636:\tlearn: 14410705.4825797\ttotal: 12.6s\tremaining: 7.19s\n", + "637:\tlearn: 14402850.1386270\ttotal: 12.6s\tremaining: 7.17s\n", + "638:\tlearn: 14394505.5011945\ttotal: 12.7s\tremaining: 7.15s\n", + "639:\tlearn: 14389880.7745743\ttotal: 12.7s\tremaining: 7.13s\n", + "640:\tlearn: 14377444.1374488\ttotal: 12.7s\tremaining: 7.11s\n", + "641:\tlearn: 14377195.5387020\ttotal: 12.7s\tremaining: 7.09s\n", + "642:\tlearn: 14376865.8516819\ttotal: 12.7s\tremaining: 7.07s\n", + "643:\tlearn: 14376087.6875871\ttotal: 12.7s\tremaining: 7.04s\n", + "644:\tlearn: 14370960.0957715\ttotal: 12.8s\tremaining: 7.03s\n", + "645:\tlearn: 14335061.7135929\ttotal: 12.8s\tremaining: 7.01s\n", + "646:\tlearn: 14333517.2675055\ttotal: 12.8s\tremaining: 6.99s\n", + "647:\tlearn: 14332926.4750268\ttotal: 12.8s\tremaining: 6.97s\n", + "648:\tlearn: 14326885.6760510\ttotal: 12.9s\tremaining: 6.95s\n", + "649:\tlearn: 14319019.0801501\ttotal: 12.9s\tremaining: 6.93s\n", + "650:\tlearn: 14311067.4550712\ttotal: 12.9s\tremaining: 6.92s\n", + "651:\tlearn: 14310524.2181344\ttotal: 12.9s\tremaining: 6.9s\n", + "652:\tlearn: 14306662.6889616\ttotal: 12.9s\tremaining: 6.88s\n", + "653:\tlearn: 14306620.9494302\ttotal: 13s\tremaining: 6.86s\n", + "654:\tlearn: 14299409.9997647\ttotal: 13s\tremaining: 6.84s\n", + "655:\tlearn: 14291944.1281620\ttotal: 13s\tremaining: 6.82s\n", + "656:\tlearn: 14287581.3648732\ttotal: 13s\tremaining: 6.8s\n", + "657:\tlearn: 14284339.1795952\ttotal: 13s\tremaining: 6.78s\n", + "658:\tlearn: 14277508.6533752\ttotal: 13.1s\tremaining: 6.76s\n", + "659:\tlearn: 14276678.1578443\ttotal: 13.1s\tremaining: 6.74s\n", + "660:\tlearn: 14275723.1778056\ttotal: 13.1s\tremaining: 6.72s\n", + "661:\tlearn: 14275270.1651625\ttotal: 13.1s\tremaining: 6.7s\n", + "662:\tlearn: 14266913.2486421\ttotal: 13.1s\tremaining: 6.67s\n", + "663:\tlearn: 14266545.0175669\ttotal: 13.1s\tremaining: 6.65s\n", + "664:\tlearn: 14233391.6468042\ttotal: 13.2s\tremaining: 6.63s\n", + "665:\tlearn: 14233247.5294223\ttotal: 13.2s\tremaining: 6.62s\n", + "666:\tlearn: 14229273.4180518\ttotal: 13.2s\tremaining: 6.59s\n", + "667:\tlearn: 14229079.1897143\ttotal: 13.2s\tremaining: 6.57s\n", + "668:\tlearn: 14221641.5929433\ttotal: 13.2s\tremaining: 6.55s\n", + "669:\tlearn: 14221550.9286433\ttotal: 13.3s\tremaining: 6.53s\n", + "670:\tlearn: 14216689.3069859\ttotal: 13.3s\tremaining: 6.51s\n", + "671:\tlearn: 14209543.5584520\ttotal: 13.3s\tremaining: 6.49s\n", + "672:\tlearn: 14202612.4178331\ttotal: 13.3s\tremaining: 6.47s\n", + "673:\tlearn: 14172077.7613748\ttotal: 13.3s\tremaining: 6.46s\n", + "674:\tlearn: 14171162.6718566\ttotal: 13.4s\tremaining: 6.44s\n", + "675:\tlearn: 14170808.6487783\ttotal: 13.4s\tremaining: 6.42s\n", + "676:\tlearn: 14169776.8913464\ttotal: 13.4s\tremaining: 6.4s\n", + "677:\tlearn: 14169502.2201892\ttotal: 13.4s\tremaining: 6.38s\n", + "678:\tlearn: 14169400.6557676\ttotal: 13.4s\tremaining: 6.36s\n", + "679:\tlearn: 14169091.1534075\ttotal: 13.5s\tremaining: 6.33s\n", + "680:\tlearn: 14155284.1836303\ttotal: 13.5s\tremaining: 6.31s\n", + "681:\tlearn: 14151372.7789899\ttotal: 13.5s\tremaining: 6.29s\n", + "682:\tlearn: 14151116.1785403\ttotal: 13.5s\tremaining: 6.27s\n", + "683:\tlearn: 14143570.0998262\ttotal: 13.5s\tremaining: 6.25s\n", + "684:\tlearn: 14143402.2361827\ttotal: 13.6s\tremaining: 6.23s\n", + "685:\tlearn: 14142685.0144028\ttotal: 13.6s\tremaining: 6.21s\n", + "686:\tlearn: 14142429.1239136\ttotal: 13.6s\tremaining: 6.19s\n", + "687:\tlearn: 14141516.7840252\ttotal: 13.6s\tremaining: 6.17s\n", + "688:\tlearn: 14141303.6839720\ttotal: 13.6s\tremaining: 6.15s\n", + "689:\tlearn: 14141063.6576353\ttotal: 13.6s\tremaining: 6.13s\n", + "690:\tlearn: 14140704.6119570\ttotal: 13.7s\tremaining: 6.11s\n", + "691:\tlearn: 14140169.1508485\ttotal: 13.7s\tremaining: 6.09s\n", + "692:\tlearn: 14139962.7510552\ttotal: 13.7s\tremaining: 6.07s\n", + "693:\tlearn: 14139698.0396403\ttotal: 13.7s\tremaining: 6.04s\n", + "694:\tlearn: 14126587.3299292\ttotal: 13.7s\tremaining: 6.03s\n", + "695:\tlearn: 14098439.5247326\ttotal: 13.8s\tremaining: 6.01s\n", + "696:\tlearn: 14092413.9670173\ttotal: 13.8s\tremaining: 5.99s\n", + "697:\tlearn: 14091976.7326058\ttotal: 13.8s\tremaining: 5.97s\n", + "698:\tlearn: 14066081.2809070\ttotal: 13.8s\tremaining: 5.95s\n", + "699:\tlearn: 14065657.9655402\ttotal: 13.8s\tremaining: 5.93s\n", + "700:\tlearn: 14065056.3116360\ttotal: 13.9s\tremaining: 5.91s\n", + "701:\tlearn: 14063060.5479925\ttotal: 13.9s\tremaining: 5.89s\n", + "702:\tlearn: 14055416.6213175\ttotal: 13.9s\tremaining: 5.87s\n", + "703:\tlearn: 14054788.4400664\ttotal: 13.9s\tremaining: 5.85s\n", + "704:\tlearn: 14054567.1177084\ttotal: 13.9s\tremaining: 5.83s\n", + "705:\tlearn: 14047841.2592422\ttotal: 13.9s\tremaining: 5.81s\n", + "706:\tlearn: 14041942.7637161\ttotal: 14s\tremaining: 5.79s\n", + "707:\tlearn: 14032003.1913976\ttotal: 14s\tremaining: 5.77s\n", + "708:\tlearn: 14031541.3957558\ttotal: 14s\tremaining: 5.75s\n", + "709:\tlearn: 14031381.0858530\ttotal: 14s\tremaining: 5.72s\n", + "710:\tlearn: 14031307.4373230\ttotal: 14s\tremaining: 5.7s\n", + "711:\tlearn: 14031104.5109335\ttotal: 14s\tremaining: 5.68s\n", + "712:\tlearn: 14031079.4384099\ttotal: 14.1s\tremaining: 5.66s\n", + "713:\tlearn: 14025939.8399957\ttotal: 14.1s\tremaining: 5.64s\n", + "714:\tlearn: 14025576.6473374\ttotal: 14.1s\tremaining: 5.62s\n", + "715:\tlearn: 14021357.5621778\ttotal: 14.1s\tremaining: 5.6s\n", + "716:\tlearn: 14010970.1733229\ttotal: 14.1s\tremaining: 5.58s\n", + "717:\tlearn: 14005353.8685640\ttotal: 14.2s\tremaining: 5.56s\n", + "718:\tlearn: 14002594.5921397\ttotal: 14.2s\tremaining: 5.54s\n", + "719:\tlearn: 13996198.0652146\ttotal: 14.2s\tremaining: 5.52s\n", + "720:\tlearn: 13995827.6286278\ttotal: 14.2s\tremaining: 5.5s\n", + "721:\tlearn: 13995625.1482069\ttotal: 14.2s\tremaining: 5.48s\n", + "722:\tlearn: 13988390.1767698\ttotal: 14.3s\tremaining: 5.46s\n", + "723:\tlearn: 13988033.4042234\ttotal: 14.3s\tremaining: 5.44s\n", + "724:\tlearn: 13987705.1114549\ttotal: 14.3s\tremaining: 5.42s\n", + "725:\tlearn: 13987572.7496949\ttotal: 14.3s\tremaining: 5.4s\n", + "726:\tlearn: 13975549.2973251\ttotal: 14.3s\tremaining: 5.38s\n", + "727:\tlearn: 13975497.6102345\ttotal: 14.3s\tremaining: 5.36s\n", + "728:\tlearn: 13974565.7941638\ttotal: 14.4s\tremaining: 5.34s\n", + "729:\tlearn: 13974089.3687881\ttotal: 14.4s\tremaining: 5.32s\n", + "730:\tlearn: 13973918.3603598\ttotal: 14.4s\tremaining: 5.3s\n", + "731:\tlearn: 13973640.7407537\ttotal: 14.4s\tremaining: 5.28s\n", + "732:\tlearn: 13973544.9664352\ttotal: 14.4s\tremaining: 5.26s\n", + "733:\tlearn: 13973335.5730618\ttotal: 14.5s\tremaining: 5.24s\n", + "734:\tlearn: 13973314.6886931\ttotal: 14.5s\tremaining: 5.22s\n", + "735:\tlearn: 13973147.6337343\ttotal: 14.5s\tremaining: 5.2s\n", + "736:\tlearn: 13971840.2578089\ttotal: 14.5s\tremaining: 5.18s\n", + "737:\tlearn: 13971217.9164376\ttotal: 14.5s\tremaining: 5.16s\n", + "738:\tlearn: 13971169.2519933\ttotal: 14.6s\tremaining: 5.14s\n", + "739:\tlearn: 13970958.9393172\ttotal: 14.6s\tremaining: 5.12s\n", + "740:\tlearn: 13970812.3031235\ttotal: 14.6s\tremaining: 5.1s\n", + "741:\tlearn: 13970614.0036168\ttotal: 14.6s\tremaining: 5.08s\n", + "742:\tlearn: 13970368.1657139\ttotal: 14.6s\tremaining: 5.06s\n", + "743:\tlearn: 13969808.4623390\ttotal: 14.6s\tremaining: 5.04s\n", + "744:\tlearn: 13969753.8877122\ttotal: 14.7s\tremaining: 5.02s\n", + "745:\tlearn: 13964356.3635191\ttotal: 14.7s\tremaining: 5s\n", + "746:\tlearn: 13964236.8387923\ttotal: 14.7s\tremaining: 4.98s\n", + "747:\tlearn: 13960803.9188307\ttotal: 14.7s\tremaining: 4.96s\n", + "748:\tlearn: 13960533.2698482\ttotal: 14.7s\tremaining: 4.94s\n", + "749:\tlearn: 13936556.5617582\ttotal: 14.8s\tremaining: 4.92s\n", + "750:\tlearn: 13930527.6037931\ttotal: 14.8s\tremaining: 4.9s\n", + "751:\tlearn: 13908519.4055456\ttotal: 14.8s\tremaining: 4.88s\n", + "752:\tlearn: 13908076.7827156\ttotal: 14.8s\tremaining: 4.86s\n", + "753:\tlearn: 13896619.8772339\ttotal: 14.8s\tremaining: 4.84s\n", + "754:\tlearn: 13896469.8030620\ttotal: 14.9s\tremaining: 4.82s\n", + "755:\tlearn: 13895917.6733027\ttotal: 14.9s\tremaining: 4.8s\n", + "756:\tlearn: 13895914.1925891\ttotal: 14.9s\tremaining: 4.78s\n", + "757:\tlearn: 13895906.5141022\ttotal: 14.9s\tremaining: 4.76s\n", + "758:\tlearn: 13894680.8452164\ttotal: 14.9s\tremaining: 4.74s\n", + "759:\tlearn: 13893900.4048769\ttotal: 14.9s\tremaining: 4.72s\n", + "760:\tlearn: 13893090.1903054\ttotal: 15s\tremaining: 4.7s\n", + "761:\tlearn: 13886422.7563719\ttotal: 15s\tremaining: 4.68s\n", + "762:\tlearn: 13882542.2101709\ttotal: 15s\tremaining: 4.66s\n", + "763:\tlearn: 13876917.9849237\ttotal: 15s\tremaining: 4.64s\n", + "764:\tlearn: 13869154.1279738\ttotal: 15s\tremaining: 4.62s\n", + "765:\tlearn: 13865721.0256688\ttotal: 15.1s\tremaining: 4.6s\n", + "766:\tlearn: 13865325.5531724\ttotal: 15.1s\tremaining: 4.58s\n", + "767:\tlearn: 13864488.0985036\ttotal: 15.1s\tremaining: 4.57s\n", + "768:\tlearn: 13862722.3729616\ttotal: 15.1s\tremaining: 4.55s\n", + "769:\tlearn: 13861558.2772508\ttotal: 15.2s\tremaining: 4.53s\n", + "770:\tlearn: 13855644.6008357\ttotal: 15.2s\tremaining: 4.51s\n", + "771:\tlearn: 13844736.5245748\ttotal: 15.2s\tremaining: 4.49s\n", + "772:\tlearn: 13844053.5084590\ttotal: 15.2s\tremaining: 4.47s\n", + "773:\tlearn: 13843767.6841598\ttotal: 15.2s\tremaining: 4.45s\n", + "774:\tlearn: 13843660.0878431\ttotal: 15.3s\tremaining: 4.43s\n", + "775:\tlearn: 13843519.5656229\ttotal: 15.3s\tremaining: 4.41s\n", + "776:\tlearn: 13842165.9056632\ttotal: 15.3s\tremaining: 4.39s\n", + "777:\tlearn: 13842058.1043917\ttotal: 15.3s\tremaining: 4.37s\n", + "778:\tlearn: 13834683.3005966\ttotal: 15.3s\tremaining: 4.34s\n", + "779:\tlearn: 13824330.7648540\ttotal: 15.3s\tremaining: 4.33s\n", + "780:\tlearn: 13821449.5219745\ttotal: 15.4s\tremaining: 4.31s\n", + "781:\tlearn: 13821357.5962736\ttotal: 15.4s\tremaining: 4.28s\n", + "782:\tlearn: 13814359.1324883\ttotal: 15.4s\tremaining: 4.26s\n", + "783:\tlearn: 13814227.2826089\ttotal: 15.4s\tremaining: 4.24s\n", + "784:\tlearn: 13808325.0086331\ttotal: 15.4s\tremaining: 4.22s\n", + "785:\tlearn: 13808274.8826495\ttotal: 15.4s\tremaining: 4.2s\n", + "786:\tlearn: 13801638.2244459\ttotal: 15.5s\tremaining: 4.18s\n", + "787:\tlearn: 13801419.2751533\ttotal: 15.5s\tremaining: 4.16s\n", + "788:\tlearn: 13801306.5570793\ttotal: 15.5s\tremaining: 4.14s\n", + "789:\tlearn: 13800184.3721073\ttotal: 15.5s\tremaining: 4.12s\n", + "790:\tlearn: 13790356.3064221\ttotal: 15.5s\tremaining: 4.1s\n", + "791:\tlearn: 13790143.8154680\ttotal: 15.6s\tremaining: 4.08s\n", + "792:\tlearn: 13789781.0560736\ttotal: 15.6s\tremaining: 4.06s\n", + "793:\tlearn: 13788151.5387251\ttotal: 15.6s\tremaining: 4.04s\n", + "794:\tlearn: 13787953.6654068\ttotal: 15.6s\tremaining: 4.02s\n", + "795:\tlearn: 13785685.4062468\ttotal: 15.6s\tremaining: 4s\n", + "796:\tlearn: 13784520.1734342\ttotal: 15.7s\tremaining: 3.99s\n", + "797:\tlearn: 13780517.1557671\ttotal: 15.7s\tremaining: 3.97s\n", + "798:\tlearn: 13779711.5747369\ttotal: 15.7s\tremaining: 3.95s\n", + "799:\tlearn: 13778313.2436650\ttotal: 15.7s\tremaining: 3.93s\n", + "800:\tlearn: 13777901.2822639\ttotal: 15.7s\tremaining: 3.91s\n", + "801:\tlearn: 13769185.5397209\ttotal: 15.7s\tremaining: 3.89s\n", + "802:\tlearn: 13769019.3830993\ttotal: 15.8s\tremaining: 3.87s\n", + "803:\tlearn: 13768538.0466150\ttotal: 15.8s\tremaining: 3.85s\n", + "804:\tlearn: 13768454.8685936\ttotal: 15.8s\tremaining: 3.83s\n", + "805:\tlearn: 13768380.6476226\ttotal: 15.8s\tremaining: 3.81s\n", + "806:\tlearn: 13768059.6021701\ttotal: 15.8s\tremaining: 3.79s\n", + "807:\tlearn: 13767932.4589060\ttotal: 15.9s\tremaining: 3.77s\n", + "808:\tlearn: 13767739.9474389\ttotal: 15.9s\tremaining: 3.75s\n", + "809:\tlearn: 13766687.9264105\ttotal: 15.9s\tremaining: 3.73s\n", + "810:\tlearn: 13760378.6609238\ttotal: 15.9s\tremaining: 3.71s\n", + "811:\tlearn: 13756092.6317829\ttotal: 15.9s\tremaining: 3.69s\n", + "812:\tlearn: 13756015.2398990\ttotal: 16s\tremaining: 3.67s\n", + "813:\tlearn: 13755919.0526178\ttotal: 16s\tremaining: 3.65s\n", + "814:\tlearn: 13755811.1990879\ttotal: 16s\tremaining: 3.63s\n", + "815:\tlearn: 13755526.6973763\ttotal: 16s\tremaining: 3.61s\n", + "816:\tlearn: 13755241.5375202\ttotal: 16s\tremaining: 3.59s\n", + "817:\tlearn: 13754829.4717828\ttotal: 16s\tremaining: 3.57s\n", + "818:\tlearn: 13750314.9701585\ttotal: 16.1s\tremaining: 3.55s\n", + "819:\tlearn: 13750081.0895962\ttotal: 16.1s\tremaining: 3.53s\n", + "820:\tlearn: 13750052.2455171\ttotal: 16.1s\tremaining: 3.51s\n", + "821:\tlearn: 13749379.8505211\ttotal: 16.1s\tremaining: 3.49s\n", + "822:\tlearn: 13744872.5921679\ttotal: 16.1s\tremaining: 3.47s\n", + "823:\tlearn: 13744473.2684358\ttotal: 16.2s\tremaining: 3.45s\n", + "824:\tlearn: 13743675.2045582\ttotal: 16.2s\tremaining: 3.43s\n", + "825:\tlearn: 13742643.2577464\ttotal: 16.2s\tremaining: 3.41s\n", + "826:\tlearn: 13742458.8752724\ttotal: 16.2s\tremaining: 3.39s\n", + "827:\tlearn: 13742240.6771652\ttotal: 16.2s\tremaining: 3.37s\n", + "828:\tlearn: 13742155.8504755\ttotal: 16.2s\tremaining: 3.35s\n", + "829:\tlearn: 13736167.4467943\ttotal: 16.3s\tremaining: 3.33s\n", + "830:\tlearn: 13735996.0708658\ttotal: 16.3s\tremaining: 3.31s\n", + "831:\tlearn: 13735493.2005146\ttotal: 16.3s\tremaining: 3.29s\n", + "832:\tlearn: 13733689.1766800\ttotal: 16.3s\tremaining: 3.27s\n", + "833:\tlearn: 13733607.5051327\ttotal: 16.3s\tremaining: 3.25s\n", + "834:\tlearn: 13721414.9907570\ttotal: 16.4s\tremaining: 3.23s\n", + "835:\tlearn: 13717127.3635539\ttotal: 16.4s\tremaining: 3.21s\n", + "836:\tlearn: 13716999.1069396\ttotal: 16.4s\tremaining: 3.19s\n", + "837:\tlearn: 13716913.2399601\ttotal: 16.4s\tremaining: 3.17s\n", + "838:\tlearn: 13711226.9169250\ttotal: 16.4s\tremaining: 3.15s\n", + "839:\tlearn: 13710970.6941366\ttotal: 16.4s\tremaining: 3.13s\n", + "840:\tlearn: 13709877.0723952\ttotal: 16.5s\tremaining: 3.11s\n", + "841:\tlearn: 13709787.1005052\ttotal: 16.5s\tremaining: 3.09s\n", + "842:\tlearn: 13709650.6948000\ttotal: 16.5s\tremaining: 3.07s\n", + "843:\tlearn: 13709188.1202366\ttotal: 16.5s\tremaining: 3.05s\n", + "844:\tlearn: 13709037.6576260\ttotal: 16.5s\tremaining: 3.03s\n", + "845:\tlearn: 13708902.2827845\ttotal: 16.5s\tremaining: 3.01s\n", + "846:\tlearn: 13708778.6197114\ttotal: 16.6s\tremaining: 2.99s\n", + "847:\tlearn: 13703386.7859640\ttotal: 16.6s\tremaining: 2.97s\n", + "848:\tlearn: 13703267.9377232\ttotal: 16.6s\tremaining: 2.95s\n", + "849:\tlearn: 13702980.6317410\ttotal: 16.6s\tremaining: 2.93s\n", + "850:\tlearn: 13701940.0387833\ttotal: 16.6s\tremaining: 2.91s\n", + "851:\tlearn: 13701787.1153741\ttotal: 16.6s\tremaining: 2.89s\n", + "852:\tlearn: 13701336.8187285\ttotal: 16.7s\tremaining: 2.87s\n", + "853:\tlearn: 13701128.5982232\ttotal: 16.7s\tremaining: 2.85s\n", + "854:\tlearn: 13700214.2512716\ttotal: 16.7s\tremaining: 2.83s\n", + "855:\tlearn: 13700030.8591834\ttotal: 16.7s\tremaining: 2.81s\n", + "856:\tlearn: 13699030.3116141\ttotal: 16.8s\tremaining: 2.79s\n", + "857:\tlearn: 13698969.4703703\ttotal: 16.8s\tremaining: 2.77s\n", + "858:\tlearn: 13698909.5329623\ttotal: 16.8s\tremaining: 2.75s\n", + "859:\tlearn: 13694731.7280367\ttotal: 16.8s\tremaining: 2.73s\n", + "860:\tlearn: 13694626.1354617\ttotal: 16.8s\tremaining: 2.71s\n", + "861:\tlearn: 13694549.7530144\ttotal: 16.8s\tremaining: 2.69s\n", + "862:\tlearn: 13685158.2690592\ttotal: 16.9s\tremaining: 2.68s\n", + "863:\tlearn: 13664600.4370607\ttotal: 16.9s\tremaining: 2.66s\n", + "864:\tlearn: 13658903.4964855\ttotal: 16.9s\tremaining: 2.64s\n", + "865:\tlearn: 13658772.1453303\ttotal: 16.9s\tremaining: 2.62s\n", + "866:\tlearn: 13658700.0253906\ttotal: 16.9s\tremaining: 2.6s\n", + "867:\tlearn: 13658541.3184384\ttotal: 17s\tremaining: 2.58s\n", + "868:\tlearn: 13657593.4117992\ttotal: 17s\tremaining: 2.56s\n", + "869:\tlearn: 13657046.9219992\ttotal: 17s\tremaining: 2.54s\n", + "870:\tlearn: 13656760.4817945\ttotal: 17s\tremaining: 2.52s\n", + "871:\tlearn: 13656606.4015105\ttotal: 17s\tremaining: 2.5s\n", + "872:\tlearn: 13655695.7827993\ttotal: 17.1s\tremaining: 2.48s\n", + "873:\tlearn: 13655316.8101662\ttotal: 17.1s\tremaining: 2.46s\n", + "874:\tlearn: 13655037.2038031\ttotal: 17.1s\tremaining: 2.44s\n", + "875:\tlearn: 13654796.5930907\ttotal: 17.1s\tremaining: 2.42s\n", + "876:\tlearn: 13654519.8205448\ttotal: 17.1s\tremaining: 2.4s\n", + "877:\tlearn: 13652603.4478047\ttotal: 17.2s\tremaining: 2.38s\n", + "878:\tlearn: 13648499.0930775\ttotal: 17.2s\tremaining: 2.37s\n", + "879:\tlearn: 13648337.7075388\ttotal: 17.2s\tremaining: 2.35s\n", + "880:\tlearn: 13648157.2472955\ttotal: 17.2s\tremaining: 2.33s\n", + "881:\tlearn: 13647331.0259756\ttotal: 17.2s\tremaining: 2.31s\n", + "882:\tlearn: 13642530.0968617\ttotal: 17.3s\tremaining: 2.29s\n", + "883:\tlearn: 13623623.8704151\ttotal: 17.3s\tremaining: 2.27s\n", + "884:\tlearn: 13616321.6923828\ttotal: 17.3s\tremaining: 2.25s\n", + "885:\tlearn: 13616254.5151889\ttotal: 17.3s\tremaining: 2.23s\n", + "886:\tlearn: 13615914.1338129\ttotal: 17.3s\tremaining: 2.21s\n", + "887:\tlearn: 13606970.2289736\ttotal: 17.4s\tremaining: 2.19s\n", + "888:\tlearn: 13606728.1952106\ttotal: 17.4s\tremaining: 2.17s\n", + "889:\tlearn: 13606589.0295901\ttotal: 17.4s\tremaining: 2.15s\n", + "890:\tlearn: 13601534.7404156\ttotal: 17.4s\tremaining: 2.13s\n", + "891:\tlearn: 13601203.9114680\ttotal: 17.4s\tremaining: 2.11s\n", + "892:\tlearn: 13600921.2715171\ttotal: 17.5s\tremaining: 2.09s\n", + "893:\tlearn: 13594610.9357134\ttotal: 17.5s\tremaining: 2.07s\n", + "894:\tlearn: 13593918.9554990\ttotal: 17.5s\tremaining: 2.05s\n", + "895:\tlearn: 13593623.3583576\ttotal: 17.5s\tremaining: 2.03s\n", + "896:\tlearn: 13592786.5943511\ttotal: 17.5s\tremaining: 2.01s\n", + "897:\tlearn: 13592494.0020987\ttotal: 17.5s\tremaining: 1.99s\n", + "898:\tlearn: 13592321.0874097\ttotal: 17.6s\tremaining: 1.97s\n", + "899:\tlearn: 13581909.8355361\ttotal: 17.6s\tremaining: 1.95s\n", + "900:\tlearn: 13581714.6855455\ttotal: 17.6s\tremaining: 1.93s\n", + "901:\tlearn: 13581583.2702309\ttotal: 17.6s\tremaining: 1.91s\n", + "902:\tlearn: 13577447.8482879\ttotal: 17.6s\tremaining: 1.89s\n", + "903:\tlearn: 13572604.7633084\ttotal: 17.7s\tremaining: 1.88s\n", + "904:\tlearn: 13572521.9981344\ttotal: 17.7s\tremaining: 1.85s\n", + "905:\tlearn: 13572394.7665040\ttotal: 17.7s\tremaining: 1.84s\n", + "906:\tlearn: 13567645.7843907\ttotal: 17.7s\tremaining: 1.82s\n", + "907:\tlearn: 13567146.9327071\ttotal: 17.7s\tremaining: 1.8s\n", + "908:\tlearn: 13566540.1488621\ttotal: 17.8s\tremaining: 1.78s\n", + "909:\tlearn: 13566264.5900869\ttotal: 17.8s\tremaining: 1.76s\n", + "910:\tlearn: 13566147.7496118\ttotal: 17.8s\tremaining: 1.74s\n", + "911:\tlearn: 13565954.9762823\ttotal: 17.8s\tremaining: 1.72s\n", + "912:\tlearn: 13565722.8330228\ttotal: 17.8s\tremaining: 1.7s\n", + "913:\tlearn: 13565374.1058795\ttotal: 17.8s\tremaining: 1.68s\n", + "914:\tlearn: 13565339.9876957\ttotal: 17.9s\tremaining: 1.66s\n", + "915:\tlearn: 13564999.4851646\ttotal: 17.9s\tremaining: 1.64s\n", + "916:\tlearn: 13564553.2714396\ttotal: 17.9s\tremaining: 1.62s\n", + "917:\tlearn: 13564493.0894889\ttotal: 17.9s\tremaining: 1.6s\n", + "918:\tlearn: 13557832.3311930\ttotal: 17.9s\tremaining: 1.58s\n", + "919:\tlearn: 13557604.1444962\ttotal: 17.9s\tremaining: 1.56s\n", + "920:\tlearn: 13549415.3576224\ttotal: 18s\tremaining: 1.54s\n", + "921:\tlearn: 13549296.1319348\ttotal: 18s\tremaining: 1.52s\n", + "922:\tlearn: 13549065.8874742\ttotal: 18s\tremaining: 1.5s\n", + "923:\tlearn: 13549013.2277371\ttotal: 18s\tremaining: 1.48s\n", + "924:\tlearn: 13541252.3964380\ttotal: 18s\tremaining: 1.46s\n", + "925:\tlearn: 13541118.4393823\ttotal: 18.1s\tremaining: 1.44s\n", + "926:\tlearn: 13523659.8134394\ttotal: 18.1s\tremaining: 1.42s\n", + "927:\tlearn: 13523372.6123983\ttotal: 18.1s\tremaining: 1.4s\n", + "928:\tlearn: 13523298.9960587\ttotal: 18.1s\tremaining: 1.38s\n", + "929:\tlearn: 13523229.7847989\ttotal: 18.1s\tremaining: 1.36s\n", + "930:\tlearn: 13523163.3403551\ttotal: 18.1s\tremaining: 1.34s\n", + "931:\tlearn: 13519155.8624765\ttotal: 18.2s\tremaining: 1.32s\n", + "932:\tlearn: 13519105.3541888\ttotal: 18.2s\tremaining: 1.3s\n", + "933:\tlearn: 13519031.1159000\ttotal: 18.2s\tremaining: 1.28s\n", + "934:\tlearn: 13518973.8153038\ttotal: 18.2s\tremaining: 1.27s\n", + "935:\tlearn: 13518376.6585356\ttotal: 18.2s\tremaining: 1.25s\n", + "936:\tlearn: 13518172.8356066\ttotal: 18.2s\tremaining: 1.23s\n", + "937:\tlearn: 13517992.8884415\ttotal: 18.3s\tremaining: 1.21s\n", + "938:\tlearn: 13517941.2098579\ttotal: 18.3s\tremaining: 1.19s\n", + "939:\tlearn: 13517200.6339096\ttotal: 18.3s\tremaining: 1.17s\n", + "940:\tlearn: 13513047.5537990\ttotal: 18.3s\tremaining: 1.15s\n", + "941:\tlearn: 13512950.9711099\ttotal: 18.3s\tremaining: 1.13s\n", + "942:\tlearn: 13512678.2375099\ttotal: 18.4s\tremaining: 1.11s\n", + "943:\tlearn: 13501004.8881561\ttotal: 18.4s\tremaining: 1.09s\n", + "944:\tlearn: 13500891.3365531\ttotal: 18.4s\tremaining: 1.07s\n", + "945:\tlearn: 13495622.4796461\ttotal: 18.4s\tremaining: 1.05s\n", + "946:\tlearn: 13491170.4852445\ttotal: 18.4s\tremaining: 1.03s\n", + "947:\tlearn: 13475101.4449889\ttotal: 18.4s\tremaining: 1.01s\n", + "948:\tlearn: 13463999.2463230\ttotal: 18.5s\tremaining: 993ms\n", + "949:\tlearn: 13463491.9256000\ttotal: 18.5s\tremaining: 973ms\n", + "950:\tlearn: 13463000.2524334\ttotal: 18.5s\tremaining: 954ms\n", + "951:\tlearn: 13462656.6373331\ttotal: 18.5s\tremaining: 934ms\n", + "952:\tlearn: 13462372.8551029\ttotal: 18.6s\tremaining: 915ms\n", + "953:\tlearn: 13460952.4213064\ttotal: 18.6s\tremaining: 896ms\n", + "954:\tlearn: 13460850.8235406\ttotal: 18.6s\tremaining: 876ms\n", + "955:\tlearn: 13460696.8538462\ttotal: 18.6s\tremaining: 856ms\n", + "956:\tlearn: 13457056.2735812\ttotal: 18.6s\tremaining: 837ms\n", + "957:\tlearn: 13456515.4174941\ttotal: 18.6s\tremaining: 817ms\n", + "958:\tlearn: 13454202.7159913\ttotal: 18.7s\tremaining: 798ms\n", + "959:\tlearn: 13449065.3874411\ttotal: 18.7s\tremaining: 779ms\n", + "960:\tlearn: 13445971.9860801\ttotal: 18.7s\tremaining: 759ms\n", + "961:\tlearn: 13431187.4696091\ttotal: 18.7s\tremaining: 740ms\n", + "962:\tlearn: 13420640.2806388\ttotal: 18.7s\tremaining: 720ms\n", + "963:\tlearn: 13420343.2454284\ttotal: 18.8s\tremaining: 701ms\n", + "964:\tlearn: 13418335.8156443\ttotal: 18.8s\tremaining: 681ms\n", + "965:\tlearn: 13412369.4847919\ttotal: 18.8s\tremaining: 662ms\n", + "966:\tlearn: 13402367.9503594\ttotal: 18.8s\tremaining: 642ms\n", + "967:\tlearn: 13402321.3484318\ttotal: 18.8s\tremaining: 623ms\n", + "968:\tlearn: 13402179.0832179\ttotal: 18.9s\tremaining: 603ms\n", + "969:\tlearn: 13397225.1329227\ttotal: 18.9s\tremaining: 584ms\n", + "970:\tlearn: 13393750.7282482\ttotal: 18.9s\tremaining: 564ms\n", + "971:\tlearn: 13390423.6410648\ttotal: 18.9s\tremaining: 545ms\n", + "972:\tlearn: 13376811.1247023\ttotal: 18.9s\tremaining: 525ms\n", + "973:\tlearn: 13376327.5071527\ttotal: 19s\tremaining: 506ms\n", + "974:\tlearn: 13375464.0838350\ttotal: 19s\tremaining: 487ms\n", + "975:\tlearn: 13375290.8335452\ttotal: 19s\tremaining: 467ms\n", + "976:\tlearn: 13374990.1920422\ttotal: 19s\tremaining: 447ms\n", + "977:\tlearn: 13374808.2609258\ttotal: 19s\tremaining: 428ms\n", + "978:\tlearn: 13367352.9513837\ttotal: 19s\tremaining: 409ms\n", + "979:\tlearn: 13362655.1953826\ttotal: 19.1s\tremaining: 389ms\n", + "980:\tlearn: 13362591.9016145\ttotal: 19.1s\tremaining: 369ms\n", + "981:\tlearn: 13362424.3720459\ttotal: 19.1s\tremaining: 350ms\n", + "982:\tlearn: 13362142.5062367\ttotal: 19.1s\tremaining: 331ms\n", + "983:\tlearn: 13361843.8288514\ttotal: 19.1s\tremaining: 311ms\n", + "984:\tlearn: 13361292.0256243\ttotal: 19.1s\tremaining: 292ms\n", + "985:\tlearn: 13361195.2903051\ttotal: 19.2s\tremaining: 272ms\n", + "986:\tlearn: 13361068.4898725\ttotal: 19.2s\tremaining: 253ms\n", + "987:\tlearn: 13353507.7997423\ttotal: 19.2s\tremaining: 233ms\n", + "988:\tlearn: 13349051.5645266\ttotal: 19.2s\tremaining: 214ms\n", + "989:\tlearn: 13344467.2156917\ttotal: 19.2s\tremaining: 194ms\n", + "990:\tlearn: 13344375.0678171\ttotal: 19.3s\tremaining: 175ms\n", + "991:\tlearn: 13343966.5070671\ttotal: 19.3s\tremaining: 156ms\n", + "992:\tlearn: 13343619.3277021\ttotal: 19.3s\tremaining: 136ms\n", + "993:\tlearn: 13343484.5481961\ttotal: 19.3s\tremaining: 117ms\n", + "994:\tlearn: 13343227.6871102\ttotal: 19.3s\tremaining: 97.2ms\n", + "995:\tlearn: 13343170.0970924\ttotal: 19.4s\tremaining: 77.7ms\n", + "996:\tlearn: 13343011.6160519\ttotal: 19.4s\tremaining: 58.3ms\n", + "997:\tlearn: 13330480.8328314\ttotal: 19.4s\tremaining: 38.9ms\n", + "998:\tlearn: 13330241.9625622\ttotal: 19.4s\tremaining: 19.4ms\n", + "999:\tlearn: 13329412.2161376\ttotal: 19.4s\tremaining: 0us\n" + ] + }, + { + "data": { + "text/html": [ + "
Pipeline(steps=[('preprocessor',\n",
+       "                 ColumnTransformer(transformers=[('num', StandardScaler(),\n",
+       "                                                  ['geo_lat', 'geo_lon',\n",
+       "                                                   'level', 'levels', 'rooms',\n",
+       "                                                   'area', 'kitchen_area']),\n",
+       "                                                 ('cat',\n",
+       "                                                  OrdinalEncoder(handle_unknown='use_encoded_value',\n",
+       "                                                                 unknown_value=99999999),\n",
+       "                                                  ['region', 'building_type',\n",
+       "                                                   'object_type']),\n",
+       "                                                 ('afr',\n",
+       "                                                  Pipeline(steps=[('autofeat',\n",
+       "                                                                   <__main__.AutoFeatWrapper object at 0x7448995df580>),\n",
+       "                                                                  ('scaler',\n",
+       "                                                                   StandardScaler())]),\n",
+       "                                                  ['geo_lat', 'geo_lon',\n",
+       "                                                   'level', 'levels', 'rooms',\n",
+       "                                                   'area', 'kitchen_area'])])),\n",
+       "                ('model',\n",
+       "                 <catboost.core.CatBoostRegressor object at 0x7448bd575f60>)])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ], + "text/plain": [ + "Pipeline(steps=[('preprocessor',\n", + " ColumnTransformer(transformers=[('num', StandardScaler(),\n", + " ['geo_lat', 'geo_lon',\n", + " 'level', 'levels', 'rooms',\n", + " 'area', 'kitchen_area']),\n", + " ('cat',\n", + " OrdinalEncoder(handle_unknown='use_encoded_value',\n", + " unknown_value=99999999),\n", + " ['region', 'building_type',\n", + " 'object_type']),\n", + " ('afr',\n", + " Pipeline(steps=[('autofeat',\n", + " <__main__.AutoFeatWrapper object at 0x7448995df580>),\n", + " ('scaler',\n", + " StandardScaler())]),\n", + " ['geo_lat', 'geo_lon',\n", + " 'level', 'levels', 'rooms',\n", + " 'area', 'kitchen_area'])])),\n", + " ('model',\n", + " )])" + ] + }, + "execution_count": 73, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "pipeline_afr = Pipeline(steps=[('preprocessor', preprocessor_afr), \n", + " ('model', regressor)])\n", + "\n", + "pipeline_afr.fit(X_train, y_train)\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 75, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "{'mae': 1381270.9152113453,\n", + " 'mape': 1.97323058892439e+18,\n", + " 'mse': 262290784020649.34}" + ] + }, + "execution_count": 75, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "predictions = pipeline_afr.predict(X_test) \n", + "\n", + "metrics = {}\n", + "metrics[\"mae\"] = mean_absolute_error(y_test, predictions) \n", + "metrics[\"mape\"] = mean_absolute_percentage_error(y_test, predictions)\n", + "metrics[\"mse\"] = mean_squared_error(y_test, predictions)\n", + "\n", + "metrics" + ] + }, + { + "cell_type": "code", + "execution_count": 76, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024/10/10 14:53:08 INFO mlflow.tracking._tracking_service.client: 🏃 View run autofeat at: http://127.0.0.1:5000/#/experiments/1/runs/65961cc40d284389b693dcaf1c25a5cd.\n", + "2024/10/10 14:53:08 INFO mlflow.tracking._tracking_service.client: 🧪 View experiment at: http://127.0.0.1:5000/#/experiments/1.\n" + ] + } + ], + "source": [ + "\n", + "experiment_id = mlflow.get_experiment_by_name(EXPERIMENT_NAME).experiment_id\n", + "\n", + "with mlflow.start_run(run_name='autofeat', experiment_id=experiment_id) as run:\n", + " # получаем уникальный идентификатор запуска эксперимента\n", + " run_id = run.info.run_id \n", + " mlflow.sklearn.log_model(pipeline_afr, \n", + " artifact_path=\"models\",\n", + " signature=signature,\n", + " input_example=input_example,\n", + " pip_requirements=req_file\n", + " )\n", + " mlflow.log_metrics(metrics)\n", + " mlflow.log_artifact(art)\n", + " mlflow.log_params(pipeline_afr.get_params())\n", + "\n", + "run = mlflow.get_run(run_id) \n", + "assert (run.info.status =='FINISHED')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] } ], "metadata": { diff --git a/lectures/lec6-feature_engineering.odp b/lectures/lec6-feature_engineering.odp new file mode 100644 index 0000000000000000000000000000000000000000..c302e8aad8c8be5d25ec4a654a7446cf153d16b7 GIT binary patch literal 127631 zcmdqHWl$d7wl#{oyK90=g1fuBySuv++}$05LvVL@cP9{n6Wrx?-hK9|d#iSxTj$UB zzwKcVKVQ_FVbq0R_xkusu(<#u;e;+u& znXtW`nWed_(?7R5Gcmf@+uIl!Ix+t5TAA6{8@ibOZ(9LVb^V{m!ut<1bg*}Db@Q zp@Z}Pa=M1b#-=u=zjn{!>sr2S^h?nCvm zUy=6TTf?>17xu4}98)bvSQq3}9vCXE!Bu6 z6Ezsa7#oO2A8YwkFM(7IXYj&1;Zhx}NjK7eX0#VmqLw0C4MqtDO0FL;ieU@Ca4oCb z5&KXt(Y$!&u}R(*?E)u?--e`X2`3#|tCjbH|19Gl!A6jk6y1B&+}eBOCZ;tR@koMZ z)ZN&kBO^ev*^Hav$(bI>96N3qQAj2!^O-mO^SvuvTCVIld^MuZ3yFh6>T#lxc?O=7 z#inUAENV?v3ZzPLm?h2|U9D}41eW5G7xLQY-~^Y0fF=B@VE7gTW-PQHq0zm)qLSE) zM1}Fan9WfUavZ&;*C`I_t5m`BWXS}`$Mm-|t@A$n7F?NWRQqM_flOl`47tTLxG*zr z>1@Vv)xB3C)uhGK8nK>@R0Q-^oDOy^=-~bI5Vse~Y^U<1nOCJHke~21>>h7nRRcb? z3H$Hc<+$%a=Vkh9xap-6ff381L?g)du^(=+%Myp$JF>78Y>cTySQg`6&eoDB6Kw~h zYM4m)P}a|a=wt*6L$s-Hm6{{XJCK<#y{+e?A8CVNm#nC)D5NMCbQ|_}xgbsFl-=;P zGVwZxjwid@1vxMlJ(ETo+<8!7ynxVJR^4pd`>%LjZEtl&!s3Hhs+ZI_a_Khm`+l?K zWZEk^C+V0Y?V*A{Wn9s(_Sf-md0%-ByHnTA#AHmEg#%9)Obo?}vDn%bVsY&L`q_Vv z&LklbA?kH3xi54HJ_n|tfM%SrO=yU-gngj;G!oIwaUKGSEw>g`Rz5vLp5-9v&fl&_ zcgojD}{OIv*Q!0BX1uBwEs+L+N=0e7NP*9>rO|${#DVsBCa6 z03i(B+5iKj&}tyCgbE0gw5ptj15W_Oc&FN5>;pD1 zp$ID;!Iq~f@+{O2XWhVuisU^a{qf;E&t8AWFW2-lh+6Q8ZaXah@}1|3#CoIMqrRpd zb4mOmt1*xJUQ^~$2yQ0?=*$tfAc(q4YJb$7JE@V_F%_Zg@e!P?xb$y$%;L7hNc6+Z zYnmJ=+b^tXc4zQL%E~+oqy}$o#;&-wuiDphg?JDN;6h6NIFOg0B}TgVzsQ)8mBya_ z7BPjJQ7fZ?eO@%&7li$g<37j=l*F505v{K+5@1S0Mes$pS0%(3kg{rj-!y5c?76M7 zkW|z9ZpZS%1yB4Z)q)Sp!V(Px-F8IrgOGN7aw66oypwK3k4Y#}n|!;m{5X_E5aPTS z#c3A#YEXCkA(+6OZ)BbcVjivuMjg$PCzKX0wXDNs0>be|!=Sd+;f+q)k1SZy@2FGg zNoyxgA~OndC)E(4_rweVpBz^n09!YVZD>L4@5Q6O{%!D_qIj5sF z-F_GHjja)5DEyB<7OVv{Htb_L+*$OLnmtdoSxA~<(5o6(aDs+nUNLO#fm_uz6m`12 zUf)gve;B)aE@^VIZAa297kZb9DWa1ok`KU*<1u8;w18}bF_%xF7)Yz=W#hDb>(|Ib zM;d3%{$*gvo)9VNjKfhwvYkQx+11SAh+j7}fw_y6=_rfMO+{KF#z9?WsyF8+6UCRv z+Pr+K8OVk3FSQw{91Wgql*UfsT@hiXcI$)SN2?M$%n@i>&S4~pKN@gMneE}2ammOb zKY~nxJUDbICo|U($M(i_93XY1=wwrWz!xG-4IW#Z9L?vkNcR)%-duciziN|(>U_@r zl*j}>lBDDO%?S+_hXl&Y)Gaato#Kjdwza!eQzm}{v(8Y{dXat9XSs5(5ejJ?i6fp< zBr|{-0R;T#Ey_H;utc$wpA*a-zKDfOA`ET}>G!d>BoT{Q9HD2)J!TuEJoKi8qG6Pi zys{i3_V8KDbxT!C1hRcYTq_3d?UB3jX5$n)s71DQ@To5YF7~MQ#CsymHKQf-UD9my zYt4gWgK=M=8+lw`^HsiO3-cf*-^$618G@GF8`OT;YD}dP$TIbkI2J^PA=N$6BVp|~ z4ff8l0d1q}mt{fTbz)01TEW|~%B5<*Vz-de!7-WWp%#2_Z*C#AJ)v?-on;x;#IkBO zpLFaLA;(pSLyd#2w8KQ^uVhq{L@9`A4!++ay81d-K|C6?jjUPiV6YiDID2 zRlq}LKnc3d@q06*98i#luE0jy^r2*cm-_`)1w{?5c|{44WBVvuqqMlQJj^< zumuO6eKC(IxKy$^e`n;~W01{)tNA5&$uEv;K_4l;`kdS^(Tu-Kko%wap9QB{K(4T;DHf7px~PrVzv`mM;e zY-)y-un^`7WY6J{po&#`uEQKu;O$SPe@DbdzT&YBoP`Dj(fcfd8wqaMh`QZMD<^K+ z-Hg+CSnh^}w02|L*s=yX5vorlQ2PO+lNLOiGLZb46GO>|3x0zvVhN;xfva&PmKdBQ1pFqYo7MfBjh=Ybm52GkM!jh;pE&o1tXEptjy!gZQ!-h4 z9*C_RQZJAcjL9B(9A00>vzlJwVe1qS@E%q1q7x#D;~|7q)*T|fUF$8! z6M~iVnaWiq#1ZRcZU0b1=qS`a$uQ;d{0@jhlaOQEH5v#Qp%Cv9$0yR)*AY_?8gG&k z+N@s*jpq%ilN{CPaUkN@md@ssRWn>lGacvHT-eH(;wDd-Ti+*#Pu#pxn>y{WcP8)W zKne~W@v0Q7j1?u~Rl%j{R&9tuAe80t9$;GBtj>hSs`3rcW(`8`Vx8aY%KRCz9bs{y zH}&}G%JM=jrh*`TvivSE!ArELg=Xz z?5DU%%OX`S+i~RBal%iolg|FoHv}Di&h9xL>ZXQfC(V_ZS(=k|+f;s9vah#WSEuSf z)zdS4aV*7Oodh4af0nKc%5M<%WS_RotLD1n{|V2Vqw;&pd@M&M>Jcn@#EnDhF)VsS zBqH=H61&3Z<9Z#A*<<(Oc?_A54E__c^-`dVw(1ljt!1e((O+lV@>@`6{`4NT<6j?g zPCIMjx{kR0=KsF*U;p5@^)_%s4f7m~{X$gpdR(+~zTk3RN1)nxufZ_(DRZ4@;()}H zZo~O#BGS^X--FWKD`bMOb0WOv2x8;I* z_An`a?7GJzLYmL<+2O(=?lBU4?@XoE+a%0MjvAdeGHr_vHn{S>zdJ0Z(}l6AH7Jt= z^PD;VOv{@5O2z4{T*HlRNx(E7B#IcQVILdlp7tAQh^iP3{b~?#mf8M&H1o&2rFjcq znT4J53-rYx1&W1crBxWz9!kzC1lS$5uh^x6fXzyTUK8KaBKw_Ed9~f-9rX{JOoq6H zujV?(^#uqmf~oKkwD{Wzd}t>@733z6f#-e4i*-gFp=usi@~^8yUfMzj3w z3hj_As#j4Wv(0cAd9;3D`F!n7IVuf2kygiuJrzdBC`K&nLv`10!{UZpbB zo?btTPNI5z+3 z?s1HKN9*2Hrq;uRH=f&>1)k(noy%YWFIj>j<{lqk(s^Li@9WVXKg&kmxaIJL|4xsi zBOS6$aczU$eVllirGlXMqeF1TexaALq!389{zjDRXD#VDwuICzSs-EAEZFn8SjYuq zIqU6CUKHYhn61v&0AbeI`{Mk1rxd+FL5y+9p`unOV{Y6zzW23QX4C^Mt~MjCh+b2t zPZsUFK;ibDToTiqPg}Z2$-dh>5td<}cyjEoE00cG1*_3<64;X=X#9-PLqq+d!nVUy zRoUgGIl;_j3{RGeKQ+hTV15#Vw4`}Z5-0GD0V4bqhRhjXh|nrYx5w(LDxS?K9dBjw zPQUlxCA{($9i>`_{`5=>jb%Fc^p@Fs>XCW1KSr`J_lp2!3RMrZuecp|Pi{lvjE!OB z{eqc1rFm2O>n7xk-@ZqMg*ZE2ct9q^=yIQGO^o&Yq_03Fp_EEfLRGmKZX-+Vlixu~ zbDg=u14+%&sMF7#m?(PsT(-+wS#~e$x0~ml)4b)K&2B$utga!D-{H-@J?f49hNrn6 z31L~It=bJ=lU1Yen>-wg+?dJp1k)(#{T3yBR#}DBZjzEHkQ=}9UoiQf!_vm?WeXjXN{VXEN@Kx6702Q zPl>Uixk+-WU-)&OeEfA_*b+>-xu8ks6ZT_&=?AoeG&GEB4~D)wDhLSAGziGQG`IgG z*T4gEjj_F*%U^BpAH8PFd1DyW|Brs4Aw`CV6&M87Nl;y~R0o27ai+7GZt#$Sxb7e9kQ1mgoB7V62oWnkSq z+&g-Bd=K7i7f{&B_fICZs>h6qC={ekGwtrwtb4jWtGji){m&`YHCiew8>_#% z8YoRvR!&~;mV#`iZXj9V8do-TAFkO-8nHEfQ%d0Em9%70`@Y+Tb&Vp;u5JxJV#U07 zVz)s?;kuE#F1B2jcORbWL1LH%kAu-(VuxW1Vkv005UR!nCCg^ay$C6^?KAf~k!P$~ zWyRL(E0ngj_MTeOssD{S+VT=_`?(~l>&zUBF&bWJG8Go{lg zEiEltO^igxdXg1?s1#_*IA3>6YW-a1~pIkTG?g%tZk6Tt& z*cbU&VHoHU@7jn*(f)j+lMfr`92es?M9HxyL(-Inl}ZPhBrD2 zZVk3MSAoZ6F_Y!UrsC8nWg10+Msa%#k_Lk zySO!~d3~*&aBp#Jsj|96H7u7%*OIeCMF7P1($nNQCwO}F^{B=nbA{K1ij%CnZP1wj z+W=4YWNEy^aiL_7Gy;1Z%5NsE2N&iO)UUDvEQpN7T%BZ=&&d|dEZpeh582TTw**EH zuB^-4K5`WVDguGfB~ZvKJyc7kP=CrM$_)2*=d<}2!6`oU--~|cy@7vNw7!(jKQ8)1 z^za|_@ZLSV9=;d!{2_h{>-_XMSf6(JXN^U6w#-KsOgGUhkMkKD!8w8J>n)MY51OXO zL=xGSh|QDm@kK*W8DW|f+{yAR8Spws#Bx+|_{!+Oa_!a%ood$Y*fIKX$Q`mNJlU38R0c^b z=SBi|Z%9||FLPUtON`~3sMkmQ^TZgY|1{GZ!5bvkfKvuT(hsuW?F%A{ik*VFL;&e= z^v$8UYGCgfN!-Bzf1unuhY8PY^WFrnJ!o3yb6Y}td!RQjl-7PV!-q1yy1!_07W|E3 zk7mp>76f7o+SWnIs!n#&HN1J>{X$Uu$Qgo&w2`vi6m)^w5KIU`=tar#%7JGf<;0$n za4xhLdwn1CN%{Tw%XDaykuSN{;4!EH)cM{mq{L^q9FV~>j0EC5l_I09`TS4ctA&bf z%B0FL&didpa@6fVY&<-d*a9XYS}OH3VC#SGVrK51_1Ahh4anhq4SDIxfTdd1+Wc(K zwS}+mZbXJmy;OK!oLZ@&L-|9ls#9t+iS5WuLsw@~=8T%&Oed6}l9kZuVNR$p;MT_s z*7_!WVcm3X|G74B@7Ww|DdKJijv23wsE{dSzHCRtsN zGv%6dlTwurop^$3hpNVBUuFEzAy4H=Mun$oLwoXTgrm~P%s#`#vmyKtsRkD!#I+)!nx_AN^v zQc`ELrKFS9)gCD*F=F$IuZ47OAcz9xxi=v(_`;Q#%Ap)}YmFVuk}ANGxLs#RQ=`~d zE+MkCxfkQ2whT848aAxoTP`0uW`D4yJ4egl^(kSHea0^Q4NDq1txYKm!Mx-|i(EK2 zwoI)62W_W>S={UvgJ!?05U9nwRk%1;y$v&B8U3#R_V)9K|5?xF!|q|h$09$$cIct6 z|BOGZ|3fG;IxGxlhbqHoQf za6Kv5f(6k#!{!GGLXlLI7>U>-ZFR|qZG8!E8dS^%7e~y-E)$lM#>HPZq%`-%Fkq&W zQL1!;cR53L%N@Ue^XQC^QnzN)lg7?A(=X3me!}gL^batxxiq2j_dJLjD5gm2;k5-})T3_9$H`m%Gr?X?$=wF* zNQ8!X5WKTK6*m}_FSo>jdM#H<`Iih->T;+CJiAGopazf)#4hME@h`KdSC`Wc?vK0k zrWijpdf(@7ev&Q{kXBHEN_-&4R!!TDO@ z=pEQ8?9q~isFfyFpo<;qFQ}Ztj4I`XU_=;Td69oe7^D&S;w-6O=;8w+i-JddwGUbA z&-;qrPO0@+|D|(;5bt-5(}frIaIdaGcO1CaqkzEVBgiV@HfDa7G0Hhe%^)1lQpe1n z;ll8^xxhN%-QNyDm#W@a>(|<|{=^U*;q8HS+)J4oD50td<7wj{q^}My#HH+qtCzkM zmlH!XeJT$E^Wf2`FBaF}z2~IJYy)9|ZV6_eM_zt@azRNEW+AhGBtd;H;liZ4{M_;| zu@GY4_krAwz=1w;%fLme!i)2?N{DiX^Er~NP=VwzRY%rNH3p+=bK-3Yvjq?rsLKk74z21H`~9FiQJJ$is$>tM4sQ6ET!NoFTNOb@a@9 zBI7boQ&+(QJJ#}ouqWNrOQhbW4}g#a8=l^w>nv zNpC+=^7xxzo2a5X=f{)+dA*uOlNv&GV`@@~EQ_+J*^9yMSD9Psm~_kLfa9TOA$R)( zKJZ``5V|wL%vg6meNUOJ8tes!KhwsO4zvT9Jwf?d78)uUoZ#~jbk2g#DhgvX!h92z z(Z7diicVCA`-f=ujZz#h)N!#T7ieWNw>xcoQPs#kev_lL5FtOMZVNiJ*T>;!3wgkN z;~s1+er4&#cL?8DYDH|PMfwxrQ#Uqpo5ZBOvigR)QzfIvy&JfT@_-%p?T?$p%Fhd% z-%4N?Hv6%*wb)NzxW2A2Qo>(C;Xl0Igg7u0MZ|L3hPW{u9X2UwO*(Jl%HXp_b7l7K zUU%NCCrQ0efY$MBNsKXcQupr|495=1}{pqGoj569T`H16G*KZwNs+?ZZ#%_BS z{h>#ru8)deUA&T3F+)r3o5kS3@80P$ zMmfIf%=||8;d0hS#Eb1_HcNe`+cFsUBA-T;2f^>3mqe2b8uaA9ANxqw?%v1P zcNLlG+m5ub;lQCx!}f_GiAtmqjNlxtkle)G+v=RKC}|AG`#b4 zlKzO4Vob5Ed&1^1Z@WX4lmn%r6u=oG%~efY17D0kR*N^?46T2 zU7uY!ED0U5cCvg&(SPh8&8MJrrUtma$|X#&es&0}!Bvk&GMp-H}fv9`%(wMj@@oMSUIO$JAJ9U8kgDngGTAY~G zuY{$BzFB1BL>7V5Fk@rnqaswOzEo5cqCJ4lw`sV_RjklHnaveFQF$8-Te?4#9DH~T zwT8=I^KwrV%cq}?BL@2J5}#qDdu-d}aoq2f{@t;VTGq#3PpZp^@!*xw+WKGlF|+&Z z;~^3VNQv#g^W%SSkC8q~VRLr@1^N4B1@dDBOJf&TV50`(*DwFCgISd0pTC$mnK=YS zn3-9=ii&;_Vqz9!V`Xr#GmlV^6Gw!@`@0k5MA{$F>LfycL^|EmWP!hpe?f3+hy z0r_&?JtO|LQQD3nYU3zXn_c{lB^ED-RQa`jye>!a0wle9p`oEQ4c~w_Cc+y6yu{qN7<_iUc9(L|C@kSH zB;6J}T@4M5EH+!OyOZLYnjV9`Z;Oj6(9mIHNtFB5W@&?igP62xqX*Xvy6rCK>#EYy zr&H=L?{6<1Zs$@`QqRA$?BU?x+|E`L5)!W79(J@fHF+)5wmRHvw3@8?gJCtWqK3xC z#&j4~rl!sjm!JLf9~3;jy}fO1&((7|za$X7;^}G~xs30@cnUQ&H3I{K@o;oG-GxS*$ru_POo&jpcZ>?Y+|vYOkBO-% zVCdc%85t9)^t9B}PmlX?d7m!aFE%rHz3<~mg;9cgt&NS1`9EH7X82yJN(On^+1;0` zG^4QSom^ae4l*W8>i0+Dh`C%~Y=c`*kB>>wXc-vHN|3BSZf5wi*zLEA#N#Lr0vLou z0!a?94<>-6dbqzwf`RckxcdZH?+GVh&{sYh+S>b=QX~6UB%T^<%*^{Yw0u%B%c5k8 zDb^`Zs z{%?0H>iWKS-@91yrLQ8ZK{5U>?SG6K9W!T$yBIBYt`E9jEsynk&%&r@nJ4@ zN~bXx)p;zG$gVyNAoCAqHdq-M8yildxbyS(P$~6%G~1Z?W6rGUyGs*y=E@NlZ*k zqLe3_8VEy1JrwYnBb~}*75*T>&5Yoh$>qh@4dx*}JrIv25KJpvh@T%BDFxj1h%A=t z{psaT&m6OIshsGe zW(fDscPsjHhvDpK=;){> zqraiLy4tWyBlJU4ON(zTfH+Eo>9w1G{r=rgm#|c4Jd)prPASLZxCc=` zBt=O{NwweRe3)L{FV)r6l>tYK0jHv>t=%CQ1ch=$sfrJV7!w-Wzs`BmgOeXqYq*%I z3mG=#2UuXKeBQ-+6K#KJdOA)z!>jAh{*d9}VTC8a?}YmP?DXWt$M2q;XsM_KtK}<( zgoFU^Jj|^|vjK`Inq?KvQ9P9Y?G_yaw@<7{EM^v*+x z+EW`D*Wd8x!SxR&AmB26m(5~}H>TQ6|7r{%^ugZVwT?Skf3;yott1tzTNs(JbAi{x z#a2{Ql<+4I*_~EL@=vbQ*_;4&MMp+{OAt0O+(2=+Xo-k`O~z2#t~Zr_WN-^iBm$7K zF9^D9{_$#mYI3sH^>n!!07?L?f(gi{+;@D{6%=|I3K=zOL6~d}4X;-5`ZX2R20{_n zv_&9M)Kso>+_(F@ij_QH?rm{A9ULy}O(*g@*Vfi()qeyxArtYm0*-wK#-CVVu22>j z2>R*@%+ca;v(0&pfEAz3MgX7lL;$8{rAA-B#f~r-$7!cf6uylATNo@l18l-yMN2LKn|u zvce|}j3$up_&(^eKyQ6}U7IRjw*stAEE)%um>3!U&q~eC(OeO#$L<&1RtHitvi_Z& zyp&iOg-QSd8^XbUB2GhB+3a+`B@J_+0x@$C{x3C|0o|MyNA%ukb==1uh46y54wDrM z>2Uvp^flz(^M4z5DTaBgwl$L-Bc0@@5Q^7~e=yNax= zEVJOT%bx>th0cf5b!2v%tqG~A_vbA;&9)nrJ7A;*0I9;K)E4F75Eb5gi1uh3x+%eRN+8b;JE8;ij}BOqHPqp@J1sI)E^9g zt_Bn2zm!P=WWZq;1?P*SxQ+`DjswA=$%qU%$;E{1M+OH;)`2(;1Ufy^zeY{axhfMI z-#>1Des-qSVuxtX>--zx7SA3Pq}N?nM~7rB-i(7m+;-;eq zYs?Ocikf=v)579+sirw~ji5^d0Mc#H=gHFrpMjNUvzP_p4TWp5SyO6&?|{*CU}NM@L7m&!;t~D>Xe1Kl_3B@@KORJSYrepkMum4y>RF8!Ynj!a`8; z0C~Y~yqvD?77$(3)z#VX5uA*%J2R4z@!1ewE;iek>FCaGj~7M>JfR>Vr}FrBMn(?X zgpf2Il^HMrSQQ}ztb(3_fr!NdKKKpN`!1qv8eou)`+Lc7>wMk z(kEIvxyeOi<=37qtuLt?XWl)mqX` zLf+TQpTZ(KK=d4;?XXWjh~}BYazWxc@nrsM2H~ZDeX0h*+WHiDIjPCX0Ex-S%CgpZ z0>DT{Y^@a)BFh_Pp5Q;?MX)iw)*8@S^_jQFvl?Rer{VEoAL4S{JT*#eSLl^Di|a! z&v|!>sBlmQoz{Tb0-(ccfHW^x;UeJBD&7L;&-dpq-(M7n;o(BwtQ$MvYLyoCKcOMw zd*S`Q3AR;bq4@5y48aOshJw3mh-?D7i-h<(&W@cs@ zFv!Sxyxz?sWC>Em4tz%(_=`wSL5n9xN4KXd_ID>su&}V?;;}X>KM39U{a!ihGnq^y zv$C{JOib=dSurs%@)g*uf9>`K+1lEYUZH5lYFXvE3td4W;`Ms5+8b7u;S93#^Akgg ziiiLtyy!YyfXn?NJ2SK0d?ptlr^1GcWLEGEHBPD<+uPleNfhwd44WF0h-HqDnD zZg&R_U0q$ylnXKehi@i;D=eA!kpA_kw7`F$#G&a=&xar2A1FrvR}mUXv8p9vdpbzd zudhSz@_q%aMYb zf~#Ygl`(B$2mrQQ-M-pX7-aHikkw~9Js%A^jXw}RWlf0?p`i4dEDjApHb>%--9G|D z0rA82_DHeQ^rlN-2bq8agfXly4xjx32wF8vQV^Ip+-^3CigT&-+VKqu38-1p(5W*g zP}t(RN4&u-Ku)z+CX6aCKLbP%q2#M(Aj+$$sT~b)I{(f$W@T@4S{z2W(W^HZ1D@gS zxHPXmY&3GnNJOM+WEJP3v=KWrrq|%*{+xpa29xjQtYJJ5`Xr)vKS`dik5qV}-Sspy zMYqFkNKp)VpZ*2^n;AAiNPmC-?5qMC=s#*jC(U1zTv-vautJUJu&EXn(;O6FLCaX0Rw}}|iz7MeQ+%88B_h)MW z9X&hU-v`nUe2Q=nc+Swd=cgyj$k2-Wlch?_g%U$6l2{2Wz@2Z*ym>9vGDf~-quL8h zu{uvQ+V21!FvZEqNd=RYzc><0D7>hsr+4w^&rXpzGR)h2vGk7TAz@vn;Xvq5Rc2BI zYw}KlYB^b1QKPev(O&@cXLcB%{;1MvyIuxzBD>A_F%v4hJ9A@$T2=rlK*2@yX;SLH zO5ka9%#QWEKc525i_(&BfdSpJJXscmre3WzWHxES!NF0b*M-+Tt;mn)etR^x3ON^B zP(TWX(8$igk-qggx-gAiJ5+CIt`F3-gPW66@B!5?pSHqFV&CP7vRt)^sw?JxIl{|$pmgq<$D_&Mw3a#nx>Y5_)X_XMz zoXETT=KFFVf?-u}DheMWWbxStW4K0v${-^%ESfK^$$2p;KMv=s(f)7@kZ1vY4tFbT z0BK!)7WbM7bqpS>&|3oYYl`m7u=XI4QyamUMq8CarlOhOl%2=mtbxKmdHI}WARRxOo{3R3z^kjOP`YZj62xv)=*~6(ulc3# zsV5D=7Vj#HIA_nHN9TXP!BslH!}S*(L7^5>}URJTZ77gf)%bFY6j&}$iJIzY9lN1C#KsbTSGPJ2k z6ga}+?*Ht305DV+fsv=;FMCj@;MppjB4#dg#%^H!E zm^i<%z|73dZnL_3c(?*Y=yC8h)EGK|&nzywjK-5zIghRb!dwkRG@Or#W)A!@_eRh$So=_|j+x#VAlbB;xg|(x}^O+3`DFEJw$; zXBLPt+bMk*yKy5qEa$zad%Iu#n}ooyP3Q5?P>h-*=P!w%0=ag#|HnHJ2vQkT~dXXES0Dj=rs0MiCoKad>5!^5xAcj)MbL9_rf z0Z?&zV#3+o-8{&IB_%F8+GaMtZ-)N^AO~Z$trjWn{pOS8Bol9T?pAybs)EjQ?3Vej|H* z#*VYtR`NIXa0UB6rG&2NB}*0SFsM^u++qx?gFQD~-<(9GLA;I4?Kzb5N2Cc9?_GNE z1PTZc%K-1kJaDk9yk4bi-Xk{1n@BQwhABiE1n=zAJ2_=E~EB)$uN5sTz)ftN?QOLTuxQw#u zqqTagsVxJcm;d!rxK2Z7VX6EkA9*H+^YF;tWTPTMr=;xo1jQAtn~m|^B)L!5xZdsW z#Dal|6;@ZD|N4c2gJWTBU8xX{-^Eeu79pafp^*+4c8P2jpu1%LP|PW9{C>O3|3b(^ zm*x79AoYt?m#|QdZ1HCrmV`^at5rMM-vxj|dsvT>Lc~9oQ{U7_lO?X-fU;`mawWPg zkQ!2?QH8s$%$Aee6D{NLOy-EQD8mHYx#2It1KR#={=ZeSYgUp|Qc_Y=GqbYP{cH-P zipR&s`tYj!n}$*WR z=SQrBw)P`U53?znr;kWz2rp7P6?<2BQ^>^99;^qLZSIfm%Q|n#aZ}tYPp?oewbi7AE-?6=@S8m!|r|@ z|MK$kk4{S<#cytI1{58?kn1qAANoB+SXj$Hbh&{H*`nPH9l;e~Jx@nKtzEPvoR`0k!3}`Cq7OUHp94%y0d#?Z=!UP;34>ezo1<;{Ayes}Jp_9Z6_(HXg`%0>HcDC1iDDX zi;Ef1{m;V&yY#?8yF9M;#EE@XOnKE}2{^BpQX#850c1VjXwh!6QhgI*-P9#PSxlMm zi?;z%v&Kf2(YjQ;K~lh29335N-O5$CfShT5UWw%imszCh+n3#%o_C>KL_8J%3<3R{ z-{-M5D{CWPFbEJqa!oHhUC7;nk5(djv^o)_LMlx4^BOw-u$>;i@jD(`5pKB@O@Z{> zwsvAqs4*x9=fzFPGU7d{ic3T9zo;iQPx2}zYp+*{3nv81Y3W?t6(OJVOuon>HQ5^! zcpR^t-Y`qMz4>$>c97aav0|l3+7tL5?JUchd{X}7({W0H#L5a3GL-qFitTi6lg3Qm-`@dE zQNe8yH>TbD;R2ceb+*gfy=r7*Z_lVeAzV(_9s>&t4+rPhuU`U3uCA_gfns7}DbB)3 z5fKrzYE=RGjC!5vmRwx-fVc)I&}}|X^+`$SQX+guUroN*B#=Jq=<5@A6ww{==_&q* z#glF3fa4H5((V-Xg21;AT2Rj>r*Zb)ckcLeV2W{k^1(e@N!4=JcW08Pa5y;#7Tf-` zr)25nL!?)1)D|nTF+8-mDx^=7IEPOPpW&eh6>Vf;%Av(T^J152Qb zgt1@E3}&O~k@g0KwDf6*-!Y9UflZqc8mVb|(L7UCh*jCLZ94R4P8R1&{hjsGtz_2@_JD{c~>iY%;N0ZFc&8g5D^I(WX8Iyol@maR}Bn|iI*Omk#5$O zn1^Yf`6p8&4up6?$>4m+buq<1)2Er1Vqux3Nqu%Ii0vI-S{&8+#E<~d=-%Ub{WBxy zXRKZ-5+WUblT*lG-WqD?wm{YO8MwVwWMU#(_Rkq2G=R)?n$E5eU$GeYSua-w>fz(z z^;7)J0qRbUuC5`}fD8rS4jVD(B`!V!i0zNh&)eO;-hg=Juz=5Q_g4|NmXemPG>_WZ zN={A&e6j&_?eVi16NaMtV2-Yld6DldxW!O~YGwWq0ENQZC@rsm4hW+K)NkGZ8G?x0 zjqb}A6fj){1(5T<6gE+H8j<$ld?PI<=k4h*Pq+s4wYKsRP?W@TVL-+wCLBosVn|F# z7}FB&pdeylK~*nA$e`C5PNslp(_awSXN)U|Aru0r$UyziVUeAggt;iUk~mQUref7?4=Q8qtZ`FK5{4Bda5sU8LMAfb1O$=^*%yPuY3Y=52!l7=R8P%7jj-NJ`* zwKFg(C-$U&8cfX^dcqtip2hlLvm*?JN?*eh^3uM$(Pn><(6V6|?CKPm2SURW@5B7w zHA?W};@qqJC2DUe9V_eI^WA9>G_sh3_5v#z%(pUJiOmQ~FhK-lGI)WBi6M(p=kP+m@rf0vt)0W12uEbLEFhQbL@ z-9%{a0)_^{UbFp9_qC||E%WXA@Gunba8*tQrr5hc6>OaAmVx1$=Gg@zuj0YYK^MIEg=Tb7dH26(KPQQpYDGrY5uD`Ie`e^^>sa64E!m%4OR&D6vUgJk6R1%I*W4D1vTQ~Y)+ zq%V&PnAe@2HgR=Vbe<^dfhcqA*e2Rmd=X=7cz#0u-ou>thD*ucz`@@zIJhpbBhu0- zGLC&Ibv>bddvyitDt1<2P9+ktdVUe&Awnl?)_8~H*y+pld-0x|%j?^4^;YbMXXTet zn=|*EExe6wKHM(XImaqZ3+RlnoG9mq7F67m#fmr* zBVC`a@1>>t6VSwW{*R`!JMt)>eom+R2#KvONdz=yN zlvm_r3UWM~A0Cnzc&?ph#Hz*AyAye#fZbfSHg{DdI}s(3tWBalNp@c2aALRmmZF}( zQlnNstW@=@&Cu@&=FZzXI(*6T1oDmV?{U_Q5*0q-yU;9DEs%>IkTwyBzTt6HQ#B<- zdLKdaYIC34lSDFInsechxu2n(=Jm7sxTGVgUR-B^glx)@=!E_IBtA5y&==>p?>UNv zPxA-1cPdESiPUY(99#~aN_@cx9pOM`LMX$Z5m=^y&BCmzO?n8VRGd}PO7A#)qZ(6`FyaPQqTBulE~V@axO}g z{a$`ac(v$rfm7$kw)Ccw7q;Ja9&DN}yzKQH4VE&$fV#e$_Kt&{s4EvEj73n7DWPce zL8S-Ep!xL%y$EJ9}E#*;SUrnUZUf3_$V3dUZYO}8VfT?8HNbw*s z2WasbZ!Sh2HPlFX^6R3g1drOOhUS9 z*c}R?+i5LJdyGE}nsHq->!|u16nx8{2r~83eVho!x zWUrRvlxAjUv-r*wZD;Dr&C_`F(R9OW%fuhC9F9IDC`k^#({<(}T&biS+D_4@pr8OF zqD2rY*5sgD)uOE8c-6ZN*96mbAsieWON{uA?ru4{qjg*qjxdMGL85W_-pBXxMSqol z3dTq=DEp9WERE7QIW@IVSB62IDR%9eAD0oO!1DL$sQ>YMafhj`UzFIXQ-;Uub8?qS zcyX0pt28+NzZrLCu@qd4bHCKgi~i!!P<+?Zr^sAU#v>Qj`KXiWKFNl<&G2GAK2wxe zR+B_Zy8a{f@$xc`K-NZ{nnUXaTK3VYmNS>pK75!EbY~(K_;0Y4CJO7$N@5=COuotf zLF6!{B9-vA;ii#>SXxfWDC~(122t+fjlrDVX_UXElKr;H%VO)7d7;mdrYhn)Av)!C z^+A2Ms%F(4S1D7~`+sB!819%D_}Y?RR+tntx74_EJbc~J%fXa}lV8$O{x$DiOvs_V zp>95Ek{*ZurU?-?EgfBUcJ{(n(rB4Ti%>Y=lR_SQnFl@<5My^EvzxV zq!zu)@3`1rIb+?wx@`sg4Vk+Z0)x*w%s7y~7r(N#^=`IEfkXjULQxcU z+z`@!a&iJo-Wk^SzwK&LMdIYKeBF|QF-b|GD}Z{%zss0wVE>tpWi$nZwFJp{)SSm~ zwObWB1t2P|d9S3Ujg8PLBj?Zme+A!R86E2qZRw26x z@2^E`<&|#z`w3CFN^MCeV;}~1K{I<`SAkajsJ3jFf!|kD^o7L?=j6Tg|2|GJ$^@T+y zjn`QpDMlW0Q6;aFU?{R+(XujR*5sRNI{irSr?s8!Egqft6WlPZ_}klIbihC^s#~h@ zp2AW5O+R5Sd(9xx=7EM)(10Qv`_r){>xu7Fri0IY+&)uET(zv80)$)f4O=LEr6)^V7WUv4V!k@nFlA~ zkg%gHwg3SV6N9_V7ArG&9%cayC+7K2ITRJxirinmd}(gJ3bhg$2?-#H7`$|ZTE#^e zB|kIs%~b)4CxXNe#S_`HDO6M^b|ht@!wa7ybx)Ed*qPrCEvX?HT)***A^g=|*qUtj zDNHJ5X7qD@YF9{}7m1^%c34Y73aQ?H7z4UpXx43u8*G@-s+~E44t~xnhN6fsF4Qq5 zUTEEfOLz?C=VJO8=rcCgv3;(RS)3TYbFi~)M&ik}aWNE1(6$waN!|BiPIExjfmSskAwMD;hf@CY@&9Y8_G zX}3h`$hY5_N9ML`-i1@KNXIUOwB2iqrQ&SqbW2mW#6yV#iUa-DWH(dCLM_JKKbdE- z-!*eWm8B#l&CJY1uNE@lNd7gUaOwsYj4mWVyplg#Z;%xJk0%I?qTX7))}ym8y5J$9D0&xjH&btfZRUbKGl zI4St#fVKBYWMe9}wf)RA0cN@JYvHW^kX0`ODKygL4as>~z$GJk)>h5&X^JS9f+&QZ z^t@wux_b1sX?fkIn5PE!yw;VYe0UAh!@=an%?_23AV_tvM{1QeXW9Bt-D*s##O0XT zPOQv{v&NC@T|?|lMJb!K(932W95yu+vgpQH4QhPOdyeg_1Vl>-1QXi}Q*M;2a`hFW z4;l%ZeJS;o`W4w&nF-xVhjKJTj8G^o8x?53`7=D@cm`OJC+aks>pnh-=4Hp9{$y>N z3IOp+DBBMINMQ`~*sXM5VQ^FrS6t>STTyER_hQ^kml%JB+i(-^@ZDq(vRg;&V@^B~c+NjCHk>cV7 zW7x&5aPMZ+`A*zE&~&z9V`5-T-hqxKed$v{CI9f@mjbmf=BB30YiqCKs;jxkw6Gpf zfF!169Qx=~Nv=(UBZ!hEkZiuYCsf2d^s|elk$bDzP+uQOA;8VWMNn_d7Jr9I^1kch zXMxT6FG^IP+EOAVVh^YmPw9zZxTC|cdLre8+IoO@EYyf9R_F0LOIsD8TXNd3X>b&rE-8Pa{`2;Ac3mfdz&gM{tUiIck~Rs}pY6v7PSwW7pU*gL+zF2Pbf z9u+Ft+pBf9HZ5WvmP%N5lKlN6QiX~#x=20$&$u+Vp;A-E#;h*Veqv=;wvR zc(16pg@v^<`{LwRJgjXB=3j}0)nFUxvx+zB9-oRoUdhGtePtjo{Yrvpk-eb0a^p?3 zt{MHZY=_eWW@HYspokHL0$2lEoE{Ai^$d3pq1eYBOfR6Js}~z6@DbeoE4u(LpXZ zh3Ol{c572|GRa8?`?3Z-Ub%4+sZJ@gZn&3NRE86CVJxXz=p#Yo*}&()$8IZpR9t0t z5>;&`#EE#8=BvgOpztex95^L1PUqaNdozb|j_xKg$?HKbyh67raVku)o>yKk{KdpA zJ|eL@$0=jAU)j){m~gXviD2OAqUHXXKAI9HaiuC~m_KDjb;=O67OFoQGfL2HiDFmyja%XVBiL2An4+pFOD`W z>V@Jx82rtioI&2kef!8-cZN%y z+k@p_;X?jj+-9eh2u2?ULmrsK=6w#aj?xv`<#8o!5zZMyqNnB}A|g=Myi65#q$ygM zj_(Ecd3i|Azwn)nOVPrAsKt5DZTr+1)2(rH3l)!LaJ71KZEJSsZl5|Wc`S3P+uok- z2JPsI*0*6JvV4<0KP%p7vagCwPJ><0g%jpgGakTC@&Sz}kzDRDjQYQRNz=v-4jAsB zp~;0M9{(I2tqRBJ%Y?62rx}Jc8?%-OSyk*+Wb1TS`&`>4 zR@Kppr^ZrKxJBqi8nx9!gl@E9P*nseJkmZZqnSUz%cB9qyvLI+RAFb={wqta*kcWK z%A!24SGfpPMJdYvZ|Yo~ft7U^CJP1z2GHgLlLbt0AT`r}zM{`exw*N3h=Shub%wAn zm~Bu*FERf!8fh#Xp6~DPE2eN}d4{5YKKakY)cz_n^w)g}+aTe%YNC35KaeBs?Kubu z2_?uK&kNuxNfoGr1ciu@5LDV~S<08E*zIbfRE!UEm;d%f;Pgn$kGr^r@rNUrS-!RO z7CSD3`#^5}X_fgMN{a21`8}j5qaeSu{5<&wBslJO2Z=(V})tH4&u@ASuAgG;!^zi>nDp7YQjTXnNeh$pfa8>KjX0 z`Vad{zfwPb9CeXIU*!6aaJLVGD*$RS0wSUbKldcLT&H^dEalz|k&LVQ3U0c%uFI>d zP%JX3#TQKoLJtl}I~fwHzPdV!W9deOAew7#jY)Z3UeR;hQi}gOJRZ(v);k>ISlmy? z)E-;^JByo+j*{`7)e6nF%dvE{`bYH)eR1Xggw0_*ky*)~vcE7CtoCGEN>5(wP;D6g%VMS9B%t4Imse3$ ztpHU8kOEPlmHGbWHkU#Av?NF&!1M7J_`l6fey3#_adGhzI%r|La4;pMr2%CF0Ja9s z1&c2T>!2`}YXS2JntYJc0uo%_QdYJH0@EZ`134w7+KP%z7={5p;>^v-$w8=%fk#x! zQ+N%QCPmJwPl4Gw`r0YPab9cPR|o8lji36q(@M@ayPlO7+B?7P44c*#wvcS=-07?iBZM+8pSEm$=jLx{?fGI?`p;2X=C@RootvH(1#jKe z5nE#mS1miA|1$qc;3Cea!K<0rt;+d-ulqM&?|3QKNj4c))#XQL~V`PDYvt8ZJwIYFWNh51v3cqscWK<^NthYErDIfT^Rf~K3U+sF$APXcQLo9S(9mBn@me(u z9iz9rp>pLpSqx7VXFRbPp0PNuCHwV)tZXfM$-h)!n`r&@mTC3Ni$dc#Z+EI0O{E-b zOA}n8WPHCTg$B_vBL1`U_Ud2ma?Ib~v_z}6GqtgDaM*ebH`_HR(~vP%y!)?#To{%- z=Ef?6#^t|TuWc!QBvTmk6-Y7>MAkn1pI37~@Hy-3>?~NEJOM|5vjDZqib&bb+4&I8 zXWR6CO%Z9$Muz`23A-tJwAQmsBt1CHTBU}OD`3S_0~&Q;AUfOcC>#dsnrko~a+vpa z_Voq-sGQx|*;zP(4^9^JWdCryJNp~d@4xN8X*^ozpGrk0rB8ZH&5v|Dqo?%aye?A$ zR`R4cYr+(t4&{_*RL>GB=i&Isga%W*qTU(j-Q%D4qwl@L`bIgH=N{AeewxOnGDV&B}LwWUCT1ghxio@1h{Sp#i) z;3cgyb6lIiz^eJm+8SZK$K)x=R9Dw5wqcGsQ@W51%??3*KJU*w)H4xgp^4$2#UZH@ zqsYtbxoYA)t!cySdIglXYdn|-R}BO%zJFKO8c@zJ_hgaKe<9-+!+?7b`2>r!+Ufo# z2JfZ^7jpr%FfAPys}_xx3)zEW3+nbHrRC}z5Vu_T4jv3&hh6rXBuQ}#_ocs1kq`7x zg!;>d-nFRqCbO}=9>776=l9Ib`GIWXilP|qum`H>)m#yVfF4XaeVOtJ2XlR>$deXZmzVW^Fg(LY!ui3 zY~*~TMrY3w(~+eWH`X>+Vlvdwqm?4$t;hH}hx$%Rxk6irYdywV{yLVa6D;oTt<%Ia zy>b@6OiGZ0c5!!^J=h{9s7{^KpRElDp~e0Od%H$2Y={>v?W}Xu^PvoUaGGkE3Z*&S zjoj{*e^=Wpz_(n@Dc za!RcBVBxpF`%h}IOB=7x^W07GJQZs{QqPI~Pm7Gqp;vAqDJB*OgsrrYfdYWIV5I_C zaZoMTt$?3lG4u4~^XVTP#J%{Gnp)I#yZci{297TgMlj%T2y8kdBO^VXVjMg{U``d? zvd(#pB2B^8W$I*s(-@wCv%_7tbHN~RFY;K4vMvobo!!3Ag+RKfSY6P&17BM}b+PjW z#lcmwxYi6Q|AH+n!S&VQ{ENfgeZA6l7a1&np0W zUcJip{0+Kj;S7H$CE&dIK1oYU`>}2*31J~g-QL+TUIq!1Th#%P+Dr_-P_%e&ct*|AKiaFT z&loFfCX*Gztm}e-PqN-2LC?=qCVg?1Z(19IEl7#8`3W~YqPUku(Hc=9L|N51tyY{& zmm=H$C3CapxwHKQSHX!^+8|w`s4Jh^$*nh0Z*7dBx)*95+?N zd;7b(2em~+zC5KKwHP^=?JD1RVdsWqdp*YC-rdvQ|LJ7zdRL5MXA&)ZXPE}qJt5)& zS@LLc$+v)TncN(4*`~YH>i}ngG)Ug>BgN?0A*he$l`S~P=#qiJ3LoTSs7>e zC=UN(6D&2V``pI4vCD`seCC>=|Me;R0}?%Tq1WlYCF+~M^`?Zrx(GjLw}#=%Lv)V< z_)Ol-AC$Pt<=QMsJ>Xt@doYAMcOkE=90L+axFBii=`>s5Pla}A2OKB&@%Ge%7gTt7 zc;w_ho1+C>*5f`G8(*}+17MMZVBakIQW8FW61JZcv7gUGYi%tJ{5&)F>X&2 z!t{oiS1)g~*{}ayUAD7${mRW|z!S&FLSC=Zo9D9)1q+hi#YTinh|=!WTH%$jqF|sd zkC(&4li~1GqpM3AX~{(KCK`vY2WR?3MqAsU!`wHX&3ox`iR7?R^a9&XnX3umc6lk^??A6fS zmrt!eys?CWf2(SpD;5tuY6GSIXJyP`L4Khx`j;R^B{x;33fiNpc}17WI6NKOv7cXQKX{tY`tH|lR{9>B!= zF5ths&I>PA5c+$=bE`p1pLPGp`P^=-YI(A1p3si!X`Xx>VapS{7@}JIb`=Nx-6!ykqpMacFH$9f}NU-qT}EimiTN9~J>`OQ~|`uIH-JuwA5= zGC%%^YqDedAz9rqjm@g$x-2(aEi>8b&l>xNU$9KLsKM~0Wo&Wwy6T&oCkdmAXozeHOkLPS$ z1Lvh0h^l{VfLOs$!x5hIIWPrwyd&D*p}@mqx0!qjD6qxlXxo`e+{VZk1i!|n77G5$ zfsfCYV5rRCd_7#5V^PTna`L56gbMjAuXW%-xc3uqTzTC8be$_s`+3MkjmswY=a(s@ z;`Aq$4ey%G)GD>+-?AzD{MFLAi60K1i14zLPEZFie8~U*_L@orAndl1*fxWCq8$7-kwK+ z#9U@Ub9t_CB8Vc*i^)(#jrce0)4qR=w2|^18|wY}oSM_q-*Mp3dsp%AQL4Y*DNf=^ z6sFQQ5V3vp7oN9pvW&x~Bz0dqs8gKw-mCU}RB29jrQL4wZ2NqnEfiQC06eBZ9t2qv z#a@$s*J^-TKJ^t+LdwylIiu^|*O6nnJ+|06B1u^RTjFchF$n zyZ#=43~QL*&(}W^;j)9JajZxS4CQYIN||grm06Rr<+j895=0Zq?lzWTENMK#*Bn@o zy*wIG8a%pBCk2@nG&#VI@NRDO7jnkAM&+F!1)rcI>@pLQ38hc_es5xq&0nvzNZ3AS z)#BwTb1$hIAQrLhE|gY2{yQ(lI~GO#k0vKuwu!{Sk2K}o+ zb5C8HxN4uJ)n%dy8h&k&9BWtgABHbAit$@3pNSH$SiOt>K&oVxdVS9+|C>T#nd38C zq0Ge>UiIn=WSj4`g1wp|m9?I}x_Zg-3WDIBPrh5AFJE-~=@<0g;7Swi9;!7ah z%WoCcR8%lAGmD$R6r1m3Wo0EMh9q2BW(!mV2(FS9G_|$EP$5`B0%;O+1LS^iQi^yv zIfaCU{dzw>Q+DtntFoThg*pi?7Z+z|W;V88hTDbyu7|7n-x$g!V0TvGHwky2^OUX> z7807BnGxjSIR$_J#l-~{88_$Cr_X(rnMb9Sl-NKA#LkWY=)kkrZOAtOyaFf$+?eqT z<1;d%wb}_%_%jh+eP=hf{;sZoI`PxCwzk+TbtXkQxmGWAVj;Ethol9Zpq%ICCQPm? z6b&714Z2PgMc5wp1IxB{G5Ws|O(>R)oWbFDM=dl8l!(3%a50(k z@gur)hKRqgzUTg8hq4805;NdToxmvWfHU}&iOJ6>3QS`PFo1$_B$`4Xq~wWi)fpUE zu7QbH)X5ncMqm-38HeLrtVG?XJqcn(SQ@gLK+)*8ebn+$>C@Y=)6FqvK0Z(Ii<3J3 zk{hKO3mK>TT~_vVIhq&cTq%X~HB8@@lKT4O_k`)_Ud=UmBCKZ z#qxa)12fWPs(p)Snz8`Sl^*hcP~&l6GKYoOp#Syx8gTS(kn%G>{|$adgB6S~3kATL zf^kOB`@{ymovA6k!bDFGCM~8n!bMHUqWuT#9}5e{mX_!mfEc1wok8GDP>*4g$Nfev zLiYr@l`kR9l38%mq48j#qyNdLicbIV!Nk_~YO-hN<;HL>ED{9a622El$z0Z&aCEIY zJ@N0CfRzgp25?WITG9!g!FS5~1L7A;85kn3FAtT7PabCY1Y)@Vol75B%VtLkxYc8M z4I0*ub=lVi1qB&?7mth6BWJ_~D~;Pwiva>+Ani~wdL!*J1itzBRuvH~lq@7o+Wq*) zY7i4}1!*-eg$#`k!37Ix89Ua3eklL%;@qw)01$wlIK&r96$oWy>G7zjGJx^(gB}2) zLzo@NKHLo!d&J)E3!ZD(7+~H95OuN^$b^`U!?l4|L~hLTkW&JO;JpFZryXFN+uPfN zhXeYGX4+~iha8{F`XG#pW$=XGf`=$dcgFaqY5{(YL31GhXoFkeJjsp+OEF|5Ap%I? zA`H{N32cM7k||8sKtR~Y=XdF0Sqp0%Y3qS>9|*p}Bg+5DsSqwkzyO0tgP@~Beuf-~ z1z|%HY$?+f&-THyI0w-J5oNb~0A&`9Y3sI~YeEQy%J6QBtE$3m6t~x{GT`$5))Ew3 z`rT5p|8Va)zmK*9>d0du7UEWj6c(6PiJy=?Z1d!_CF)I>{TzV;1Hoc`FlE1is%oIm z%de}etDio7vLWK(;YsvAZq%X~0hH+qgcxSKeZfB4=;u+8Lo(EfR=BgW(!tHg&rkb6vEN3n8xZz^bOE9T zG<5-RcvBjgc_nY^hIFXk*K~ zwRQT_eqn9>G6bE&M-B^YJT{#Q(}U$Mq=3wZJT1?6mk5jFpFaq+WL~;mD&dv%kGNStF>+8QSp44ZB z9v~r?@38V%|2zD~!sp9z;6DaA59oMsFgr`lUNJ5tdls8b9s-<)y$aJ4S!< zRa4YW*3L`CT+GiCEO@MyZtSl6mE4>o-d{t##-B`N40Hgd42Jm?$XH?_d&sFN>vY?aosMwY*V21l0%G>8H%9l1e78Vu;1_TA@ zuY9$fZQ!ECLO}%xa&RV5moK2cp3me1R0JI2r?)&cc|3|mn}(`htC>1&3^U-n10#eZ zeD)`l%I_n)S$C1JOZ!a_f+_&zC+6nEoNjFUI!ZV0@!p@x^V&oghWg6mIYH93O2z}^)lFh`_8{Xv-qeVd;dI;9c0>Cj75FY9;^6QQhLD##a1(5D zC1^J$N;95QKYjKrKtueZNZ(GnB*q+SPs)_}M{ORfQS^Xc(48P*Cm;c#@HP_^5`q~6 zdg}z(RiK1MT|8E?oNoNv@{Sf)w->^#tdEY41O){-m5`54Po=j5642!-e``U4gMO38 zKBV)M6PX}$N~n?WHdIz}ynB?$Y7k+%{wz#bu2gs?%!A_p=o>&*^g{jZ+|PoWPk;JN zq$K3J~#-KC;qK8{-8Sf3kgBF!eB_4QWE$2d*x7=z8xK~?JIUkLls zcx+GniloUCN}|L8B8PGfeI^2uO=6uem!qniMpQ1ed&CjvC8YKG*hN2!0DCrP)_1$U zL|6;~TLV`j`SC6Hn7BA_oZZ6txtk_!#Y*@N23}#uMG4-bs3`2zrQCzDt)`|X2;%w+ zM#iuDXg1*6l)Nrm6S+P^CU{SS^p9fk8({ z=k?DuY91}E37B$9l(EuMOG1y}=~Xr|%4n9A~O(j>b0W&5X_=}4>_T$N3KJ)yeiLHVpP@vL$+NtL9RBBS@qkWgDQ~aE0)}& z)|z@8>?7KdFbc#pDYtX4>=SiI-q3OV$htIM=eCyu?#`4M?#G=$&i?aPCL;pi1pU@0 zxICS2GPi$VBCBGU?Cva08J!xZs;;WtzhcF;wJ}?jEp6NK(n#mg!mkuhGA_SfLXbH# zm#*c9Mm;qvwaMKc%F@pN5N)2U*mzCzkYG<^7$VZ?@=OcJ}rnlJ2H1cgOz5cgS&Yksr4ezF}kr6|`c)WJ`z@ z?^x@-MJ~|RdJm0{huA@Z!*>*!Ex2MY`+WmJh5HTm^TK_Cx7sm-gz?{XY=rYAMa{;_ zKQ5#!hjgs$5Y( zO)iAI@GWRogUOaAgKgQdib?8c+crp0CxvoS*TO~qlIwaa&TG5xJuCtLgO*>@&GPtf z(R8v$R^@df;Y^2L)s2kIu>4}-a4wu`J))nh4IWv* zn#e2J{yYCugH_|@Df5ls_Y0o+&pgpG^rXBFnSk}bd)Jf7M&Z`tups6Z+sK}eleVy- zVKto71q-KzwRO+et&y_C26WiV?@&E_@E{Im<1wIzL60Cnk86ML$XO>mM;6Wf8)1K! z%2YT`1sq*AiZI^#o$-f7;MpS4%)hviW=gEwqL2%QC|q5p*u!&up`ZxIF>K0J0=)z# zA#r0_QBSO@waB2a4(i#S{}`1F@E_u$16kftz@7MWa?;~$Z$YU5Z(L7X8fY(8^m*TSblygL_bJ;1mb?SHbO#dV1`m!boD|72GE)35OTq{;3W&&B~2()N(yglpGh61idoGaYKC)JJx^E|&v(~?~dxTs;!ftqbb-DkR=-G(%Vf9)IO?JV2s<^#5BX>I~nE~Us3Np z?0Z%|>Te+>Ag*@o6Q)aEN{t5oK{-#`jCyUxIN&V*MMIB(Dn)?N0p@iOJY_(hR1H%? zH$16eE-u%$`4W=tH~v~zm(&}#V3kh-1T4@mw6i_Q3s7z3WMv`rl>6B;5DZQX4H<8( zIPqba3^08t{Gn;uqhfRu6&VQ|9axj$!sip4LVy-Y)*V9dHUJKY(7tZM-J*V0XliF? z=*j4@974VqK_2sv=}jR7ozWJ7-Dg=xQ*(85)5^@OgCEu%Ojwu>0YX1Ix`%<$isbzU zAF8Yb4s|4K6%bAM)vhCs5OPI^9!5n)X@A!o?e6V;&lpsA|G(4Qa?3q5@bF9@R6wV< z1h{Ps3Y5Fe_-PJxmc8@t0^=?X4 zk%wihI&JAnt>9;`m!%-bhfr4IqB=_@_QXwuCsHC z`oLNSnR^1T(=`Q4{RNLMt)A~{EnM!uvyDVwd)E#tD=VRsmSHZpG&T7E5&=8FclL2K zG&Cvz8epbzc5wL4nvYsV$%<_d@|=$TBT(-W44D-kC1uOYA4u)UVOTK; zFb6~s0e#tCu*OTN--=t^_we)c0|??1aMj(dUZXQG#6pTI4azW}-`B7<<L-CTR4U%?XpR*XiSII7WV3_)Cbd3%fR!i71T zvdADIBhyM&zdw%oh3c6J3Oxe1hsV64s%m@W3(i{)29N4N`R$B=LatU?$&w z?u`5S`%t+zIK#PV7{=+G@(;GP7qR}+xJd|ohD{29lh6tHChma8*)a4HMEL>Vcm}az zDcBmp+a?S(i>?qv3ro?P5-96lprK2%ihhoV5e&<(BsOHbAghyEgBRWSaFUdSB)lX= zbb*G3CKQ{ZRm{zeTN7CR*2?=blarGlKYmp2e-L}CBJh8ETTae0z+`}tN+RWe(kqO= zOc`eITJuI~o^7YIJ>%`b&fB+t!VQMq_yH#7J2(`ed^BxU`9<8R&qC--d6kY583`dN zfnbR3!61-|d6Te?!a|q!+~YepO)Z4lP&av&gob;t{D(e_(!B&sqNU6kfS*1EULscg zkNmDoFgT?26f^bN=ZNk#dttRCrMd-;e;ea~q z!dE5lLBBV+40#YXUKb9G+Z}Eeic&T~Y31)jGD|tTl)Dd)c=nn8Pl|^A+3y+bq%s_h z=DqqSpa6?GfW))&v$Oc5B%=lw*`*_yTU)cu-ta2b09NsRa0Si7md501k^~@MB&E;= z1Q@sLc^txDYBPe$*~Wx0@c@wI6epS=P?50j-VJz0gpW@TfW)0UuMgCjZqoCA4GrCY z40@X`^N`yR#RotLz~7yTX6VWTP)X_Vo^L@o9$?jdaNo=CPOIF2t3&uT6LGMZF_?p% z-kP%73^9YoGugz5WITH9DKc;J|M~)As(JkAo7b7>K@f``Pv!UiPq5a>nRTp;9pH6k zxlRl(Ij1)jPh-8ddHMMt;g$|W`0_@xzU8=g?;Z-$?QxLygKo0{l0RF!DIYT>^?;+k z2_8rYH;1HTeBB!;EztZ#^O)BPT1so-yy3KWbkHb*z23p$X~n_u@hqe(00;q(_$@N? zI2l?!NStZE3w!taT(bbaCq&{RkVM^;2HrVt>+#;-zrA5IhZ-R~3|1-X8?f5{XlxgLo~-%X;}0 zuv~k97-0!FK7^jq1alZ<*b+U_%^^8L`ZM0S3}KG&BgzkF%_w0m!kwx*AeDsqSm%LCz&}4(VTw05c)N27*6pmN}e6 zI3V#yX#t_oj+c*2PkAiii~AJP#*SbBDyyxPeygmnf6rH1QJLq}t0nuEbJUF~g~Ku* z&O-?G6J3M|8ePaG4Dnn`cOTTYk(2?=zWo1*cf-8i5^Blks&9~NE?W@(dYkLX(()uCLiCaxz&Gv}1&VB_NTDeP90-oWBvWul8`?t z0;$&2)PJBd$JlVEG(jFGAgv%k0r%%46EM|pQ!1;!gxTrQJF_UdVc^FgF%447k`>Hf zy+U3yU+Ryg`5e~SDZ`S@LhNQ|$6B-o;7RGi*K_AhEN-ekTv;*a=XSmkKS}>gJ1+AD{ac)eFBoPl<_h(xl zA_Y+M@@mIDY{c&)VFTC`}20NpD(tKRD8e?OqmpE zN>+{082((Vv_0!mFLC?blkO+8*GtJ|m5zO0oNrSNgo%^!69B3R%_``e6uW_d6MC~@ zB~2Ve(!*IsEhIZk*QJb`gI6bIRFXs9>sWE%V_V#T4<0%W^ezM%2~QW`W5~$J9^9kk zsXjhFkjpe&X`w_Jos^Qo4|Ktq%Zl~6(twVh9$8Dvt6$jKEzQloFfhW10x{FqEPdw# z;~GXrzI#`g_!3R`{@d67#`NHS3OJeSy5o4t9@yX!lKw6zh$Lmpv0k?@;x$G!EWbNK zy{^Z>S6`|)IY;EfIu)V^f>yT3wra?Rc`w{HBbRUnt||PEqRGe;td^4~ShgdCCFs1el1|5H>|0O1u75MZO)0VN)+o6uQ!AXyP_K1cQeH|>fJt2rGOJM;vE zat34}aj~(tm!M{Ri;ToF!1UMqPd@D>9bCs#9gQXbu0B15H2v_8ZbfBub!>gUNl0=; z*X@K8N<#1=K@h+TyXHP3he(zqnq5gxGlHe1<6t^-*Dy@^`I~<$pVKO6Mt<{p>%~@! z+~sEZ)!SPPQeli>q#m1T*Aq01kM81Jc;Rm&z19Q|aq$y5424EJP1%=Av$Ms=uL%)R zdr^QX$@p_vNHd&U@B~S@o z6sW5!xTDwCILgMyS-s{pR{dUupz%&O{-l0yi0oE>ugFI-{z0g?eG%Q z9s6hh-72bBSpuveih-R5GB-v|eMV=?ms#0gO1xYwtn@v+E@zJzDTKUQ$O3ES8Lte= zMQz{ee;S`lAaYj46!-Qt;&<@h3LNQ_)-ux0E-Ba%G;=w_d|%`LGcCOK|8Wrh`Mm7m zO|PYK$h1b_qAO? z=dnrsQGYr4PuzbmzczaNH*;Pdp2(S%wD28I$;!*80uyM&HSj^0g$1#Mz;3lZ@xA0Niy^*Dk^aF20pd`o508zqS)jC8!7De2!q3~Uk|~0K}-7?hDI}{Y}JB@ckdvz zn;k-jle*S496`|m36o&-)TGB7b1~SY`|rj?$HaF{W82Mqu8ppalhrAwuPgdelA3If zNt6ZdG$x9t)eOZ3dP;+ZAg&8HoSAi@<4}pi-UWzW|5KtmEnD~FEndOQD}n1D27=X9 zRquO`qw%S9!)Spj{?()%O-1YYU2RZ@BP!xW30q;iuwb}aC)+;^+h856Z zDEqVcgMko6nu>}FudWOFhYtagXiN;oLxS@PR51+ySr%!}tP!D4v^yjP!Us0)myLB@QqYKsx32 z_BMd{pgjid-Zre$kZcQ+X|uxuQi`Us@=F^VLC^*D<$r?V6`{L2hQK$FK<>hJs6avT znH^nTd5zEtUcv*zM+D_}O@Qa23Jy{Ko51)#MtYQpX_Xd#kLg=*(jny&%&HrUJ5QZ* zip!ieI^KBnMcsV$tioi8B=K3{vj5D$)((DoKw7g_0X~6J3QB^2SwtT5(?2i)5nCZB zR@shjpy93n4w_tnJryA_Bo{x;`U0b=9q{Hko`ad9aEgT?NdG-mNl3SN9UP5YBh>(2 z)YND`!iT{f_~k9w^KPIH1G3dvS67A>W!UIu3*Le-+A=}tHFaERl()Zw$pRqmSX(o9 zc#(?qaal}iYR}qXuIg(zZ(03-qO9zDKOYnc;(;@W@ytni~VeDL2hY2@*TVR@T|yl$r+rx~QdDfs&JD?GS+uR(NF{Cd8n8PJt4)AU=gW|!%)q$Vf-&|H|& zmgwygvL5J3-=wTgQLBsdn8@Jx62CM=nc zNPqgD?i%DsbSyiaLngKgM(TdENrNvJi^cXk1?v4l$1_e%+6PsOP$CjKDA=(MrMysZ zJbuHD13n&Vac#(pmbn$a&kZr`3DUi9j)CJIotS{jEUuWj5vBy)<;^gh<{?Q-KQAx- zMYIHXQr=^3?sJSNCrQG(_!W2RXRLLnlut2M;tFQNj2nztBL++6{?M!opxkAY; z-+&DmT1=T$UlhPMp!evE3m16D$ytj{JsB1iCFL7KqZf3R1yx5|r@kE`#lynTN8?CD zn7RbXRwKzFc}IH{&{;4O<`Zm_CoJc`y0`sqQ*;E71oYHe$QOo{pr*#$KRbVT7^==I zAqk^OR25MLgLaBsS{C-A5f2`VPB?+|k! zj+szFfD@>X2VlVyLWs^$lN0^wuCB|#_h62y7JjWy&Hd~%PV0b(!#q3)lyK@|{D^QQ zB;x5x#w8~s4urBjK{7ffCV-G1OnuA`v2>8&xmRots!eC|!WY2B<+sH-iaiz-E~2r3 zqZR4bR>R89!ntJZiRur`>Aj=Cv?^H@P{+kWyg(IJ^#OPfV?E5|F1kE%q z*nrJpifj;Fi)ZZ1y``)c{xMCitQecQY9Ht3+#M6iu`X^j89NR~69mzwz@S^2n|q*d z!bOy6i*%Cg=Hc#c24Q5V+_G#7Ot!lQCmiBKHbWdBg+9~*=`K2iBK7D1qE1Y-z@^um@Pm#qA z`0-u|UTmgap)p6N3F-W8%vH&pZ*NnWfYMM{&zKB(2Z~j3{t5O)~z|-T&p256(xjq9kYRQ<4&7E2S8#rfK&0} zps4B9T`2vVZz{aX9)yrV+H3Xb35AyhEqHu9m!@K(qsPX_>)1%^|M7;}2bk}@8(eXz zsa!0q{E<1(0Uj3@AH>4%s^Gt4BWOO=1ewCPE>EBDkS{Q}Ps+{DZwnMSSPwDc7?zCA zCDJYZzdHxh?b7mhziOHJDE(L*jL*lA$H+f%`<5q-&7e?cre@x3@z@m`?Sc;ijZt~6 z;T4b4bdL!A3NJ9bZtZU=KIl7H_2I)bM1xnj-iVu1K9Rfor{Y@wPHLGuu3eQ9-oBcubepQd~x4-(bwL9ed5IhF9CBP$Y3N7=D&WOAmKFXPqt4? zAbHjv%N}jb-c}D76fwQn@Gqt*oT|(I*AxpzPuw}{(Cdtq^xZ3qqM^SAS~+ICcFumP z-l#k*{?2KW^WpuL&d~DMw^wJHx<5aU|NgU`y~yGHI%m?Y&DO#!Z$wfj#|`}+-b3z^ zJw$X#`q;5!Uo-Ve7S38FM0#(eEnPT!QMOgbHE&_Pw=(hVSTy;Ix1Xb3vl6G(PRsBm zH9XR&;Qb3Rnx;O;&MO%{8~!$WYScQ%F^$>f_-SMQMVpzuta*2ad0+pTq;|dg)+zYH z@r=j$uN_GoG=dxF_BxMLEKAE=-oIM*?VGM|X4rqC6ct`A|65BHu44RzY{=%3Pk%^< zH4jHc%@uU!9AS8I!f%slkED635tk9Oe&NRFe=2-ghpPqzMRt289?AG3Pw$(l^#A=+ zF+sOhzLZaCE4Y$HwpY3z@|gSGbBJrO;YEpdmhHfofYghZRr90bOrLg)G(_Ct3@#)7 zJAk^4j6UxPY#Zu1R;<$!mX7>wxTe;=M7kUBW+-||@&t^A`W)k~BAU1QdrSIqMK>BP z&zIhs`{VbsY3Hwjz3R90nh#TX)5f0tnNn=W>ouOqJ{OXoR#Nh~VJt*BrMW5J^iUK> zf##8yf5ty0G$n8?ArsJ-?L5JHzwo2g z0WSank?h$6j^d@WXvsIAlNS;sj$!S@87kw*2khq^a;_63Z!(>{zBx7Z$DtsTh_ObiZi`!<=QKi8;HhT_6C2N?(QP5rY zSpRJUH!BsZRAv`&XcS@Lr+SXJKi(3v0GKA!oT-_b4pBVxRegq4<+=&y2~knsLyK#9 z257tZN??F1V*7p2w}%N2H#MNfw>Z$CM|n_k{O{wNnDRj2!%NC<>?|Z#^|#0NuBVB| zHpGNnJsa@qrNOJ|E>EZpgmrEmVCrzW^`JzTpP}VW`LXl4rcKYS;|lKNYbXDsHsLby z{DP|O#0%_Vn-Gfi1eYr&-<}&$z@){TQLRPYe=j&T7AJvoW=n?sc5d!TB+crj3)Ti*Pk@A<|75WcO#kusR{ALK+t zXt=t*K{=_%ONEa|%>eBB`tC|=cm>_3%VDdc)3nz}eMu~REe)!K4MLiq{nOY{S4ZR` z{|0*7)@qM5%VM#SB6Z3W$#Ltp`4pcJrI50c&nT*5#D5_uax?0RF)x#c#bGdCBb}g&2~`qA|_?J z(R=Qc;u%F!^3m=+VAhXrK>JVM^ZLBoVTK96;HnR86am$ALvaHDqv zf&n$*jwe{~0$&-Mn4qfS!H#AW^W?*0p7(Zr3MP1hdRQD2t#*1i;M8kW{5vWGWdVaM zq(+`vn4iZ|cvtgeMLn-&37~m^X36x*e3_iDJ{d^3D)U`41i(Cn!oU$AtTpxZJckaM zf27G_z1P**0wqny_DQzxv<2(I8Hw98K8@i;=i9wDXl94heS zkziYB_?iWcYcb5xlM@fHE`;+cy`|mzjHE+sI=BwZyjgg2TN^<<-h7%G4!X9u<0LJ? zzW93vnx1awqI6Yh&rkShK-KzooAm$HhQGL$^i9a#1&VG>wO~&G=&;p6LWXBB|JD!f zG@lFQ<>j@XDfjN>D1mDo1K9(u0~s)6FxpyucpL>7orPrp6dpe@r&8u|OU7^H87fEf z@r#1dH`hUl2L$w*(22ks8LPFki|7k^YkfaTJQZvUKH=$sCHD}ORu4CdSP%l}0n#1w zUT`tCYx#AYcwDlw@Cz*7mU$gEcz7Lafj~?%>_BFfXt3nofV0rPht;24^X6}|FU&MF z_ej6uZMe!w^2Wp6J-JDJUZ4d|2r<#1c#xF8G0J1|_XRZMxBN*m^)nW60MV@+2YXEIMAp?+f@ZWRR>Yz}syTtxq z{>`6VmtVhnWeck`epif->}+gQrGotYqnRpIwSt88HH=l#*%FH8D7sgkxbVUYYcifb zecIQtHWgRvChokKCr74&$Po_{ieX;w_Ot$(-I)8FBW*I1xC^}dUb{t>!%eO>uSZ*!M%@}5WS&&x=HZGqU)^m6bNu1&Rif$4)<+!=Yf&>y5W z02Q zUwcBa*P$aQX4?g^lAxSiQtF!_?CwC));BgpjNa_EgMS>$r9@=i3bru9S~6p8ZC$G} z`5{inJ%!=?p9_XoZ0%O1O)V{j1qIbL2xUop-Euc2FE8@QAxX(gg-E;3g1#Tk*Vt!c zbdZkxtN?rZv;zoVbB2dJ^yFV#4Tiz^0F#MWzi#qf-1*Mg^ZAY*9%az(111|nk*PAl zb&HRTiJ7gIl!5}n7`8Wx^Wqlvz4woO7i-;|*K}oz)L(m}pCbAnczsVKhCcK--NFZl z@cTkbNed6ttrZB`Lraf6?Xp^KIlD%C)6X%w&i)`gppFnF2fVAv;W~5({vWeN2+4!I z|6uuk@C9;sHt3w74Hqx<^7pDy_e+j$2HO zzDNoi2#yI$Cas|2;$lE56B+M!A#xGqCR{a8{iINt7aZ8tn(KbjuE!n^4;%(o`^@j) zAkRFU5}%<Hjm+xft+}HpPgAWjZTR)y(@A^D5%KLCa)H7YPwS950 zcx+;VIC;3pj&Ypr3fU{)`NDTQI06;sc5F1*p+Gi)O|q&HP}r66?hIX(m}g&bO>N>w z(jqz(x4F3qc=pgX0(ZL(+C9)yR8&Ml5MqJ0pWfCLR++nJ!mIu`-}RT_Ev{W5D6XUT zCneXo(Ag;S_b|l!zJb!8Mop4E?d}NwB!V0#>$*Bb!)Dc8fBNgya$e})6bKzDOOtra2ugh%Y=Zh@$4M$yG zSeTwu^;=zdtmxK?3)dz4lpM8af|w26&-984nOmtrK_n`&7SvPsM|KvOys}iAIJ&z9 zyg$p*m_1AxUh|`;>!=A5k*W6NCibMgf6pX_8>NPEYQ#_i|HE!&eVF0KnqhLK7gMoQ z_kMD6=d&{=!I5MaKJQ-``F`TMm8lO^hTDL`UJjat3l8tbK7Xpuh|^&$n`yIWc5Qo< z{N;}A<3CKy%)o*kEjDsI?apWAOwG0J9gmD3sOZWsw&y?4fxyZ(uJoC%tZ3e2rsXa_ zFRtZ0VOo9zomzs5o0l!!vq1YE*NclB;t6LXT1@+lTK&w}pUCMrtV%t3{Lk01u~fDD zCr;?;NbaK+k1y8ymRH9lAgDs|F>nPvTX>*G)6j> z9(3VKw;^%M+j-}6Z_ zLwkYSa={5v#%@-dyb-m<=W;e>E7l*TF44)1!3!)hx2#5%s5@zGyq%OT!;`(O4px`` z?}9mU&8cYNc&HG|edZH4e^a{bCtX82?~RoY zA_9+I&&(D(%-cll&p3ZqE=xK2ftGc!NVT&yYa0&XlGy6IFKxRJi`P{2mw) z=8OIHXF_rMsTh8a(1UDxELqAI=SSEeMQ`xk=6Ol`+~vDNA04G-YeY zM9O}uw_UkehqZ+3=J zZQd!ZP&U-&Xlp$ZL3XZ1O~}8)=nHfCT3@fHL1P+&^i=Tc#p9uiqG5VxV?tsS&Yuk9 z82V41erG3glU0V#wzeFx(y?pd4!!4k9eF@XW$uHgSfk?}UpC}EA-WGiu?+zk6#gsEHXFOr7OoeGLUo$!C~ENOtT z&2_u3tGCSDqm;9>ZJfFy3m)>>4}NI1-@D=6duY?%*h4YVdw1Bf(&pIA^ZEg`<6?}} zKP34zyp;DO47JsIp)ms3gD!u>lcd%MKODjqn(qrKM|vmhL#80j|NhvSDo4i2;b}SS znB9D{>Y83jIo*0&LSjp&_7UAr&T-jEIV$@+(-nE58&MWSc5U+ggcAdKx_0tIHG2M- zv+bH0tyO1w+2?nB>ZlVXnm7F4n*pto$D;%3uZ`Dwb9%Y6UX$@h-nGBPt2c5v=H25dJ%y_^!~Y+6kv1@1dbaMS9m0%Sn`<-}OssL*dXV&J(=;U-lvcQX}2&8abG} z)y~r9)9Yz_PTa@CvQ)qNkMF{sop0t=<8!$by6paY{pEYYKP-8^`)nMRou0$>Ao%K2 z6>+^I56ku_#wmaN!+Mt?C^bjvi#&}unfES&ox*nnZva~o#ret1F2|S`p<2fnYOVQC z{5bd{^YcZ)Y?5xcB6k1p#UZL*({Ip=bT=w6zACFgBfj(h{MW5JM<`)$lb(J2D68YII#k3;H-v5Pg+npLit%As`m(SYL%*1!N^?!Fl-bBy{qAMHWAC z)t0%R=DTi(26THx6M#X>FwxMX=qk+1c6Y>AfLAO6-tZ=3$Eqe0bdMbBQDP z5&#$6(M=qk0Ga$sl4L#ik6yR20qLv}u}*i&6uR==V+|gb!RK#y-->*@*AeAP0^~o$&(Xo2b;Z#d73;|Q&vz9> znHcycr$5_%x+5#oFT9kM{k5~YUSjJFQdjDLRO%=?@2j&_x!1uvV#G^=tY2S)$Ep3p z2N^U_?MEm8WeeorK+iTH6NH=umWj)%k}dn64}n((%Az=f0Ta+1BLhRg5@GB-Sz7a% z_9(~g1E;(s>b$}6#Pb7=&2jcr8xD~o(pbf((i0N>s9_BIbW3MvH1V$@c3?p7KhoA2 z5vH^XJ(<&_bo=(jl}JlSkOtK78=-@JPC7gQ3QfIqr(*2G_ri<}4b4j{Dp>xsY7Mh} zLHBdSx~La~n8}}+oFd_c2@gNL4vK1&Ad}t8Hgy-+G(c;+I$16*LQyp#w9wP;8Lzao zCrD5l59HH;psQN#>&fVKe@esQYc`@zMM|3a!%&6<{3OB092R!&y#Bc4Gqc3o<%Dx& zeOFh)po^h_fx*@p?U$B}Uy_$%?k?w32qtXpA7xmMjDe5_zz$1M^~T1=O@0%n_U2~u zyxFWQW1?L7WQe`h+zHELS38mKb455bgVq#7@ zmb3(wcL5Kkv6BcMx0}q|etaEvqAczi`(0<@;p^z=XkG*G+~3z{GZg>giloJ5M*!r0w28)Mz5VOW z@&n*w_c}p}gMtKSRaI4FVf~KRQ5(QP;BxEM*|TRy$Hw~B@kP_{D=o0n)5BJmI@QMu z_ttr22Z%cMyo|#$2hEG_9r!O8MNB_5H4&bXlj^3XY@;{#7ti(;PU3KUDNvkASJzL8 zr@V`zH~1m&ItOskbs_tpn1@Iz4{$aXc?5hoIf3+{4^hUMZF#drX8(~pIC|y;s6>*$ zKTpd^NijJ}#?%cqHQC^Pg9E1@M{LAiyeOKZ5c&G~`LWyms%mX@z|%eKy~4WutFQ3z zCU6&XAZ?^a2?Dd1zYzGK2Hr&1z@YIFj!y|hSP=erl(=7%y?4&!>hdG+)C~tL4ju^r zEyA%G&!3YM?ZQ(7tPp~<;H?vdEsHliN=iUg^s@?l{S6F-AU4901;G&F#)!I2-`Mwo ztEAbDY$vX`^z;cFg*0RwtzbU`=%7G@A(oJ8QrjpwRc;~~ljzyp953VUj7%i@yEr=l z6Ioq_0$~eUt&Vl3pQ65*dDCBt1b`z65f0y$>F4ym9JYl08Ow9b*E$?EmEIe!*t~Uh zb%F3?XWIdehCsuBsnyaJx6xSUXVGU&at+ zmiXatjv+w^S@aSo?@4&UoOA2rQx01e!jE|QsMh=J!uaMVy&kPZqW1h}G%){^3mZK5tRPKWklN$+DyWsj!Jj%Lv z&-CMCl3V|tIz?^5OFDY~@6=oOjs+EDxT)6H;fx^I>IvVi@b4TpFG7%4AR+thA1IP& zJX*zcFx`;;4-!+*1tRT%@P;g1L+@yHUTK_XP%SJzRFOq)QCV^%J7E^0VHl|NgkLhM zWugY~bPMJsAl$lLW%FcMenw~uMZY%P^c{v*v%n8EFfXC`{$!f4w075-`_%OsesJ> zjB2{}rK`58sI|^i_`e;|b?XNPHvOnik&===EA1!y|8mN=mD08QK)%EsS6O^y<-;BP z<-g5fJCZ+Mqep|L=>L$T{){ce)7*R%s+M~5V_QwL6Ixlic+yZIe&GUj3j&^{sDEBS zu+^!g3tMXoNSZ7)&P5FF)3t>Wio&!s|CY(ADN?&KVHb9x2;LT*8bZ)C<2#tejlH=*W-z~PKwW$fm`>_)z*d%Unq!kh(5qwI(G|u4rL;42i?0CEg zwZ=!#$#=Han8{$`v7o#$hNo&;9W8!|y8Iv0 z&)SsUZy7@}{L~3ULzY{q`!lZB$*mIc@bW^#1`Q#{fm}4A6KN@I;aw)Mn0!+`SC{d_ zfuVbi3vnN>dPiqx_P<~A--eSQXQkV|8@pnMVtZ5L)I)^NIQsT zvn2QTBDv6FofJoVmEqe%D;~zjeFxBX@V571#Eq-T5zAN=U<7<&Jonz{rPs76#DLMQ^OzU1M{a?Pk zWqr5x{aEh@T)~%a!v5OPO=bJ%=H*R#hAj7qW70ubNBftxy2CJoGo!!C7kK1jm@Q?h+SewjvbrF*Mm7w2?Jiv z4B1+R`%S@T%;=^LzLy3_@lcd%=aGc$Uc35G1_NrRyF+wZayfQ(cX2+s?7CiK%h@YK6|rb+jMJXyr>fkq!W*?MIcr(SM){Perc{PS|}rsp~np*c+2X529J=f*Qz|Ni|&nMa%3m+vbC(uF{| zKPlvCAjtg<0~Jm+<=rS3MSDFoz;}pIPX_(eCOxCFoIeuU>+35G?(4c;ng0qQEuooR z@|sVF1y5Ul#$Ffig3%j23Q}0&@sv!@&PrG~jZ8ftnOpPn54*cLPjV{fyYOSFwsc&w zTr7&Z1K5MGa(YqgO0GQUZEKssHIV=6mG5m?$K)f#JN2+S&GX%HIq!Z`m*r>8LM?Od zuq^vZL ze70UICiX1(&E{N|GRL2`CuTNs^<2j(Zaa{OQ%(2E^WKS&8qx99M!p5rrt8Gi*T$JK z-^7_8NayDzfuk4gb7WyqSul(KB;nHhdy3a9nAvKC+Nk@@JoRGYa zJ4k{EEP;H`*b#1S?zHrDh0=c;Pdc_TGmm^{tl<_klclTWfTf8#r1q#O8xBo zB9dE7%!OK4Wi?K>{_r90VV)M5hKkyL-}*R)G6_$tSO_xjD{@a5CuNL_{5ClWV5MRq z#>JJ?fIH+f^1Ki_D9G{Z^x=0k*&B{~5`|hKHJWhy`tbt?Uh$N7wpqJ-K^>3^YtPb$gVM)}McM6;JUb#*!#y~yg5y9Vr>{MEUdw)ZY9 zLup$y+o9*)uxQk5ec-w~aQn7EckOUQ?L=DrxA?vThP`@wYue6>G;0XGb~+K$M2}Wp zbgZ((^bCeMA;y($=Q5h7x_VhM-Tj_Z1@e~hFS-0IpB%gXH1wcoWq5$nY2;@!CJoJu zDfvrz30A#mpV`}AyzU(7ri3^T+T%CejVV0Gj1IF)-G*#umy0MzET8DU#rmt-4y-;! z(IN7U)~rSemzU?P&bnuH4YYM2yh#5tv%3dTFR|Nvg^SettSkRq$c#2VOTBi8bhK4r z^pAEv=T7VPM1pJ@W^m(6iz{{Dkw#ZskOrtez$j>N6vBPH>7%jYP{cwCigXO09wqNN zftEW@qL}YwYy;N8om5g*wx5yFq7{r9=enAjS}(EO^3fe&i;+aw&_g0Oj%flL!^04g)AiHZVngy`Y&6it#35?F`;JjqRBx zd0L2j5NcRpcLV`+P>@pJ6_ZBFCK|!Yoap^>C+>9Cd3Kylh@LwKR}o}LoZmofMa0A~ zyupRD|1iHjvutWkPWVkaYiTKW|+qw+Wz09Y(};z7E1u zZy^oEe`u`VVn7eJd?5Lj{lOm?iaR`hd!YdVv%$x&07~Sm?nPr0ac}yE=ld`05QjqL z^|9#_JOz!7q^#w3pM_bd9!sTgdF?jq#Q)u{toB}L`y{yw@!G+9v}lKp<&2iB^+&i) z`TIFqa!M5*-&{>Psq$fJhkoFM%AmUB{NyjweM)Fbinv~uOeM~cPCVg$u%0!Dq^cC= z#$3m=o40GjUHzCz(B*FS!fnOeH>=3!Y5Bp}jqd{&v+jRTR+Lq!FN}CCVbVbK%ka+E zo~&zvEQUAray9uUn#VJzS7Lwr{vMXxS3F*q2Gm(P!(utNY_CY{xpBp)8(q^88k9mP zA70*co*$|wI^uanBSiCZ`S!y{5cLWee(e1=`p5Gt<7mYb=^jgxeMDN@98P!n<~nyp z3xkbs82PN59+>)eAeh|*y-mT=o=>8q3$Z*l@~BG*2>gVf_>fLui7FnU4i2PbLb3$I5tm7OEc45*3_JP zWa9L>YBQv=#_*Y6KFdp$gqv`%{AN%l2_#}aeKMsMqM&<2@+aSJ%#?q3xUG$F(Ovv3 zry6-afgH8Q0y(YOb8BqqNQa#SdrAioooeEbj)17;M@YtG5GbYi4B8?k66CIH{J_0e z_2c8|VAuRHsP9`_TiK-({3sXR7gW(6gz9|#;G>Z14K|mHJq}Pds4U=|1j3*^apmiG z$^$YRojW(e4tfv}?e9{*-(w7Ar=|5V>5yAJ@JfgNGW}|rQq~hO8())-Xi)Ri``ozVM0ln%C(*u=z$+5I{? zI@e?u6gDq2KcCa#*Ry&Pn6t5=@YF?l*Q}Msn_R2FZ-*=>yuJ9yU%OA#|FijlF2I1(dpgTsjLKQn6@T$`QqhGn~hAaGUEv$vgVhPeG)kCmEU>@9>Fpna=5i$E?csaKncu zdFOM~0l94m1%a9PaAKI;SYN;XtAy_MIu zI7!*9z1Se-Cr&yO#Bs~#=~#5w1u;Yap5!$sRg;sGS09W%RC24%!JOBznQ!n6De~S{gM+E?rC9U8S=@AMViX7X2M2qZ{FuoGX+llH04Q-;`cem zAMEc#-yPIuAH+m&8LK>~?z&WH(>+XW1F=x&sB`4?vZw=5HeZHJIL4nC9;_Q~J<=vn z?Bp&j5{4C1nv-x)`GsD-(y~5l^s>?VkGQu_i@a`Y(%>_0@UlH>{Ipgx%l?mB$`Q9;0+*^t)ytFJacjCS=71hbi4{Kw;6 z)l3e_^4h$9Z}r(?A|@UpSkk|!aUOgj4o@BWs1G8(w}-SPUWXcmvsm_f!CyQlR|w)**%`ExVQPHrbtxVil~ zs>x<%v@2deBo7!(PuCKN_VAFA|mbcV^8PVs(8`J~|N$&*)yHxA23RZY~)F3KlX zR^ly)i)ja4aPhp!c^#F^ECI2j-A;3_ZY>q|Ew!wLd9t7QF_qqS*Xb@B(NQlKL8GsK z7>Y*)*iPrH9+@p|i*`&>zt|=a`}v7s{h)WC0r%566O!`|lE|Z% z$jRxKh4>hD)64%noo?>(Qt>U9zHV+={M)QIm9eKa_4Dqkyo=g?Y&&MY>m4KMn&A1R z&rB%BJBK%1!0fiLuy9EzdQUPSuL~VzV`lb5crtFvn>S1PhhC03Dk~FhR7wj8<>co- zFo|(0LcHeO;-ZEAHRHhKOZnSyS!@gL{e;BHDt;{ZkN4%T)KnSUH3n=B`UHaW-OXQ! zF}BxLEC`2)nHk4a>9c1XoSezpvBRNm&#u|lyha@%p@~6x|9ijR!A}$We=4ObY?+A! zlmlvJkK$ZD@}YXh11LyJ%AvZ1-eFBo%#(9kjZ#zA7%K5(Rg2lud47a=(U>o>+BZSv zE$MPTiLa6Jn_%6(eAVf!pKcZP-DLdaDnB7l2XqcVqyG8>i+(UG`JBBh``>R`pL%J{ zYCV+ckXRA4cLVN>SGW)a->4)?Nb2VFznJ~0u#|Ej(`dhD!Q^K<#if)}3x((Yn0lPz zl(BgKT*1&IzJh%1V&|F7OrAznL4jtGTCvN1mlG~uq$Pi(MtA5_q4vUMt3@;!i63{g zFHer6B9c0ps>vJ0C#aWeon|sGD5$SzuSj`*IbrD1k$(ze*lx*5=Ty}UsqzY42yZEM zJJpQhARVK}PfWjeeh3j6vGBgiV;-XV!uq99OykHpPr%hx{ht?d04Is54h)mKa@|Da zOi3m{nG&^VUf^!;1CGnccmatx+6d>Qo*_}h_g}wqGBVsSQ5tADIUU0X-bVH68d78h z2ohK?j9~dN&b^jC8#XxK67Zn4811Bio z??am%AU$jhpiMct6wdR4)J#izb}`Sm)b}hpb4b(Y7?(rH1W~4JksJqqOBZhU#w{#j z5a7bncO0gD!*4UaEWPe%PQz9NCtg>Xi=8g9?QGt=KjwLPiU<4}Ct)@R@BviKGxHNc zGJc8s{Y-czy$_A;5*WW;7yO=KQp&s{mql!u9em7oNRLBLNHc)8aL&ok-a(npR|d`U z86?=W7Qj7mEGSe|0!hX??cdk?UVkkn(8a181tai0=HEnq&*<$*Ze~xu7oTy_ksUo| zR&{TCx~P7=e?-hziEZOudHDuXo1&;aNAyaop14pJF%gvy$}(3jPA)18R_3srCwC4N zp7B25ys6e-y@Y%Q(ddQgfe+4~D0s|&ELnVA5qUpHoX6zMaaUH!VYklUeSet{WnSH` zt!a@wVn_Br5`+t15dE2+41If$KUcEA}9(jYi7avnjTLV8K z#0R^?gu2V~44=8%nG^9{7E+M93x>*vPKyN3B@JD@FH9NqsuBHCc#sk%57jPUgM#IO<_M+3guvgkbK*oMo4r zDn(|)FxQBsv3N4pF!=id7KXsTF2dH*(w}4B{9#)g99yJ!+M!U<|*iV*v^V-%=l7vz;r_sg56gDvmcx)8A3h~h& zG@9x!jb}B!>4Kn_6S>dZPWANht!X!9zyBa=Q|h*_p`_tQ3SEKbPr1CSwxR9$&Su-< zZ#tMIio{dG+KqBQeWLX^@j9rQLYx?F%?ILsPH!!Z<|R!l{aH@@X7XNluZo}Tg|K$D zkT&NmYX_YMbwi=Ndpy_>#0lp`Pslz-Luu)9oD~JvdjKwBoWH>|b3zh{Zl!mKS9@_3 zfD?y?M{!@=z?gt>M$~hVA@J9h{60j2$UW{rv#(49QFNio7JtmnVZ|OiNRhuE9LJ>r zg1P$hJgwHwKd&@s`N!2nNEnFM{Jfo;8`t+0ntrk8ReXxGmgn$cW+tYZv-90>JH>T! z?f0ddAD@bRM{_oq(Uy5B@S+k`e)g9YXU8|G=DU_dSVhpG@Fe6JN(G5rYCM-UmH>g6 zlhmG?{_&$JaQJJJ0C^b4k?^oE3nd0l9@oS8xa%|?=YFwfEf4^c5T>4g*IWo&&ksap z;Pgx3dLwYktva5}Q?&d2dn=SoOO%ux*2^z`<~+yPFOHh`#SY2b+nKLZ&U8O_IAGwUX_t9US$m(r7V-A|{ns+XTKE&*+7y?a zm+~YNTQ!fW%T6AwF8U0)=F_s3mp=js^Zu3aM!)Nqu( zZTlp2MtI8BWz+uopCLH`9`^b-t|#hQxRv799R_5Wq|kf)%$~Pr&s6)j!>6y3`TLeB zXfV{@{<-pw4otoj_l#r9J8H8g&&)9{&sgj3c~c4LNm^e6Pd4u634K_NUyYk3U<@6swcDw(EMoWbK)=GmUyO*+*s;jX&J; z>g}8aVJwL-kAcDA;cW~-!HNXeQe87E{XhLgZ{>4rNgi*@?OW(p&slQsS&vyJTAtBn zE6!+g%-l9A-sSR$GvQRG=b(!ISj0Ai=<1UbQhN-UYPK)E&~_(c@~YR<_v#wiexe71P&_Du%n(8 zEaWUp7(BNkKi8NSU zaA`g=XN7f3zQM%=j6)w%jzIDyBqYE-Yy3;#5UC==zv|Z8lNHT56DQwPQU40!I&C_n z^K&pK`-{HrpZkQ!K#3b>U^h_9A+xr)7wAcMxv(#XEM8JJ*L(9h z#l&7~Gfhy@C_^mq*FQfbFug`}WP_k11aklt5@xbk?r%*he{$Ai;=R5nPren1%^7VS z;yFxDe);zw6U)Z;AG`-;`ZQJBx5$^~-p{(4af#f$CMQPA{?|4%|3`Cj5hkFEFAb_q zBj=KSVCRF8@M3q&*AAdKxOvCek~Fy?Ohb?X#_rCZo;x|CXqP2Sg)J>VONRsf{XZl~ zF1nB~pVk7uK`@vOI3g&y!mQf2v9WRTRM{4t=6z^(ZsRD0uqCgELm~68)q36(-QhRR zBfbgj$EQ!)uFM;!LlC5h#F-=kLPw{FN9Dcvc&3X!#FzC4V2=)4Ai7}(mt>UcZ^M(N zTDpM68NI{0uPAADeWr6OFqp!{O4XhQc)1GWe?#H;CLYUeE>&wb?NON zt9hmW1}uS|kK1a2?cQ>Jkl5-w550WQ&YDDW9Na7PM*5ZR9=C5pTe{x5;HB0^q$Tk0 z=gbkj5)@ydfAT`^lLre2okvGUVS9KD7=rL%KsWjD;f{E-GdxWuq)59wq}erf zu8f@~kJQsv^U$M5`@_1s)gp&NL;Wh1KMd;VdkJNyCSZhz)fw>}*I=py7u#@Zv#hd- z%r|q;OMW!sxP`867npz>TKcY+beWCqM`8i-#l^+JW_7ZAvcLA>7U&Hu>0-b8`}fh( z?%V*vOMP0wkvgfpy-}6=9nbUr_-=dzx`OBLtL415=4X=8*u^z&364YNHqB>2V5e6f zLXRKc2;e|w?S>&AAD`n1nf=M6%7vMks)wR9y)RM-t}0v&RX#+wQcT3~7>wyVaPYcu zj*(!;4ES3$j;Pa&N0+{nmADdGKT*M`%bk6t#{#x8%QTktufZIo5!9*!PHYgntpFd6 zF?sMsUO}N6k~ah@>gO1yPrvHgLqjSlApyl&q-m6a5P_S(zOaa21z&BbXjU5-F6|NV zg&7&1-yuhWGopC;z?nHJetQ|6&9kg3@SqCBD=EoG0s>2O;+{~1>Y3HsTAn7GUx`ON%zN( zTCPr{qq2Wf^!!6P>B-@_N2c6E6B^&!d4F|73N-%Y_$RuqrD@xM426?EUHCqN7xDSR zF9oF}IBC_3y1JIvt{K>%A*dIkOE~fpz4B?JSjI2f;LJ^X`BMD|JsLwQ%&Q+9H@yqc);a~ePHN{`9JR5?P15AL3N=zWr9_@*NC|(^tLaYe2BNpQ*IwC@UQ5;yV z3)fFSKFGurkLEHftNjNK1iLutd%qk7CTKwDZ7j{)5xNZ?2^C))0#KMUGR9(2y+bPZ zx!DrcM)^o=x_tR^*FUeq7EP-YY1r5zG&|6uwn27s8^a*!kXd&g+kz8^k}jX9SLs!A zYQ&q@Vay9p9iu#JY9}u=VfiiQ(L>fgcVS_E$r8_fhux%&4hv9H?zBI}E5o^|psuJM z8d92^9qX4Jo0{ko5nuMD((X#>r8Q&T;O#BH3d;MxAC_?9LUP<4;(1=-P*fEM^rMCD zi&pB!=yp-M%@}DQh)?7c(iIVU!P$z)3#VJREH5iG<5f?Ji?gw`L`k49=5^4OU;6ux zw_Ji43WqJ}Z1o2)AW%V9Z*ZQx;r8JQH9y1t{pM}Bwg9n!iwo}EkxW~Yj0KL6>w{Fi zM)KpaLPmicJ2m!%s2oHf!7j#z1|*ju&H&N8^uz|`u0pqx81{w{?&y5i?}UVai{T$= zuXT@Jq11Hj9eVJ9My!tg?xs&B8Ew=niX~Hpx&=Z^*{KA?yP{;PWNkJ0fE8dGM~<1^ zAowkpH)9Xi9`hYk*U?!lqA&@os*Z_!R5_c6=|{^{!$(`CeKzPQ5h-LI z;hqsEHsF1Ur$7bt*)U>wgx<)X#=D3U;)1IdttgJW8)!F^I9sBLN@ zi&DbgtEo#*Ct0yg?&T=szF>)~7fe>|@1VIwMV?i!x9RCThir-BfaO>{&ASJ&>++!) z@g-VXTAb~x!Evet#@9Qqq@)+AUDe#*xK3HS%@i<0En8MzE_Zuw$H(WRW%)W+8W7x| zC_3FuO4b{8+VNZV79+%ofsWwJ!V;YrD>++u%kjV^{fo@zQm&zKMo5?i{IZH{#5+Vk6F3dsZ(h= zn!#tjhCexR?YsX>LM9^v{W)&!sdGv+{-KJaraD#KU;6sok>Oi1swfa+lCYv|WySlU z1PnbCpYR`y=01mrj4&3#&H~11?jvNHU0n}7wSGb^lgx*dcI47_+)DF*Yjy`w1RUyf z_K6irEr4Q7jE$d(%soc#3&+9ZZP;7o-(ES56B0#y3aV8J{fP8Su7O$&KG+>bH4PU| ztyuyFxglOJTAK+8=p_~RI_i8L?%bu5b2yF&O@Z%`xRvs%C544X{gASMIiaYsEkV*a zNjL(6ek6O-mWwTiKYU299MU_HAIZ5(xvCQxnCH*8LN7q*JA{b>Zs~F!f3{M-P-$0g zq67*?VPwTw?$+h9s`8!gC;{)ULDjqU;Q8y<^yNp|4j?8DZ?faZlzO_kMWnedaMSfmUZPw}Q>hGOBX&P6%(k4H zNL4@^l6g+hz}O%qBQH8u$AIr7j{^y?-+scUa8yie8XDsdbo#%kC(C?uW_k;_F3j`| zsFzUqTfqK2kN{a&vc|KK>7Za733wyOZX$UkPUTUHr29e|p2Ax}E9wJLdpmbJ8(c)- zSYxX(l_D|G?#4&n0UzLYMi&9CbT=wt0WCM?AiGQZ=4;z-Tf9>pdC>RbIEMQdYq(7XWC$3@}7&cqkZuHNL0TH=5+ntb5ReAxl5S~` z5-C9%rBg!bP626@?vj#}`qqAC=KcPf`!_SU!rs?)o@*WHkx&wn#HL-|0uw7l<=ZTD zx(h)p%sNT;IYnQk`+;%(2nd8c4u2;)Bth~w!*V6d!-uVWmLS9fH_9K#kFRB`S8aD{ zU)>V7{|!MUB)-6P_^GN6LCre+L3<#aZ~%=NMPYVgpd#mv2h56a5dMZV7N|*`vI|TB zhYds~z`L>RawL63iQqiuJpj*m3>{UT`sp zJZ8HdOc?HwtDqO#!xm0@n&%tUh(hu9a;EKzDgnp+`bX|kxGCTN$r&6y;+n7@|As%3 zCgkpNa+W8h+}eh%(McCxSz8eDJ|)Eiwz7|$yOeOg!>QcI(a9QsMJ~WaA=v<7*)l&+ zvk~bzq@<*mt6d9$pT>%`Vz4>ebJsyoi2!rc%wcTVNoz#XTTh8 zbwt|G6I7#FTs$-fYZxj!UA7rj(Jai&uHZ?7;p*R7s{K?Nw8}skB&^c2zkgrozd%uo z_h&>@d`KF?x%92ov^)^x)^u-p`S^m549ikQJSri@T+v?$4g$c2IaEv6IQ@<&`&k|s zYIzRy{`dh+KM$pJc5FU61-1FzVGC`A*fNORW#{KVfq6ktLR&oLrIVaQ6cY>a7Y0+Z zRC4WKA0?MyS`q(OxtI0F;>7mxw6)9~7cm36MBdlCgkv>L&ot_s<`Sd&U2C^z+eW(H zES%YK7brb{z%jsyV|8yhGey-@JME)qNq(Xm@(J(GOUAs;C}FQna?8p6-%~6BAfTH3<3j>(JI+l#BUVd+LiY(Q}x6r$%F;ei>1DE66Mc5;<>ZupEbH{Ah2UdKCRx z#VeHWkOG6{x(q>F`-L<8TKZ!wIL6h>UK|u;nl+rIe#=i*6qA zj_f@uE=jC0x2)*tpMSydz9)AiOs+!BsNy(q>zAN54YgW9er~Rku2!TR$3Jxkr&X*m zFBSC4Q?4WhE6?r&3LICpfY^*fUaY0kol1mp;`F1Gkr6FN`!fmjBBci)mBXj}IWyA$ z&eT4RU&1J<7`PLoqo6baA#qzxXlN)zeoPGyJ3z21+?PNiK#&;!#VpDA2)Jf}R)7~X zvfkwYp4aviNJ4cy+0iloTMoLDP#8SHsTG_TaBvAy#JRaSaLAeV#wG*@qu#XyxgA0m zEQ0aRWo>|bYARDQu-qNGO+g~@tUM?Ph3M`mkfyJ3AP>vW?*`~D5Y2kHyJrSz*T8|) zqpag{c3&bY%Sl_NBmR^O3{)Z?Ge-gwsL}JWc~9Xmf~DI9E>T$VphU@?=R|wQX-fB9 zR1_$b%Q;&o@$Z11i2MQp!Z1+b{ZD9bik^(;Ag2aUPqTDTON-PSFNdI6V{B}k zf$m|axC|d+K=S{_VbG*!U;wGHU`O3S$nA*QFsy3pq;rgd(6!3Xhd;r`XY>~wtQdSj z!(N`AyYtNkY$R13$>GiA>u~5kU<4i@M74blNXh15Tt>81z-@UT{@?8FwV<;CuOOPul-D#+X_ZbqR!AXEvw!OMsJXax?WgE(x# zKr(^~0oQw!ov?pTDOPB}F986sE%@~G!JMcTFjb2PScO;_LsHxMrYq3IL+$+}*lty3 z;y#0<0}k5Ix5C20CVQQ)FonBO8vMYYknjB!RdaD`3sYkrI@e)H1E1D5OadRKcq*hJ zC!VD`q>y~aS3Uge?@U8cg=A?7mN4+bI%{P~y zvBs?H!lmGqXC_+eW5+|!L~>9D(Uw`#_UuT$8okUz8yPBj+ERik2Xsd@wVyCvV{2(P z*p>fYK^!Fn4!Ed@O!k?x-Ma%pO9+FZj|>iYE`W;+T+;ZGY@YLEi8X8thR($eL}HkJ zh1dEsx_si|PsgE>p2>vf0IR?+A0q0e4WWqqm~hC7(pf1F1r-|BKNGTY!ka3u5Ejp$ zeJwWfr(!CE0`gxQEh485h9<*jcn3*U;20T?#NlkoMJ&Zxjx!g*HcFlD~!dXF7ps-Rcgw= z47B23B;>Sy_s$MZ^s3V(E}0HHy9;Y^h6W`QrRgaRL~Wss{T!c+8^KDpBD31ELY?kB z9h5a`ww>w)v18AtGgOMp{n)eNdk4q5^^Y&K)rMv9U^7QbD**?32rl*YOqta-A6n{! zjLLY~vwIQ9*34~ln(aYz`h(8#=NA`Myw7a?gofB}ZIRfa1god@#$z=u4JeEyXz&%0 z=4rW&`PoK^wh;58QDu5&*t34{=#1(i-g6Nzgd)n}0>PJlNh> z#fZvY4m%6YJiDudF1#SGBCmwJNMazU`sU*INBJefjaOxkvag?WuqK5un*QO#b8)G@ zEYALIK@X(cJGLXFU*E&~SK6=h4@}jk|M0|x`ZXJwe%>ZARD@6i_(0w!>B(Ci@U7YM z|9O#kHN-olyl$LlFS_j z!|LM%B%8=S1zC=CoPCi;M0zY{KXKNy?o!bwyigvBufK3!C*;E{_^PLRg5 zBYIu#MtTUjMxaF7C(f?+=eLt$4|r_w=VVh}-me|!e%Ng%Zp^rZ;7@VyYvXC=5bkLj z^cScf?OwkIHHX|T{U%#hb`Jx$mz#6`NqfKQZ)^;_gr|p9k?}cIZ|?FQD}P<(#MK=7 z!7Ng->H1PCI-tPrsbor>(k~--e_bElti-N%R@E{p-{KNhA}g%yUB*!T$a~!E+C%cf zVsuxB*TZ$?Uw6Jr5wbQ4TV!wXJKD_k%T1>$;@P7kT!TS{hb*G5=u(Wa@}|*qBm@<_ zhjvat&mv*_ZdBJ*s#nCvn0dQDCO{9ml+#+eAKD?XxwSezb$i0ss{fDT1+~s&GV5*Y zSK^1-bctuxhrHE~hQ~;DU%6(rhe_y5vtwt`99z2DnX6UZcTp}rR<-*U<@dvLA}=ub z{|Aw|S1X2aVWWH(_}w#gnxbXr zFE^&Z4`Ea+w<7!BmG+|g$jZn(`r-2G0zHc@ERs*8N*n8i`WqZcRhujITF|xT$z?O?Ini(!!SvqaaB+@x zPIt#}EW5U>DZEDMB~EInpYV7!OV7i(lVzO&8{yoh^f=9L#g|<_9Xiwas`;i+yl4q3 z>8zOmRr?PvZ*^-`1Y|Tne{=76&P?l{0>LMW^yhs}RqlJt`3ase?wYZor<`w!?6(&z zIpP!6Ka*kx5uZ-lQ)2l!F**LqK;Ei$XiX~$l|HcT6x`SAIzrOOuK8b2Vw42idx?qV zj9l?LB4jK9u5s8HiZ)i&Upb~WsL93ri5%2_ykGu%stXeJdNOFa4ogia$Wyj0x6ujd zxYK-!M+lwVI2(gx)zi(_&4}YG39Ik^@5e>~0m!#canVhm6XL_=kn1C~tP>QJO77nM z@kW-BpjoptuQOYhiJs){d{#*rM(W^N{EcY zMc}ET95YAJhOfrYf;Dm#my-O9yIJ4=%PnN_f+_{-cVaDVgI-dYbsDErl$qJjcYP9# zVn0%d?11mCi+v~CBKL(R3nXh{3wzd$*Mi#Z#d?vYm1XiYg`rf`mx)UIMo)JA^17Pz z;kMSPzUf@Svgi3Fx6SN-{YY)h4c9vOP&l4D?@5uo#q1W3u@rUg-=Y5?9zt-yX8H&S z=xH-GTbA@yP8LiI4Cv+l{I8evi8cme|$o3(9CD09?o#+c=u0A#6m997*X=?dqxrCO2OJi1quBuVpf}&E=PQ3 z^3Q+%Y}AuxE_dL1AJZ1CT32}^Y@i|6{;L;d#Na?u)6THi_H}mq;AeT=IHhYzMY;T( zFTG+T&vt`Y4#8Y2O@k+Y<732s5VBy}oz(syUu>R0Gpa5EBK(ZsmTuVN;y9msk(~E& z$lLn=NkeLqXf(bQCHPc}jHXk{fv@;MKl&r;@3KC#)f;jB!Rwm~wa@#+iaMpk)2s96 zt2fbPf9*dW-uTfRr!?KvyXVcIOI4rwcTmY$R+;|*lprA~>B~-KK0mBcz`wy3_lS>A zAHsViB?BrIbAK5Q4fORrcG)5-{0gfG5qhI2><4ntL{Mqgzpu z06bP43u_!!G8^3=36N>b8E7{{#iOiWEz zV0wqzmJ*O1LX=NiN5={=sf-8o&8sCN>Xkq!07NlBy(*YJ^oo;(de?4!+5+CFf|XgB-% z0ORaZVg~^*FksKZ>>TF-Q?|kOw@AR}q-PM8K!AyB@!g8QfBpKLfuPa&AEM|33XlMo zgA5pER@UX-jC8-{>LQ>v5RPg|W@bE;f_qyALcIr|o<2%=j`g6H|KVF;d7QlJE z{a}n>Y^mvG>>+fbGG9OKLGhn2n8%r39?ny3N=BiXc4Z)2CUW zS`2Ev7T+@}LPbXqQW+R{0B)4Ziv7|2W|QC=5>XAMG&HJUokLV2ZULwUa7w)fy@nSOP#73wl$A@MW)3K78()x9@#EwYtuLTUgVqZ@4C~wf?~@8ZyV=Yc3$!jHPu{wlF+12I&m{ik ztZnWNYx844(^AEO&KTwdl38Cki-$Xp*5sBOiJ~lufh`;AC>AD+MzoZ4r z&;=Ab;ok>CRrmaSvz40BVh`m{Bu(s#&{pS{cQrDA@kHb+f!Gcd1)xBHzBS^C+lMmb zo}jbLl6!Ul9iBg+LnsbI)prykto3N{An}S63xaLPc;apVFo|o2_vYJix|okA{B@wE z01fNg9P8JwX=o&Y`{Y{jS0DrY0b(;Cr+48XpE6E9Hzy~s4uP<1qtMD6^QTn)CkEg- za+Ckf*{kPvt^un4snJdRqANP7&oq`3h$GWy6b4k-(YJPIU6hvJw{Z`Ah{&55eDJU9 zSGM{=?~hu0N^JX|leHd0-=C~`uAk0ycc*=A3-9-oXB_ciEj4|x>cd;xK6>Y&>Mnz) zfa~kIex@ZA^^Q*+x%tH>(%bd_z|6Gx8}O=~tt|j!F;l>{A5!1`Gpj*q1|nh*dA4T< zP^HDi#Lw>v%2CLp07MlEs>gwDTpP-O)GWxb)jjh8Qyb&PWfRaoRxR$FoyFVobN?*Y$pAfB{_6+Y6{o2X#F(Q)7*p{a6 zW-*i+c0b3*Exp@tf&j)M_=wMDDpnC9io%kw;d}&ad!?89tz;#z@!;)(+%bn&TOWn?bBj z0%C}u|JZ2K#m%kzQBhIR#dTyrwclYc6Fhpk`*ZYl{JFO0B%9rTU6Tb+?uriXm~SM$dFPf=+piHE>b4Mk+?avsCStU`W!_DQA$ zI@<{p!Jf9YP7l6yXE^({v1C$K^~zp}3J3@& z!81Zl2rwCN%mk$r4sY~09?cIAcgOgwA`&t0j{|w;41(h`Gjo$3*Ct#VmAno|5J<6vqNS(zj_}R{P;f^g-%k*MRFDkc zf8k8js#d_VIdIDSL7)FIFyPQ;EV_kue#W`Xbq}W_N>FKP zXrr2p-S)?)ZQQ+=Zh{<-brj{Cw+sC~)?g_nRwDe>c#cE^i$35P9%%}% z#ks|&e>E#uWA|3)K45$Jl!xJ4}^ka1Pw7bLmKPRsQZ- zlHqtX_iZD;z8CNGw4A*Rs|w*<@P>jR!ETt(h_^f-SS8KI=y%-`Gul0Qm|H@_0lVeC zF?ajnXlu+F)y3C#4 zAh-#ji}Tel{G!a~+E_J2ejs6v@m#jL`)z!{{|`2eW7$K)W{V{pYFUv+r`0~-{2XHbOquk7Zlr72zNE#HMJ`JdD zpM_}kz)r`v{?Fhkh@W-iWM}Whz*h80ios227kCsI1 zWgL_6?5?W>8k4j>mL)&UGu^2n@1RO5`gMh5qHMb;uB%#N@YQ4~Z82Nj@rhMpN&L9| z{B^bcXgGenClkY$;+N$vJxsIW|FX-{0|q%q{pKr9p4GqlL31FkdPASD-0~sV1mDt) zeAv@C*peBZ<7*j*pyRxcO0XW zQs|Wn-Zsn3o~nzI4{_!$eIw#wPY7blwl*o(2#B3D3x{M4&|o<{G#zqm!sH~K0fJ#+ ztb6y!x%g!@=Ez$f+P!J(r**|sfB3y$;3tcEaSCqd1tPJguj=XP(JY~BIXJ2+DeZuL zLv?ZNbjE3^qrdU~y@%69rw(EryjwrOFRDvlx~y<6d!!H78huME2{G|83^kA<6UNjW zjAn+5iCOrM!JQphj_Q(Z3;A!CP-H4K2?^RA3aBcJ_Mi*^b+DkcqbQ=sBTxf*3-fby zd?#GZ(-l-?Wa!EkcUVxapl3z!1+8mnTzgF@_Pr3(Vmky-X4!!ZMon!|xYOw#-93hB z707SEw{2$X4gg@q<6=%kv|~WU`%bT$lEl2@T=DU#Fg|D2eR!SJe)n-i9!`IAW^zr3 z$hE3@k7-mB2)#>gx%U5bhX8CpgbJ2 zq|p1~3R<=SVJ{zT+8=#Y7hb8Vf^D&|PRN`uaxkzps^P zS(g)!7U1VQvRt_DDo|J~GgjzKPu2hsWMa+4=hX`lZ#}f=z0b5n9}++KTwlq%xbW$C zc$mJnSsMKi&HA9?Z=jqg=?%?ZY$PrSr(~91nZ!lpL3izJm%!$^O$hI6c_0{qd zW78)Cfn#X4vn=F#o%O|qM?P&!=&xPw@;$P(v&jAYsv@>DZtddXO=>{D0tkeiot&Uf zEV?KQ6sr8UnXq?ihp29T4h}{$QJbNXK1{!CI9TZ=1SOT8ddk`o{S2v53&<_@@wL>w zi;|JO)Tk!fW>FXBkb9J+-?BdCQSkvW&rVR+k2G<=YuLCzu-4Uq2$HIEttx&WAFtE> zT+hR33NFe$P^fT+?q}}9A`+1pL-;_&C;vA87UMS6(R*ej16sDlkoO&1a5HnoAICyG zhvcK=;&#ke*5&;aIN|k&t#fG25MbB5M5#9m3G}Q~mesg*+&KF3w4@||S5gfOaQ1Y* zV#kPsRtUtV+DK=8K%{Z~RKQUTT)qJo5?oD?zzTsozP!))ES(2^~CLcZy&I0@ynwkS9CKP_YO8 zLVIn6Ntmq>J2JI%6Z)QuYb+{KrIfEZd6IkV+-mc{1K;gEH-R%%Bm2Ft&Z8qgH@J!e z65Lq$Y!z{iKe;|o`cmB^Y%d!(S@7Lp-iw{3yZe@;79=q>nW@XV1ii?~#-eQYByMZXy zfES1o!u1T80vA2MXph0U_XZ~MkslXOtAx1=U0jcqX#7^*ZPTmh2ujixA-qP7fZ%`r``DqqO@vTz~Z(yVvLm)6Bu| zhU-|S@dfA72ohfeS4Y|lWUAGUS>ER09|%Netnb!Uqq!!(_*7QcXyXZVHyjB&BY9Zc zOvtH{s}8q$x1Zm_>J~!%g6o2G0`FB;OnUhScHNb zDe5+6K|E{BF;I{=`o^i^@ zh)VFj?%~Q&;`;n3$B$)uWAffXsZpL{}i~6;xk4P^1XEad5ak zc<=yyFY9_I)LO6>B7@9v$hkWpCmM)O7-}Ex`IVK(L+0`z=>CdPNsZGYqans3ZCWFD z-z052uZ;#f4nRQ*spHu|4-0;J!O~SU z$upQLSVEHl73QF$b{82;HSoJ~$yT1k{+7w<>xp9W z(ib+*P!k`pu|Z};YJV9y(a)Iu zJE*TNE~g5R$5+N#FXmHAic)cp@Vc(N3w>z$8v{dF2e|VjfecXhE zsVTu!vF{Z1X4ji<)0Q-^wi5d3`ppi@i_@6!dT{eC?tGTW$u8?|*7@+)>Tp|Yx*(^P zGJU2n0J4+ax>436pw8>A+CMNPPwSwDAO=+f-*ZNYF5n;d#;NMd+6u<%xLY<&nVJ4q zZ^|}`lkd~4e+#att@S%x!AN?4A^V7niwn}kM!YqhI%!Mr`1R@>XoZBxzdV>A&0yNI z3FlcyMDYCh=yGvlYud>jP*BC;$0qIW4X7hPjw)F%utIo0tvWbo-f6HHx;#o~7c*|$ z_xf=A&|f}NJ>>-zC{$3}beIj{DuO|Hut@e|`kudSH~Af3JwODeDAqYO@4*fX^>7H8 zIbR?T5zZjwrZ4DK0j!#cHs=f;@idA|K^q(uMu6YK+8`>ZWhLE`P!pe;dZ>g*?Y2r1 z=Qp3|BXG`HPL(g_K8=qR8l-_69i0V_-`&%*DaMob17T;nVqc6T^FVaYI!OR{RTomw#5u^S{5)SG_+x ztc>8AY;70GudmGubbM-*^6k^&nXEqTZb8K~huDPYlb0koPtJ8@Uw&_EC|-)rp&(gL z+9KDS(N1{U@8vV`^{afxdQnd6Inwd9LG6byt6hJo2u<6Ep%*>aAz_g?+hc>GKT8V= z!m;SeBG}7O{yEjHFBg?=k1M{gjJp|7e)RIz&f@b#vZS>_yFejOX#mD$EFNApc*?1> zskr7mZbmOxK~`2b&PPdbO3?ud8-QdGY5|DBAFI6D!IoE0px`lomytpJhI0MEF4>be zrSA2jyu5E z?Vwz1z32;*#V=tkkz%xxp}~QYib2afbEaQT0^G~2@>X1`bu6h6S0V4Xu9Ow7jYj38$8Ip;KFf% z9(VS=mrdCR2;`m*^<+4_qT*4pC@X7Z@H?AK0W;D+RO;H==$u9|sS9jrsj2NQlMXJ$4D;=0nr|R7}Cr@_)XZ*haLQPANb7ciK_p^PYF($?M^z>_( ziBM5c01{3r(oKT`Mg96Wo0TB@0Vypg64PO{251tx#=r{_C{9dD3Rn?S;lujjyuQ`vMITw4Iq z9Q^&OLHPk$ry>+(>h-Q5YD@m^Tw_sIR=}2-OJwo0A?*v0K!VcF;j@YL(ZPYyre4+TjS{<|W{UQ|Mnn{d5E82cYWySJ++;tdeJwyI+>qO3zH6y|UZQ}pzkx=crPZ2D8ci*{{HzzgWJ_(aF&?9F1BAbe6%up|DNBh z+uydprEG#r{Ysz2=+Zb zaI!Bw%qe1@*x1`MupHm{BCF?Mpw2J=m7wUDn4V85-UfCwKDnja2ooDDm_!*|@$w81 z?ySHz?se-9i=Iq{s!$v8>tU!_fufJBDcNMi>|@__9&$m9)FlFnx?FW_Wd-U|p-9qcZ-2SB zo5jV^(de`>_Wt53!u@hQB^34UUAc6nD4qeV2w36JiSc&15#}o38x4nyAOqBpc*eo9 z#~a}0VK?5}Y|Vf%P9fvp6MX!5ZFP0r7+yj@kKu_?3IUgv^6c?N2o-y z54)I3I4*hKDXxcI=$Y20?v44?p18LBOagA{OkM1~i{&Jwek&D zdOlnA2Q2?odU}3gNCh)Pwiz??@Xw!F2Vvm())M1vjvZU-SK<@%JT$jVPhG%HEbO#B zauS=uUcZ_w$te_wedT81)ytl*Mfo!8BmkR@{h{@IJ!(c72MZw-U^Kl~$W>2&{^+3j zZyxuqXhC&T1bV^~y113qRk-QOzm($*VWGUmwf!Q?*fPn-%lix&2$)w@=E~GUeH>cY z|8a&t!(LgAkY`x(`;wqfx%A0DguRcm|0yecsl42iFV=5vUVl*!XA`alp9)XS*yM2W z3nY4y$USh>LZ~7z9aZTmFg4hZsxVfQ+G(euAueWzGq??hf+N$~g zN}r_i@U1NMM2Tt~cb|f6D6rYm#N3)W{>HqGG@q&4)V9R^SiN_OTDbn~<>coIV>){J zJ9o*3fqI9%1RVuwL%0J`!vpXyD<`KaAizyv;Q6}}E@QxtdRCH!o~H+TC{JdDTBG$H@q zV0_mtWpd4y9^v_wT8)h`n^UG_pxa!IJ>y5dl3^r%TblI{U5LcaG>a3G{76J^f-wpe zp{zwps*SVxvTcX54=cFQWt*gfiu#LE%F@mb*yoYbz0pH~2$oOO%EvfVW_#uXkR@Au^Db&ViE+F88-SC2CLW`9vS|J=U$DntlQ421jv*0^B7+uLoGFf4?AFd>BL0fC;< zwWb@5FyA*O;7u$kv6YZ$fuDm(B`s_q4gm`QGFL(3=5`8W1){tI_*jTKz&n6z$Z)j0 z2l#0GmLu44c>s60}Z=AoE~h6%Lm(_p5&{CkaXq zepi^>sd7!foI3|PV2Vsre&E$CbgS@MjxcN-!kvpKCWRW-%G+^52*f!*6Hf7c!ypvk z*mx3yrP~V;wI`5az=nPUh#Z$G)lZ}u3hSUJQ2KR;|1OglG#P-UT%KWMe0&yIhV8ha zwE^hry^V|%4k`(WvaAL9h4uB&Vz44|d7n8Dp{0hPKvO-aF<3b`LN;!|)E*`Qv@leO zh+Q3N*th_c46=z*y>BBxNJ9GWe|*+eSqv3wVT5@0E!SDcVbqz24sQV02alm)9Pqm| zhT4^;7&QnZD~Q59nz7HW|JV9!P8%m*lBVo5ie=oc{>i&|Y<%p#UPOSO^|qxsqgI0@ z*}rQby9!Eb6(eP`)t_oAG({C{_Ut_t^wgOXC3W#0{V|ku>U3Ds*U$2b?>y6Q`|%Z1 z?aE*4(mEJ8+cXh^EqFY@l-6Dj^__rYX56!Zj4_8s$sno4FnoFh2e|_XK-W8RXQ&Dr z8++b)1RoeY=HQ6`0}xtUAb19$Dx{^8b{A;tptnP*0x?lWh~8JgFhk9Jfv(Ios4zxV z@^gbeT8f?ECn$9Y=Obv@WJRCccHaRhN%buwhmSfp18XU>NPhQZ!i@! zHHm5d>m)T;V6}Nb7X%tE1jEc$(LvZ-M}5lUO4u|9C<^23l+c4B@{h8sq?b_Lj6 zhvae!3w|U&LUcK`YC*_F{+(U@v1)?<$@V}hI>y=aaOb7XK>OI^>zXa^unP@o4Jv7# z#-ln)2E5n%dj?@4Nq)Rv^so3`vZ-?icA6GX3gn%2+wT*;1y=jMwu{T9s{ScQv#GJ! z@C~bWy=N2^cS00*1Be}ft={`xi#rjCAt%=jC}kg153lx1*-i^DfCmQ>FEA|3e1?ff zTNqdAcN zwrio^%UmvB2h$LkH!CQf^s4zXNCz33n6#5f8iR=h6f!i}#Xisl!#sp&8|Wr8g=qZF z_?wVt2Y0_;+S1-Njcx<46(nX?3V}*3Sqv5$aQQk_ef@g0zi((jYtGEfY{+?DTJm50 zAym95m;zRVp*j+u&C{>Yv08c9(W5lKHVBGg8M@Vi1BJ-WStb-lSu-ni)U3(nSP2a| zdR9j8O@%9T27akY`R&IuaJ|_1#h`RI=T+cd^ccHFFQc*E7FG!+& zQ;lqcTh$jiwZ1xL!03EN5#yU1McYNi3Gh1P_wH2*m3t)fs$2CK@x;+OwWw=Tk)&{& zVM>L;IamJ09A>sx>Ozc+jLfvz=yO%1n#n0CS%&4_B!3;OjrcTp+N-wy8~PIV5j9AO zX2pxo6;rgkkooCi#gIW4vj_ao(N-s;%oWYMW#Qfds0p`b24*$1>Nkoa^odj%Q zC~GGWWIQQcvuNI!$l>{GSFnmq!luh`4y zz#jyDJ3F0ScKa9e^n*~6Eds^P%~0n=CA|L#@)`ims)<76rt~xbjFkrrgzU8q8Tp5x zsC(^A&~dMEmrM}UUV+}A#~Cfw;o)auXa9jX7fE2pbpGU8{vD`f>G^3qUii-&pfrGc zp!asUx%MIR&NB=|F15F^ia^C5+E0%{PEAe@8-U;owO@stI%Tu}do-5ulAeM%yTOBw z?jbkNLzdB*iI!i}3|p_bMU#W_0OD}5`nysXf#DtG*B*ixsO=4M zzYPUr_n@b6fa(;Sae(%MNdPcdY~m1CC@q0QroX%Uj^$vwX#e-(mT*YIfvk_V_I4=k z$`J8@*o~N!7d)r2fg&K*x_kF-<`{yHfPO+a^GV6b>Y$t*N(|-srQX76S8;Kwtp)>3 zTpWI1AdM+R?SQerm>N*cj#^M>t%_dE`bSyqt**MjMhT{A`1!LL&_&=rwt!q_0!`^Z zMMVEqyp0KhjJG5;B|O0wE>H!VXg@_#XJLA^lw7?B%YVl;OsWX!c|r!z?PkmF8?+25 z^oDjsDlqQMZKqJ3F13w5DBG;JI&39OwL8w3QP(UlR zQ_1a3KzeKdU<`PLm9a6Bh1vTb@u%VyVCwEQ#WM!QZs~;rBk)lR6L4rkfSXYmCtW~V z8rWS=Ai)A>6wceu%2@Lbw4R97zmTcG2?iD*7h&?##!N5C%2GQB3{eV6aB43CtC935 zWbIu+$Y=7$k5UI}Z4j*X!4{mv0AL}6dgW?RGHA2nvT~~#9Qlv9Lw}fvt<=MN8K#T4 zx@Tg^P_2hEZ}+vw;wWx0)WQu&^{+R3G`S1Y$zTd%fYxzYJ2+Jna@ z=_JoQ!sS`wrH1+-qiykEkrv|&%OJ8LwRli#Xow}+?|roW&~5{kPVFhz&BkvW>n5>{ zxt1Cuxgm<}Kg{J{zO2L3kjV@rR$F*c6s0`Ng)7wFO?klB01YsppQ!?h`-z;~GAsCj z$VwnPAv$`H)WAEy{!ta2b_tIWGdMiOUJz^mA964Xf*8?kgcy2|M|tzv&D4RA80POq zP6ja%k+i2mrXL%Jh!ba^LMDulGN^*3(|dA$qO9BMsf~1-*}I z3vYevXble}xwCa=ymp5w;Bm($2Gi2nnS0?-?VThNJqElN>Sj$4Jyc^k`jK;3zDbqV zlaD4rUbtqvIHx3M=k1tYUmmxKqNtaEvoaO7V@e18u-KYpL|kyXJ{6BQZ}o)Ran;YU ziE=0YUH+zn)vFj5&z^nv(&OUaclTKb*gf9xX+B6mb@^*I(@Pk!H`$kWwU71)S402G zro!e2(X9#TSq+uEGVL1|n9{aG}A_43GjGNRkKmv@M9TyIgIN%Rg*#6;o%CtiM32 zfFD3}M{mLR;*Y=-f4q&jke;WtH6jGJYBSww>IkF~->O0W2OC0;`s!K#;zXGDR9!6J zWn4(@Vj!N|gf9e^A)IfO{pyMmK2D#8KIkX@y0t}E8s5k7vzVmZIa|SUy_kcgD~Wl% zw{NpHv-Wet((AH~P1*(4dS*V-aP6%v3tJ=mT2hR*dNq~fon3xzyNh|1sR8dhmW1*v40vrArVwKAU zD*-}DTBoJ2IDVP(V|`@>{$I=R5tO{F{_c0<3w8onjEB5@ihL9L3O;Oc@IrMFE+0?viF$ zV)Qdz)v!-wT(eA|ELKDo}uxa|@RzPR6O zdeFST>kv^Zn1X2JOtsXE>_mmU^&x1T{^2A0e|@ozOEU05(e6U90LZnE-Uk8*s_NzL z-q2V_e*@}(iJ;b7z&B7s@-Ub|AsxyrX_YGbGUCDi<#y#`8f2zZe)725Fro5M=DCOL zEG(UJH2=ss_~l!|RQoTTt8XZO!qlrvK4|FsTU!JE`(Z8l zabY- z0mfkv45TbDkxwQ zQ9ADvCn8Ef=Ni4B39(`UHZqv{83>-r$jC@aN_u#lgX$d|rT|_7d57jB4!mtO$h+L{os11KNS$ zm%*a|LuiF1%(3RE=-_h(G6Dp+eNgH)0`CZ0a0kp5%HWgyPo;2-2oDdB+C6Y$^K)_K zYn9!P2rcnFWj`PMPp}_)wz>KsX?{|FUuWL?uTyDlGwRB=x5xCPdzu?=!u6@Dgg+t` z)EPIpS(6fK+(XW)pa0P!B;Ux{_00**i5HIjs#rVQ5Fpe#=>ul8hwdwlf@h#lHGhf{bpXvkpl z_}$=9m6Nj!V1CdgJg9NWL+kX;#YpD3^__(E2*)@~29J9pYcXtx11JBF|f1|m#VXomx|21u5B#y?~w zL8T4Q;+dfN0NDpj54eA85u7Uwotd-GoMlJ>yI^{g@>_{$4ak)+=JEh-EMttGfAC&D zV-bAZhmi_&${8?3U~kGg1Mp7j%>h*RX4+^AV+OilCj^E=_#`-JPJyBTAtC5|ftY}Q zkj(>T4kuwB9(q?8wOIV1rxcv!LE{!bVwPS(g3W)xH6r=M{O3EPV-FAQ2*zgUhocP! zVb*+CcJ%J+*cdCrUGqeWR6^2^4@jOa9;w3Zw;a&a_)k|Bj@oL`32A3a2=1J{pC&4mLwTbii5#5fHG{04&p1 zJM0M0v<%XY{ui6}tj&FMwob1Tl3W07Az?xYhW!~R99;?ky)QIZd%-*M3*tfJKnSI% zelA=RdutB&B9xRJ><%dgSVQJ@_L*8gVh0Fx6`OfMf683~f;p8)#7$r0sQ2SG1=F%*WU z7dq^;ppumiA_U{iYY6i)Z(BGZ75b2vIQPx@PKpW7yhcvbz5g}87UUmjPbMv0R-Wx2 zrG1s+5pIuNUIawLod4#gn8hG-aB_j?5$WrFm%Cqo;oxxbj=B`t*SjpH)URhhzH6P? zRrYNO=OQSztz=pb7Z*2-Zy>>EpugXQFrwm)2yqjZ^?ENK1^5?fF-3zXHD99cP3x+W z59_gw0Hp;Ntel)2;B;YOVGxi{Uj%5{WpJcHgFF7ixB=*pL~&QR}FeVu>p(`n$#N z(onuGX{A>a-DK`A?+yM9@Wq=}u8TynB=mWIe2iI*!sRO}%^n~jDqNBoB3^~LjA(&5 zt+`F4aiHsNHGbqUt-!;I{pT>FnTx9c?l}aQWNK>q{P_pqd+yPU3=eZLF2h|xc1}gy_Yc`dK`eWJL|{vFQeH; zXMm&+5!0SnuofA6mFd@qpl4Yb#R`gvUDxk5!R2S{k$tIMU8pM@^PihK>|-Q_OhZGl z_DFe7cerw`divT2l5p@kvr|Q}elHQX$0HvY6__yUJ1gc?{?Rx&C}9bI3ZnA>020A% zctnjVU?APm1?^AaCWu6d?K6`4Jpy0A;6mwo0X`kr^1&7Ky}sVW90oZ^FEr{(hV?bh z7xA37v?QS*LjpUqx|GygFu<`60h^;UNrl585ps5Mc3`psyF&xqmH-1IVk|B^Zj+Id zXUQ)vF2ef?=`qQHm9)vNmB9Nv^O$JA^A{fPFi#D=#MXcHCbDqmEcJ%AL9>@OSjeoL z+zPaU{`+fFL{O+IKZtwUp-!miUYV zQy`m!mk3G!*|TzR48TLZI`%7su?_v(yXil2w#-yhS!nUjC400R3Lu*ZLAq-td29MTf8qOTbv` ztXaSx090@yzo*WCOFcbH)9y)DWu&I91ncqfF*y1V-qZd4eGurpqZUsuI7g&oHthj0 zdi(b60msDx`X}!eNZ_e86X^FiTz>|z8>16Av?#!hn4*oRd9}aT4!KJ`9Ah(meGz%x zo`^t6cm7FfDvZoCgo_8-F+p<-A(g-&XI?j0==bVyhh4bM``p=tmNDi(u-!Dn^s(ai zcTL&5+Nd3kL)uOfZaZL<>B$=?dEj$+3?#${LNx_I67ZPHI{*bn2!enUSIH2IvOmBN z?(pWnMpVnf=(h-x2s0Qbg7~^RJ831sYS)$y4Yu%L!WKd$=&%6CLGm)VNWZsfX5NZe z)sp>xhvf4&330u-s{oLHDHyGY5-9nkT>NRJtm}E6?$%9|Hk;je;R{gg8fE}jQ`|4; z-|Xa;12ic3!&@7w0MbE-PW~=VgFj|%w-prqh-PcKK}3R;=dYH2IKQzY$H&HGaQ5{% z?)%@S7o9-*y(Y?I^ZE+J+<;i!vVZd2(9*K^!ob>M6p?hw!m>n?!eu0-qJqPJoAt#& z6w}^Zpk>E>bQVZNeGlxP5Hqp5F)?sNT?Qu+j0}Y6DEZhBB1=RTN!GI{T2CI6FU>$; zgp*AQ(Yd*ce*f_iB;AM+-AwrWxt>hCBD0}Wp9Q?oIBTSpYrPP}5&rra> z2fxfZ>G=LX%H9N=%f9UzRmd!4iWJHmGL?{dO3E0LISNIFh{!w(Ng^sknPna$B4sEt zM2IpZg=8ox4SQWZ_xFUjD754zAjKken>$7J-=H(C9Z*HfsQZnGlqfb+ko!|VF`VS11d>*{iHadP(m3?QpRC4B^p5hOl4a&OjW+vA0QyeEQnUy9bZ95QI zAj<#coglwSl@#!2cs@h{F9Wz`h5{fDe&N^$qa~4P^#J6{v=Kfw} zn_F;eFQS%&GdwvBE0*{j`uAxwhna^Q+1SbT``4uui8%AGr2@i|A2`%}e20<{6Lb!I;M6SH_dG{r9Li&3?^vdU$w1B#e3a5%y1l(ZRtgCr)gq zj($>Ko6Q6(<9}*P^5HwiyrLqENBAUV-kdyCP*75`CN2GSs57dd`J3sh;=U>R0U0T& zfmg3!ZW>-tASBGg%WH=^6<{jE57c64NSs0C=`Vq%Q5nI?v)J?a3XkyHgF|Yyh#>3t1@?4bv4rD99{&-#Sn5DVwQFX`l3H5opo9#_Q z11BmTIKXy{nAmNoujk1C!@E=N0%bko7F{_R%~l3cY3G+8@Q%$sg4rh45{p&pwY@|+ko|C@XXJ>)pg(UUy8y=Fz{pToPYI~u(1ytLIv(4z=PuCeA!b!m^89p zFa7<+pzCf?dCn7Fimfz%hDt^sq(ufav$~5M<`-3}`<-U0Mp62dUdH?R4V3Z)&miU` zQ%D#1Z1M*27m9$)=TK&GVE@gYIzk5S-~d)Vd`#T(!AsP3G{hjf}&QQD@(}p7BKXQgxpuU)DnHYk*?2;3^KZpdDSofs~6*F081$fj-NR$JF0k^%1 zlg!6NHFVO@aHUUc7kHy6fxyc`L$PC+M31wXKqHS46A8N@Fe1F3VPt9%wDp01n(FHh zzRpI0gL#eP2a``W`~wB4k~2_#ZwRPo@D}aq?;Gy#{E$&?TFT}7b2+Zy{@q8t*ev8EN<$B9A_c4DOk;V_ta=l54O-TZ|gW(KAPFeHrGU*Z}~i z2TcET7v&%NdgThOf2_vReKY!JCk%46uCTkqObo6MF5?ddpFJzWfg*_S(Q2-nTe;+z z`&xlIv|V~{V~IW%9FGl^Z~&=+X}fOJ#I`@{jNlJGA>#xso?)vNjBvjO;3r z%3*p1xxkd<;b!rZzZIwRl{4Q2_;52_+&)ejC?oyH+QazB_N-S5^)8;CcQ$PDb@`ul zW2>K3Q5x+*RH)V195}8>!4%#0^r_|^rsG{b9EP@#5r}h zc=&^}6cH)X>BRlV)Fa5gk#jU|>XDVW>MsqZ>O)k&N%r9>9Qu;WT)y_G4 zC#Y$DsM^@O;O$l1R<#i@28%N@xCSnB=Ce>HEZ`R@sq2R+* zuP3xU6xfnVQXJc;FL-XW>LZ%dvKwHR?A{|{2K@T^3W2uTd2nw0)lek)zlnoRhP_^$ zH#sV2p)xoA>GV}H?R$`uAc8`~_O2}{`NTSWncsO`;^XSpx2jiwTcU z4gL%54@dtCd*pffF48LYNZgAXheZ}Y-bjuKrC9fqu4?%C;_(LC-MGD9cD{|m@-CR% z?{*bX9ZAZr6kojjGQcP3Zi<72+2{;uf{Dd%-TWNekB&Ro?)?7XT%22@InJ%+aDGUn zWheV5IDDU5$Io!}D<-tm+D1gKKDhX~XN$dkCoLTvl&fTyF}54ll%Ae$jY;$^31pB7 zjbI4>o|2s0s2;skz@7*zc{!63x_4XBl^}E4`v*I^kfXxqE03*4Z7(wnBNL>0c8ioT z5oC1*xa?ma?j~?Ar*30rez&loJM>%ZU%k;&eJe4B6))QAm=zrw&m+T){k#Up%x$r^ z$1wWBpBRw?@ev!+o;w>dJ-L+9F^3U%|BGJ$(*=)bw@HgiPvx@{NuFWt;tPODyGGM}Vbye6*cw z@nDuJzwy|?&?(1kk8=DJFRHy$yXvELu0K<`&=L|;cyrs}f1Sv17)>OVSYb>^Fn%9B z>Gi04-$$M_b>Et=EqQD-js7Kdz(Mz}>vXy{pX5#Bp^eJiUrmlZ3zbtR%Z-JT>wk>+ zKWy6N5!Cj~en!|nDY@%=c7pimt=XPH-#ewUvn-~0!e5I${*6pVK}#wP8rb*dPwvp( z9GLe+@@LWYCJPfWu_VDB-vc@PQQ!Vo^(tA`=6}AzKYKw4^&I{uO!B`dZvXpV7S;v4 zfBJMTL@<~!h>oW6yY2PqF2cZp1E#!*@BzUDFYG)-8q|p|n?;7epDlM-N2jZ!BghT_ zXKNuKjeCX#v9UB4lKhq_n6M)TN|MJ~lbKfy(7@uK!^Y&5C1f|lpa!nm!mn8~j>F;1 z8PjpjFFVS-P{V-vU)ar$3q-a#SYJ?j-#-1Z{yAP@VP|xoKnuYI%pqFV)Ode=|3{eH zLt7UvC|NS)50{shQ@uSs)f&J2rcuMs?4OD!q3-2@S`xj$qkG>LMz%3ARRsLHH2fXZ z2V_t;dr&dhK#G7l;)J-kxajEMufX5oNKP1rqpyVP?4|pIEgX!(w-kI0kMuN*Vx^2LP5LAD;tiJ@myO0YW1qwPZ@l8yjzh^W}@_njX31n2Xf9d#(;Y zPSJH%azga*jMr7-z=8I+*trcz)>(*_I8=?)XhIBbfB5iWUnERB;A;F>x{Yg_g0D4= z_cPWCS-Z9{Chr8&?Ju9L7-;d(#u8MPoo0K^6tA&eb1B-2$i*{JAW-?)!Y2rbh`17D? zPSEW>V6ugoYMIIJ*2KeKU4L(<0;Tvs<=DeU z2ux=SP?Lr0LnXu_Cm;CxC#0M+H*m9FRs@n( z^Q*L{6AZJiA`@%Ypr2JYfWVh~k#_UAh2c$48olOC(0%RO zb=nsf2?)4yc-FYSrl61l%Mg9S!pnjrb>867(+&C%5M&gFeu$sr!1>?6VmmIN z^;{UkeBAJG>i8@yPgVh55y)QZrh`OyaC5+#BE0vkswKZV8Dma`25 ztU+NF7W??6?@E)VnTE153F`=+FG{$pvk>euEInFh?3^e~Ofl*r`XEw%M!cLB8?cq-2_E>81M>v4BurA~d=UQ1uT?D4EsxEACR+Ov|nbeYe)0ph~NHj~Q@cn$HOn zFm!p267y;a-yz1+0I6qZXU9EcMlJR0*DrLDF1EItUgPicso(mz5*ws>$`Sy1$0m*N z`Nyr3KL`Y{b|H_1T`HTI(p_Kwsxmi-qD55C3?7+bAh(0l^78WT?!lDn7#3eU^UKPz zhpXHX6i^7jh)TyKAr*eV`p(k9tSYG3=S4u*L|3Rx1?Q@ErPBbvnY`ppnX|0U+z8jO zZh)>WEb3GeLATNZmBv!6$TX=r4phr%n| z{)q%{<~o#>DM{4|qDe^8!(-#PR8$&>K${qH;F6CY=dRGIBPo$ydKzW(DAz!4cf)HDZu8C993~h5b7rve1_La=p9sL)-V}Z>hc> zz7Ebb`�(quOv*blqU}gs!yXM)>i^>62~xwjboBETIz_IupTC01ac52oll{d_3^d z;3vj`g~0w3JGrvb(_hEdP*xC%kv7tfT{m|>4YECz-T%v}=HNN{0oTP~^ZncU$kRi4 zv#R#e)x^@+lu<8E8be;x-94j-d@C3bUICd|IfqYj`jmF4&1uvgNZqO=p*le=kr5F) z<-KWhIPlp5-zzh;$!iEFo@huNu1`P0XLz&fR^HQ9%BuDjQP#sxcn_mTgJF-sO|q$D z&7q9d>l)fpuO9z~sG`j6 zjg4y-t>rwUUO>McHBp zH#;jUW`yX@DQ;#j<{EtRQ0N9XWlqpEN0V66?>`S|qqVy(=g8gY`Aq( zM%)I)_+s2Hwy_jkS5nM=KBk4DQ(U|hN!{wf1t$axGfOM0H{j10i=^R%=Ag-O{`_ZD zXv%ejtww^@cAFOqRGZH2HgR@wi6bFdg)XNSNgklGS{FtNM@lKq(afYN*VGa3xk^%v)dM2F7+cBr|9ECr9aW&}}j^II=wVa5O$ z;0-FR5dDDVCv0Xn4q0}&npV~_fBn!G>a+|59R>4b!&P%!t2Qcy@MA(THbP`1xLg{` zKuUOu8O8vbvhN>Sxxq=Vaaylk(q;~) zQNyCq#~nTQzI7m$SdKwKG1Y23+Nw!^n`<0tL0YJUsOJ?-UdA!+a{V zg>4%_Fsf~copmkaU%!083U^m$XX+>K4mHHN zI5`~z=ZGeV;9XsD6roC^lzH*&*<f0udF34aP!o!0)aT)0VQ}@>^!=0u1`NS9I0EP^q|e{HL69I2S#d_pXkZw`))GUA zV@bD**%UoA3m6{Si170A4yQ^1U;l{Zp)|t!N>_c*Q=d~ke3(k10p$u(hc3j z^*dEGIR)-2iVhf?>7IyUPo>S=JvU+FxPKetW{Vx^7;e@owuA6d!B;l;D7#XC0xzrj zQ8V!x=9W-IMNm>F_}JOm2~USa#LE@OVJ)<$*;@#Y^UqX^UPquXh;&1NQ+bi4#X|I* zn=_@9E9OUGjw5~D&s}JPRha5-9E<<~F=}=2`MtVUvg1yp)=?nKSy^s&+M+8LJv^pf zZr4=wnlnz+Cd)XjmQ`tp?Qpw_LJ!~7#=^4aH6yRX3rB@bHW@@LcwFwzh>eVFFFao4 zL;cvg!prty6kiz+$cu$f#!)-WXdfMVKy9SOz6z(Y)B~K?-$P2EX&+Ky`$;?0Maw_r z68xCjq=&*Kamp>rOZEjqQQP|q8TRF#n(_Ab$U{H7nk)^=;rRdgTBxp%H# zH#Rm7D^#Q7-hN=uF|P%30{U2f|MwJydR`Pb7-Q}3-0_lC?Bqw0o=cW9Zf{Z5h~(N6 zDZAjyQ=_%yo401))AGQQ^}g{$A*8$v3=B)zWMlqhEEX?ziETF}HSxfX?%sKw^vnp~LZ7H$8oyPi+UVOmFk_Fka`G5T#mdl;? zjq6Jn9N5CRqh|lLyeCJNf^cKn>WxK`fvKsUo*stnd+~hz&w6x|9fu9tvpe?%qH6z9 zwnkB#Bjli^O#5R&p+o7OA#q9*6epJawQqTRe=_Qc-I}>$h|hmXso0gHa=ppnw5Kq&NA0 z_W+y&8AtNTJ*aN}{q8!js{^7TYL@Ap94BdyIArwQyYm+UvvG+HM{m zhE~y(Q@{RQuF&PSJf)HPVCojiDfn&<9KATD_OD_(dDB$S*Cg9pvhH@^Lh)rI|HbnP zRdjW)*ePv)ilJa2LzoS{ZitVR@$X0CuxRo>sHoU9_Y0!MTO(+RN=U%bZ}AXYH1`{+f1f4^>M5)c zz0_9=ohrO2veFmTKWTn#`-KJ!wpnnraq{)8uXv`~xCOSIjukL1zNs7W9*T*VA;M&5 zn=pl;3loBD#$-$0DH-LW=qFR)rjX3gISi$Bi7_v8WOm;3U=_0fpX2GkrsV`f;-9Ej zhHpY51?3JI{OHE z2Ji-!FFS0OIiuEYlhV`G)%uaL;R5=4OpQ56ZZDDS6602B^u%{|w})3XGUdP_rNPZ~ z0t}R(1D0NObwR!qYp!1%RJVJKi8|a-k?G=KNVI$!wmp#uyzsy{KYrkepC=3rLsn-a z$J~a!WezT&J70Jg(|O%cHNh0&qu%tdB7#zt95$|f*Hcm|c}nVLXQI=#a8P!K zZ7Ez6$j<-zR_1eaM2?|Qx(}QDa|-&RlZC@ui5=8(hnRM8>{-;=G8Z%-2lW#b2o#Zk z@Kpn>6M*j(V6NM@Z`U7DH274;Z*%J1WZ{dHRpD2pd(=p=ru?5n*rM%XhQ-sgBKLp1 zxe8Sw@*r{zBnXD{u}|!(7}HNb9xXe4iSr01?Jln`WnlwybF)&)_t&9MRh`H;tK>}C zCg8Kche7$|sYssYpL*)NjHBVlUq0jH@`}xchKHJ#R=r!~Zdu#L7K+gn>i``6!@Fs= z?P9(VVkB9&7iIc9M$(bPE_!+zeUY*Va1%nm^cpJQ{5GZ`*VJNkkHjDXRKole2HqDg zoO_v{7#jLGu#m;iZ&#WVsAn)h*bJbb1f}b#97125TmRimCXDdaNsWkQW@Br1?Wv~9 z()R&`glmnV+2fqsw@-#Z9db&yd3ou&{v#+T(b7%|`h|WQwK+f8`EApa>gh5%8X7ci zHo%w|whHR(p+0YGi~Y%0P{=i_I4MbEq!&1pJEqoO?khCVJBFh1vcEsr8<8i2>ruIl z5C-sK)%w;=JC2fWq1FGXow#o~SXn@pFVvGx><>C&)Et88j{}<8t=D#Q{}!$dt7bR( zzV9PE06t_4GWw1-Wek=K*@pI(_Q65GQYeJHL*#zTr$b=(u|^twm}6nl_;^MBYk29Bj0xw*3ALQcKf5=gP}r2+I8QUpJHA$ zOw;`PT;^u@t6+MZp$@Wb$R0$kn|5@^3BG#s=FVC5wQX9+A%I}w+)_McO+Llkf3exC zX5!>;Z1qUdanU+@*k87nPyKt@+rSa&5<#rMgF$f}^JG_8>{X9vCJidxTAG@7Aexqg zU`?5wWo}m>%!3KNHmYZ{PL)`j(+FcwSk~6oLaf4ss`1fd1YPEh^HJ)&vJM^|k}JX4 zC(O;~K76oItRl4wZiW;a-8^=&C&#pu$Q8fVg(MNH3mO>;V~+QhKH8B#N5CEO@=~0z z+r?05$8bCAvX zUvvAptAN+=&5K0kV(s7>u=$M-eY;PuF2t3|7B3|?G|0;@4jY7;Xx`pUuXXzrg)(W@ z6UIH$6Hj#2&!#%Lx>{Aoa;Pw1nBIqUIFd_5Y)qb1+yi%{^GT3^f1t{YohD; zaO@gyud6fz&1vplfXsAMR3G8DxZ@IJ4v6eqX=t$2K_C7V0Kr@Y_M&3cf)X9EJ(fe$1&H%`7!qK}*!@ZAaR8hmds%{D2@idUaGFX;XWe#F5w72hmt%(xX3AiKJ&kbSBs z$++5u%>MwLa=Z&KNh;BJc0aT?3ce%}LH8kWh?=?j0=*j`t*AlZhN zR*l#gMs6_yqh?H%VF*HE)hbno9XFLMyjUOpS)N1=;VnL zEmUypa5q4S1|wN*fVWTMxoA!h)xs@!z#MkG3jloq8X486t|Wx-1oX)nvhKY8jD?Zj zU%q{d^nXm#7Z0`&k~lSbabZIMOdt&~{PPF)2@r(a<>)6t0MOiSUhpyVX0LPcGirOT z%?EI+h61C%9!-O#rn!`a1Y^t^rbw~|GhMGP5E>#>jDv$Lu4m4jBZhc+GgYO-Wn;$4 zwfOfbM5&iKFMw&a1po#fE9sKh?mKx88KT8dDe=G@c(Ma<78n87r6@Nv+*d* zVAzt8&;%PCt63>G&C_p^IkS~Ra>Au|vm@)39jCXRa=UmjQt$Cg9%{;bOl;@D_iume zPk=L&2WUS7#5OBLPGE>dC3Fw%O+DBKl$*XQ$Mn~!=T%+BFU3>O$AGPL!4}BfD!}A# zU3;1|sm%Ab(RbVNwoAeUG!Vb|@nff#Bi2|1L37PN{F#$7d*aMG_2f2tu1KbS@b+cU z`#8GSafb*>8zFJW4pADZ-&Fwnm6ZlDXc|kAnoWdei(&CanYCMNUvE71FQ#WCl;xb< zPcY@aJls%}ZN8<_jZ9ls!j7YzocH9dod$7xb3=;nCifLvqzF7$rz9q%wWEJU41uRX zYR%??X<6?SjB6;r@7u*F{9)v(Y;Jvq-u2F+ zb!%#cskhY%SOO))g@l475CJn;dc~k(+oza2Xa7%@>FCP*Ngy>sd zzJ7j{b(D0@PEOIKy;1otLx#H5jIOqdNBEhk6ZOo^M`MSg91>Y1gR6HNtEEP2^a?t1 zNDl>imp__{nnHm$thF{z_x186uil}kBydmqpBCI85~FyYe#rmIX|Tz5E&FOI=iQ>H z;5GfGQm$Q^3I0oL?HhzXBu729J;dJO+zu>bPlp|Ps12z_nXQD+%0|mVe4@vZ<0o5K zxKjK*Jy-Dr+S=4H4^px!(bT#_SNYZO)!pYZ+1EtpU$#ZO(K?kCH*Ct#fXV`Ept$wK zB>YB`zfp(oEXx`z=YG8@)s@%t{VuYg18@zBIh5D2<-4tAM*V{s&1sdRH3K{)KSvs-1mDk%on!YQG3K|Mw8O{tYyyZ>Lb6kB95>50bcr_ z5LI`{I5Tr_iAHIj`8B?r@|DL}Vc$Wsch9$&YTUm2o6)#((V4P8086Ky?Bu-9UYYu5{ka50!R|Wezz0eFfLs9JU#6R>I3Rb^Ajh^4J_Pl!d{Bd zPFB4c7%*Di+_~6i7l3zVWwnzvm;&RAk_llm&?rpHMN-`vz_sQXQaI4&6}MiYhOMZ0czdB+;B&schA4v=47vOmKe5jpL;2^ zzz!v_cEs_dx@NiJQ2Dc8bZQ5!t0)U!wX|%C^tQ0r5wG}Mg&_d#9}Ir$>*^-7XP>gE z*2s<^DCO(KmQ9|8z#;TIlW{Qg5&3R+s1x$aj=E+}Ypy)VjxuOqhHGbF`#Eq$S47V}I-{@A`B3BKXh55_@ zM~VxZUa|kpDLR#rKjBUoLq#3^yak4o=|b#ZSU5Co3-Z;1?8$skyB9I zjpp%mypCchOGYb>BRoM?0R?63Ws;+zLrS1hfDpJ>98acq5}bwWq5J84<_9)XUKcL7 z)%YgXoyusx(R5=RmxykMY#BbRxXwQC>aRG;QZ+VLo_P_lD8y*HijIRI@up`6SKEA7 z(3oV^-e%YZvyLu`2U{+OWmA!bU}yqRDkVjUoNQu?kZt7HYx^kYuwN1JBHP7|U9G_h z!wQ##uj=p(?*R+jE0~j$!yt*dxc$2VHFz0I?12Mf1Vppp(c`rJF4XDm(iN~r;H|Pj z>RgYKMvdWwJ5tWu@vx|9-t+u6JN^|cB5pfQCmKJ$2;~AW-8d)33_`DgwNjTn=w)nR zWO9cpV6mL!Y)kDqj^iP6hyKBUnX^qdW-~7gAtqdzL>R{r@r>J zELcoTlA!wof|(y*oWjV=#Meaf8-s&e{O&6bhCS)DV5`u1TXDd$>{Z6{yFvK>$L~&w zo0qfiUR>r++{jZoT)`HrPykUzNcsajCLGaxX?*L{=ZtjWdMGUQoZ_kV6cIG2GUjZR) zl9HJv(pxWWwxThEht4iV&~OzhV_94=T)xqBA-@!kg{ZF!hn zsBQLHVQVR4lfVu%-3&!|ztTv#i`suO6sCyT2fGLV@FRO~g>qhK1m+U%F&8GrDe91??^I|^75R!POS3GV#mjhv&h3{m%V#Cl0GZpa%$-;SW=Rf?aeJ0Xta0Z2?;j;ge?p` zSCjA9+hqQBx2ed#!j+222UX7UyR=%%ge+l5Yjc&PhYj6JW%~Ct;i}KuD+;~bcz&`= zJ2H5`WaIN$JQjZQj>nsGIpSTFptc5+dEQTS*p=$%mncwW*QidF=nh}D@TS{nOq}^K?&k+#au%k&u||yq)lD^72a!x`=SI1>3s9>J|=z<&$fvLO~*{@7}zS z*|yT+fBE3Kb$-_}?)Pr8UzjhFxNf!mJZgXbZSl+Em>j#yZ+upEr_r5^{*tQ6-TtsP znSRbRcyQkV%df9>D00o{KaV6hIm&ontyny~aFFZJp<_zBF-d8+2dIKt-4c{IHOGR0JJ_`zvHHp$~?lZoNE1I>(N&Mype% z6CRjHuWIi>hac4f%no!|^jx%jBeBd8N1OkXk?1hjH#9ZVt-xh>fqlECG^i?^jOLs= zycJ+zKi9><%@me!E%)}W0R!mfR#p+K)<}g&3uo=D0e%(#31D&4)GyJ#gJu$G;BHGN zYz|W2f>6YFoH{D&J{1-p5BRWpWmOV1JV1PJGX#2IA{}+Em;X(fiBvQE;~QpvGx`*00<_qON@T` z5{9WD(+1-lfH@kZzE{{M;2j;6l9-ZGi@=RM+_E_|Bm`20$Z7gj*em?U&?gzL2S_2- z)^8zA8GZZKdSr%AIPE=}9wZ643SfF+5y4{=)0-9NVH8mYzhY?2{ITg{On0B>`{xj! z-UCfE2d){~kYrh(uxM zcOcx1m$MlALfjfay+KIzm>wVSG9Cd+f2NBw98AWXpG3xumJn)Btn~e7m z&wIDEwM~e(O06+|ihpz^VB2m=)f}F|d`A7Z?8#eHm4`}HjpfEv)zq-x^hfuBw~+X{ zNnyN7+Od~=>(&Ik6%;Ht1kO9i9ZQaCXxPAL_AwANm>`E}h1e+sW?$;`bLFCk9|UyB zPOYD{wQcDRZWG|=Z^~l=Th&O~v7d6vg(t(v09Ran3Apn*bn=*}>NYA{1>Ey^oW~c^ z0?hID;8{~gn|k{yS(r5dKL^*NXKH!?YJBJ&@Mo%Z;ad!iJvGb%^Dr^JnCll1@H>|H z9{FU%i}sI#V{QqIJ%D!YQ$nxflQnh7>(r@fsB{X8iqyWsuMN@|HKvPvA>N%mJrNaf z-kmCqhck8|xE20giN0$`*&5%G8(2^J6+?2k z(T!)mU4XVx2^OVjSMdfUqJc9lt*v>3^zw*SI(N@dXp}tnH&%V|q+RQ@hgbtVH5BLK zzP75sW0ns{N>)IQpy0dh@$c8wXTSaKItm3Fy#HQN!d;N?nPH}0U0q#1fK5htQf@Sq zWPUdB`IAuk)@EjsTr)K^VS!!E&69aJQV7@?orU&l`9i5Q=7IK9B z`}gmsOFLBnJ4CZ=!^^|7X#j1+m(Bb$t+T5bjFNEbzlU&wcMU@4Hq7jSWLFP`;?@I0osL}yKUE8W(;>)GI@7bm;XEDD)~jL5nZA+Hkz>c6 zdTIZqt89q9!0zrGlJ=C*4LVEcg_L|($R6n9&=FETI_e}@OjvLjjA-l@HZn1Zc*MR9 zz>+#=`M0l8>eMVf=Q}%d>t{jUv|4G10d43AZazTg51M2w3wpUHkW%>EN^ z76mXo-y?ob`n2hKv-jsqc7^!;)O-%u^GW!w>p zQ-AxA-~O}K&(-&79zDzuHm7Qp?>~H8oqJZ}ZPOn^#!`AtHZa}>tAm18Gj5@;k2~u- zYt6KdaeWGLl9j7;8{$2;Z-{Mt`cW;12lK4H5fd4)*IXnc#aiP7htaC62?N_~u>jpFY)Cm81c$87L_AD{U97yeIY zE&ln-K{_#hkmDRX7O%^1Ja|3yO$t7S*(2gWozF3EaA8_vCZH)&i)Qf+PpjYi`ZN0l~9Bt+x_ zL+=wPHuwt#7G(o*;9+Uh7p_%p^I0IW^zlywee~n`@X;LvX2lZi4Z6?+k0+5%_o5~Ez^{{o2O|wI8F;4=2Og`f ziQ0P-Vcy4Zo&YC|g>(LvO>N(eMsi@!-HF=muVet%^8fGK_3KvUPl7Q!3Rq-MqM$n) z`u>QXp4RHkw6vA){tj%BJ}|=CBOnm@@MMsIl@-B)+t_5L(mEL$86oBz)zy9Dw>{dF zch8=-SFcj!ZL_(ceifnfc0GMM_ewS+nToLza&tIRFv=&S`6yg0SmZt$)CS%(GQ(?? zjQRsI?47c1nYc!{asuw9otfn*gd@3c*2^pCX=&wQs079c6aAQ2;X+~r&5wtAcKktg zf0vFz0(g6f7qxTxk#~Gh#9+n?c3Nt?W7M;eUQ9W@dVuff5*7fx^Hr5t&Nw5mXOBJh z2fzszMKTW3)d&n1^uNA;JRDesAfd|#-S_eH=TboH3rvjSyD?OKDX+Bj^T&@u$){Xg z#<7K2Y?JYC49z1|S5;Ofb=vFf6*gc|w}AK%B7%zAT44dK(Be$peaUt{j}3o%0?>ra04fP_Tn^XFkL^B+I*eBnJSK$$(HJXPZFb~E@u znS9_vk3RUsV%T%8b%}8->tqhrS+QOywd4z`d%OhkU?bqU;X9R;o;VMvXEQJ9`>0Wp^;rK~B6*%JU?j40k89<`Vq@<{WFnYX` znHgI(Fkofbb~-iTBmOA6KH1XkpN?P)PqGfNI|8Qy4;uCC#-GuK&t>@~jaw(CX-U;6 zHkAi)Yh0mx%0#|5h}r!U=d|k;oxNo(75t1R7oOIv)oE@kKC4x(8QPcH7Q!dIWs?D! z#&!CbFTu7xf%4zReSKZFGQ9HCTMzi1dggr(SH|0Qzn(*HrN-(9s+^2&rBN-twQ+HI zX!nE~1UH1CmwbTLqS=6i53brf6#Y~0u-xcg--xE{<4OE6m8%k4f>IUk#tZ0~VDAeX zeaBZN;}CK0;NmK%ui|vs4s^uJ8`A$`8VV90Qiic7*@cC3_}fUpwPj_Q2cKe^bob_+Q$K{22L*+@&{i~&*7IWW+j=RQ*~pD1u}=41o`|D;sSIJ z+Mh`N@HtnfF}=YId`M_0yj}VFe)cjsG~D0-=nUVQoMmm>hwu~kutk1@H^hA5E$+M}+Y3rAG?*Y%bg01C2|FYM*CLBA~gZ|ucrV+&Jr#r15`*6$sK0o|cc7v5Pi)NSqQVZT$z7gu6;{1GA zXJtkrx`I)ahwcoI9!-vUm1CB^U=zITA+m;h^Hr}&5|Z;tb%Og8 z=zFerUBF&)dx$(6K!O`-X*%w~Y3o=Ngwt&lLeM244a2=1*K~4_j5peqewV{{bn@gC z=HI!P{jzJkE8%Est3sCyU~Y>0*2FO_4V_t{8b>;O=`1sDD7WjN$Gw7rkGLi)D=QE! zK}VB+83KO{zGNFjd$E!Hl;Ez54T+9EjNOe#rK%5@n3|^FyeUygId=w66fFKWRWT*~oE28SK2DVWW50Fh$Vhyq z@x}ap-3t-Y`e`+=yGrkt)o9A>{-J;<#hAEW#>u-EI|EHj9Ji7DoaCt^B_TG zMZ!Tvc9DemMbAzAza%7XBrROT??I%Li^T6FJCxlhh~FnjLgb0xNhlb`HW9ze22Ict zzdzda@0VC&8-qR=qc;~ZQVNJ`s9O0Ku^9ps6bG6rfs|)`Doy&$=$8S9DJut=-#=|h z|B{sMP>$hEbP(6BdBCV*XhvBr-?$KF)$j%w3MU~s@{|0c&Q_?-SHFBwUznzMXo8Hl z?|w!`1}2*1I@rvmj27;)m_mT=K#6~RNZ+j#CYjHlUvPK-I8rGPu`?k*6z9GrZF^?o zw0IqC$f2PEn`yT=Nuxp0A+Dy#jcVws}N?41s5LTmKcNj()6tl0W}O!S)DjONm+s`_WJjz)lR_Oz<46(Fn=B zckdklDU>w05EsIVFFQR$ba8pxWC`V#uWuZ?gsR_68(=~>_8jElX?gKNoSXX*e9sL& z;d7DTbLl(plC+PyfBzF??tW-saY1|FZ2)tD;`xvVa*%7h{k>M(*r*T~7>HwIY0wNm z#sXtujMIdPf?q-<(U1DvX&Gkd{_mgPguTm^orls~ZePvFAoLVFr`M;gxRZC5a41Y*+_pF^97EMf&uQKG$2o+uUvuG8+*>7pH)2K`9?UB zsNOi~Y6}a6M^`|5>V%;xkBfqot%AM)wL~k5C=FCp-}gjdFDp9a){K@^iCv z+K&%?-a?WY%s{OXEZG7l8UUq&)dOlk79e=(Wj+qAWZfh3xp63B1P^q~cn}Xb=iWVi z)_cY&H*X3RUxncjkdFPB@4)23CCBsUk%QVnkJ+(HqDPsTo70KQoPl$&hnpK#-tbuUyOI9u2h4oS&kJV;puU5_g1v)-jPHtl*JA65h3lJ%r_=Z4 z`qOcssz4~-ef(H3-dguT6Zz2v6}yN@u_H7`GTErI_wLp8DMX(Rq08tYbo6g(&s*Qr z^`7nFy5#8Ma{qG78c;~TXUU`_jVhFj3r+(;O9h`Ga>O}95fxtA#Um26>rxw+IZp=k zH*#`ms#draN4^nZ!T)CBK z-Mmf;Qn3$k3Z{RsMNjo`JTF(AV)Xmsi#K&OFB0V{XkADb(7gb=pPcrt3ET4ZB~B+> z_2>VzFdG124xAfS4z|iZ7;r3o|0fjfaE#Bi7w@~psaA_Sj25M}SuIxj)>kqT3pF=3 z@)#Nj1u-V}OBx0(uN>DmHXxr&^ztGJ6vQAQ&U@#J7eMF-F!BH$mE#AL>sU%K1Xy%@ z{1?WAaQ9zybYS6P>BEPI4COtYt5MRKp%=5ZZbnIle+E!S0oNS49gToc-2goi0x&Y} zJ{_?}^9!bLOx+g^t2=+AH)%Z_O+s=DHB~sa^vDH8Y-EX|d;^StLH<`%h z=_U zd5pt+V(t0>v~wS9D{pkeSpXWSyoF=;F*SePv8i0=#S~k$S=No}b0d<#=%Gjeie^+` zoSd8t4*)!WLY(K7#AiXRl+aF>8q9TgdL==#oVB{zd!_JkvjLc>-Ga|5QG<9Xcq#=@&BD|)j)h|&z)dZUj-w;1Feqe+qyEpE058rY)?oQIPOWnI&jc5)6^tT|^cImDatOla0X z^W|FN7H)@cqbbZ`wR*DJYlBfSF%OWU@-hOP?d{d|utMh3NM+ty@3u8@F)`MrMPkMl zMz2`aa+_EFP@4)1NnON*7 zx%0!i(1ojBJS30UQPJ4Fg`|H^8rchc6^b+;!9#oa_zvd~n}J}$nH}n%m4ymOQ=|lX zKk8*JAC=G&WN@4_etzuEpioQD!VKkwp9F_tNl8-HX6=@@{R6zn@sd;h>ePrP|L$vy zlXL}s;I@A0?1dSv`%NX*&Bu7|cpcQ=+B#c}*mHfn;BxV^;E4Rf!mbxD@=Pkk$Vkkl z#OsV)Q(-`fs(oHHlLxMo*+oUfgz8YZvKlRy-1VX7IFY9;DW%(h=||FTmvkMI#h;OO zXDDS%A=8iq@%;|*1CmJeUBa2c*L8AwRk{L;&j$Y}C@#L&t6W=56C5EZCT3D#EbFrp zE2rJA=RZZ?+p)~{pYoT3=ijpTe|!~xD`!`bC`f16GmAAW6(!^(6D|n2w_$+=H-&q- z=wsvt`}?oq2R22OT93gQ9UUFCC+KkzGK=*E&Dh9Gje!oJ>cO>hu(z+Phu(4MKA`jJ zi?9umg$y1G@=zc^od2xwSg){%$g`fF++fX?hcM)T$OeRiNBylI>}+g#pqNv1?rk6t z_4a-LMus3?4)koU0=G0T?+g^2sZCTQL2t=hEG=-;!ke+GyBJblY{Q~lY|Q*nhAWNb zpqDNI^~8t4lFh74?qnTA3vmMri=SAxL0s{N-`+?;HH&gWSNAzsALoQRz!O-Cl2~Db zGCf4z6U&1vs-2HJIPh_Cc6WCJ%E8EOF5+0jg9mWp_zFvf1$fwa7Ze{|df5Vm1v@u- zhwnw;SPc|mE857zt^a&V1rsL`C7Tq)MHM$H2TyqGTK9P2c*? za&;{i7lo!^gVr#L0>C)fgMckc1abpp4z%qdK|$@|;53XPNSJHCe*OB?o|qQQEue8` z6SL4h3V;do1Gf<|iEk!)63qc3FvpG^y7!9<3K-W?I_;}2_hoCGYLCZof|RuMy*nTp zuCY2nacj3>MMA2c&>lHqj*_Z;fP2%01Vh?p!A-?!?HIQ`dNx5pTh*m5onQl z^w*cx77ia077|K)HiK(~qDy+wk#;EHFzg%9;%EH%d4?$VwB@T@$C3Cs(Dto|Pl&M| zKYknwOdcbr<8UlvQm{?Ys#o<0mZ_|-JJyah2U_1Z!HjGiFL}d*iR65T>#IcJ6e*lM zhsb`N28~6yC?x4bkwU--dR@K?!r_0W7Z;B`c19CdHIzLgq2X;VK%>`p8D6?@tI!WQ zAT3Q`JkaTWLUgsfWa4K>%XJC|V|se}#6xnt>a8u7dtsDVg+q@3Vsr_IOM2cTv*eCE z+~(?h@Yr!kP(vL$=EjycYmVNm7EhVu5;NULTevU5!$3hnfyhf2*#{|KAjg@5_rbil z<53N*BhE%pcM~p4c6Np*PWb-*>3eAy^)Olu2)6MsOE07+Ch`=pj1_!Fng`jv_w5#v zN1sEHFz^yl_ooBMgT2I_G(EBtret*3^4^Sl9Ih&Xu}#pz(YeX8YUWkO!h<+Y%|GUf zaKw+^Q8u8uy5le7Rwt*+ii&}^_a{=YbT z3!tjP_gi=$y1S83K{}jo z-`x9s*BQryoYBL(_kLnMYpn-(1}u00K{S;+xUWFT1Ar%KFxdt%TR#+#!#=tWMw6cf z>T1nAwhE4pegTd~K>=fV6|_v?e_FJ}@Ydv{<2FWso@8@lBUo$E*~uxGt!vuD=XeWq z+dJhqsD&?{JDA7v3Y%eoZnIH8Gb5uSKR+=u^My0cs2~Z|OFeW~CxHEde#8rabpZJ6 zy`YI-_!1?z1ZV&N%@+h93m~I|t|r$g=?GZz?Sktaq4ES!ymJ14x=q3TW$IE%a%$@I zSF6RQ15gmKPG(fD07r3Q(K0^pB%51WDu$`ACxRGIttD7${ie+&Hw+Ao7Ro&}9g3Id z{u60H7bE8Ddy6lAX$+u^-qme@%K?XHUqB%MTnwRS&uXji_DzBKfdmiC9V}X9e{wEp zg4utov$>CiP6pT?4vL0F!0`s37+4UL^MlrfrIG7pjpSSy+7EDdf^-1H6|+D@QFJEY zHWLZ50;V|+*pNp}MFj*XW}waaXG{ZpaKLa02rGcfng`(UV5mBoFw)z*0YYv*9~}in z3)l$-!1(VW-aK$KnpRgau9v0&YC>M@j~NrP4@=W@P2?y z0`^{|rBST^#Y|cN=qZrUKSyyV%!~2^Q$|oqgMH=dKICL0N`<`4%ocz-3kD|>hoG{Q ztK{S1$=&9Yd&Jgz4@Uc&faC}`0AMRR$9nGpsM`FmhK+XVp8*(Xq0Z~fYG{{lL)h(C z3-I7rZ*C|7i6s|YL5))FnHLMNY=A+1Cs?BNwj?<^J{I;quOOh0BkhaTTm}z48*n)D zFMpY7=00sLJ#0(bb#p)(F-+DVS`=O8k(9?J#PRNXJH}zj;5jq zf*S!eKspf%RUFLBFvk*q1RMzxgA!U#ED|pwVuEmJccHwg`?BJ^t2PwH*Yw5L!!{}q z|AF|bz=iX*oz6?Ik5)5&A0~Jows*JO+LGZzM`#aQB2^C*QHi52JGt3-jS{sYv})ubcAzPsMr!g z(+YCbtd3--#%o|PR=fywDYQMAA~XN?u&bkp0uH*k;|{iKWnSJRfKu1hB?Pt~HFK?D zSEhhc@gG1V^`Ek>9oX`MDF}eBUXml>@_-da;86o`i6-5v??`6?VB!Rx)vsR_K)u(g zVG7!|U9g`AjI>^UN5TbjPI3-?8DG=?RHSAez1e$LI+f3z@dHKNORTMM|J9G`zw%es z*9N($;4Vj|Z%`rb&?A)p73uu1GyQ)9OFc{gg$b0C0G+IA9Kl|vlE(_!XjT3+eGV$} zk19p^`9bAbkwbv(Tw(@#Ro8A_Qdg})O#rUl>R}^89@qfW5BpI7_z~&v!BQb+q&iOY z{a`6Wkp17jeG7hQP}qXP!^*@&x;EDJ&5=_>A1hcMIX*ruPBHTe}Y2azJa|4rUM^KU&JqgDxJVtJ2ti zI`yiIJ<0S^KwgsvOIX1TJ=cj0@-TMK8yB!+ysV(_5_iGJ!dhr^WfKF|D*r5>0LC;$QBVv#BmiJ@ z2c%`SoUOI_ZtBB+cUR`Z+*6+cmrLIc)anG|Aw60K{jo{S&u|L4kUjFP50`gn+VNf+QV^3}DgJ=c>p1}5`0WkOZ|LwE{tQ+t%gG)6d zgB6%WSXy6EZZ}i%7>MDX$)1291Kz6*>sG zYxlKn^>YGqS`Z*XDglks5a?3>ZK1TXQc|w?*BBlEwHTtS;T28 z+(DotJKM9; zrq%olQaPY|tr$)MsGADRxBy>wDmW=Cs|(cCgV?VKJmNDmX$T1&!JQl7)k_*Mp~M4# z9g@-evub~h%YS4)nZ(VIl)-H$EKdL;;2jXt(AWZO(8A(t+e&}rb)Gsiz~q3x;Rz`D z?~aGL?r(M^EW^P*CSC@HEgd5P-$P`NM~Q=0*~-Wef(FI?+` zd*Bu_B3%L+cgFPX28Sk`X!&dTVBqEh5Iytiii%&L21>$VZUiW@{*y#d)r2FVY=?qK z@TkQ(7Z{R*>1h%U@C)b6ScsY7d=JtP$UI5k!C)RRN7=8`x|!zymJ{Z*4jBKzVqgUk zAu+M^X1rx*G|lZjh&79|v$~hpEMg4+taWh62{LzeJu5f-23&1{oAlNeFkCfX*CE+3 zay7(-AaXD@n+B6u2Vglue}kL)J{`C~umj*0XxG6}fQC^R5GCvZOy3MNhDw;d*1_ zpi~|DInx5LZ!mIQ8W~{++=G8Y&avxd2;zI7iE2VJ%hCeq0$t~zJ(f=|v2X&km|pbe zjbd$9|I2lBHMMadCf3%`!K7wU1eTJr-m=&P_*7b;B+d7@06bdY60SpGG}8h`uvmIn zvEkvDpyvVCkA)pfEGO+isfSQ7mA%Z)<~54|90OcFeWTX)_J2n`H3i_mBs-RLfLfb#$bRb_NYG|3tB;$-H(xydI%F7QYT z3Z}|!fHpHDGn0|x!>@_^4hK}VePsoO@FTH1K!^g13^BUnl-M{I!WIBzRd(y0dX$d^ zCSna}$}e6N^3MTlOc1!hC8tct3s9m#m`@or1yDLLr{C&idZ#z0VV=g2$FQH7T+#dWdF}oyNv)c z{!kAF_nhYDJHWtUf}yXC%mU-13K+Dy4Yv_kU^K)UHEaL^fD6W#oOmk+qS0%zX)&?7Z414X_1(?1a@e>pK3_dA1PNc48$3ACj{^@ zh877N{<5JBuPyuy6zsx+&VO9j6Y`n*P~V^OCQ1nK%7+W%yc2F8nL}pt&bsnC`;H1! z=Xk8oe_V3r@v_AwS%nK}PM0!Q(mn+>f-_5}Cg*^tjGlWeW%f2!sK!WTm-(zS^ySd3 zf2W}jj>wYW+Yw^y%3(N)a^EKGU4=Tqii#pFE$u5^!2Vb{<(~Y zYGBmO8=pAxt)Euly@PA5;qbl@nevtu^)vm|S<_TXMyEZ@{IxG$yc#EdSx@_fO5|!cZsL5JMpet}MpTZaEKu z*rYDwjIGt^Y}bI)FC=-Wfs3!1!&fRCVt5~P`NZ}yTOc@iLbis8c=Z-IkN@y#>`(~w z>C+h;Y4u|u@j1KVhf69Rag6OY#o!055cpzsz^2*T*BCE|^V|_QjZpSm)~Uyn*)LQy zB1oF2(YF*L%RgBrx}<@{KdA{810sFo-{De$!HP&VDw7_K>t&Iw==EK~77Wp&rG&rU zVKMA;*WrAd)fOx$-OKCHyc^-(sQSU7M`h5k68q$3Sxs{^KFYO^zOK}f*^hnUmX4@_ zNsSn_z=l`pBc3ux-$`1Hq#Hl7BsXUjPW26QbWHfkbJNLP7KS9^{v`t{LNj~EbLA)I`>eD7crvrZ z5?F+QpVP#MS=Nd%S_V=XOk0>dvWOR-H2j5PD2igwN`8!<tL#g)$S(z*xon1h3>oMWx7AQ)p*QDGvq2&gj-RwoUgonZq)m<{gt;i!BipM zJ50Da-5b@;P%V=3=*8&9LDb<>UZV)>{y+7?SnvmT9dD5L3U*F+9H*zK`z&`qiio=w zbT=!ebMhBB=7R#GtniS3h#TTRKXh|;gspDQAo zO0WL5yyJnqF(iwQyOI`eY>gOGUNd@ghTB+uvB`1K9uwL`e7f-c{(?;@glzH**5hO# zrg%N}kbMI;`N`gDdBPu@r=%}u;QujSw zesuVaW3$ShleM7(CI6`juN}$X6c%*XpPsyTZci!Q3*XqHRPN*&IYwHIOoUnCQIOt_ z)!b}NtQ_0PL`5w6Iqe4vjrU6#sr`(6R@$8grA0BBmi4n-(K6Y^N~Vgcpi##-_`sqT3Z8P? z+myflNfx2hwa!v;M&f_aUuEFJYvcX3QNkVlO}&IL_F9Xex^wiRJRAPcouXHEX>ao9 znygdA1iYl%n`-2AYsI*|bUB#6ueDTcr8O6a(6;&>g=*vwAy6t6f@gj)T1AWvZs%ZlEkeUEeC$T|0bm zP5R?ox4JSqM!!4X{Aaf6s~#25|cXSy3_YhcT-%1pWI>0kWdDg{$p5ySJ7$ z&Rj14^^()U&isSAsyr?h6&834uA+jBCIlgYmoR#aeEYfsN-GANvzQUN#{iL23>u8Xh@%E=fiKMHVqlK2aGF zMIH_*J>lnyd~&aZRNjhdnoDs|$?-f;5O^Xd%BLa5q9MU8D=#Xk%PFPHtNWb&l^nm4 zsep!!u&%TCyB9*{nqrEI3R242N}8{9RAdY^mB188Qq5RK&r0!)la7JOD+?7Pd(GG0 zQbrD%FW$PndS|2c-qX<0-%?lRoq?9)Yk3DFb<219@7?s=->G<+YXe7mGaF}n2YYKr zcNg3DP7b#2?(Q#b1K&7>nYu@q`^3BX1{uL+&B7IJ;f6Nxs&+~0E+2L61C3q6Onnoq zeG2t$15>>NKe&dadqm~Npzc)OwJKz6f;oI55!12=D;nL3G z+Su;K{K?wr+3x(`;}w8OZ6BQMom`*({Bid47`QkAI9#Orkt+nr?<&eX*Y;RA@Wu5Z z`Eqstgd8uJDA@SWN{C(EP$uVPuK`LG;`cZ@v!>G6=L)HW{uwzE6eKI}<3C%BNj{CoN`|af3s(eU~sW`W@ z;s(0x3%9Nl-z-mG=8{2Ij-+!8KFA^uCP|D7rCUG~GRW{JR1n`Jl=A@%dhvhrEqM48 z0{d*?X_p_2J~pSM_CK#2qLgwLE^`O^N8cc4QLHhX*Qjm-y zAo>tG9YNUck#H{42C>xnwyJ9Ei{9U4Ozi!RYH#K9vsqnztc;Kla#rEtl1EW(e`_W| znM9m|X~Co=;bIygjv9GKuuyG~^Ib!+{|rlCmsOr=!`8bWikMo}kjBe+O3B5H?_ah@ zKY#Q-S|yL@hN&t|9TuB*s>l+VY%$UN47c>*t-ICWiU{RNI*HAD(>070jiSV%uEB;l z8@}IPx#Ji4x~P(-^H0%y@-OY^Wd4RRu?OK|pDHR68`2LvePNv({!uWtckud9?jp(M zY3uWOzWrDCJ?%-#6}-F5tY#m>e%+4neUMS^_N2LLJ~<;vwy0x)6!qxl(-*`s;g&ND zHt3u>?mL=(&&S>HzCF2w;p#QhF=&`Giws)S%NE-(z3xLFQ%_1Tsf+P~+f?9*YtX7z zb{70vEqm<}=_T}0bTXghj0Mv#GEcSSf$Vv9PKkY7G)q9c7Y_#XdBM(&=K1;eHQVRz z{4YAG+F16-ciqhw?}Lr@X6}O%XscQK*F;_`-ZA>b=Ha2<8q^`Lzql)ZV)!cvl2rSP zB!Dvb^dY1G*zNmdm-Up}ZC3OdMI|B*0YXLj{mjsj9_>W^>qv?RrmW*zQODYV=Y*G^ zklC_^qow*lm9iuZLJ4EBYB;VTE>oWsEMkczSe{~1_g&$73o#2J1~r>=R!@YjEO37$ zIIlx3-~DxP;4icT??lej%cx6lS}B#@4<)kHb$?KjVbr~&3~-}{iU=Y4D7=U6qLQtf z$rvhIydC*AO3*3|xs1=GEL<^PMS=%6JmyBov91TwNrjO-dZ-e?3u6jaHQJZBv2lUEYB3YT5KkCeoFVo6 z4_R_oO(nRHw+1T(H2w;&EIzdv9YMLRiqUYtVGdbFB&>fXRbRW8#VbnK8a$}@=?n=K z-%W?z-fW{o4}R>JO9-foSKyjnP8mZFaF4=z8P;*QM`H0$E69i-2A>~-l8#w0!K31K z0ZFCjkh(VFt<}b76lk~ApNtw3V3xPbx5Frbr@)d3THY`oFQ}7BI^a{0-ulV4qP&H_ znaO#=sKB2#Mx2s#3$;8@D8PpqH(K5MD`=ogLk#IXPPRTxh9WjJM@|M5_6In(2)OQRY^2rXW4?u#$2|i)Oqa_yzM$P<% ztyLSW0rEA;O;H&Wo}qK@ILz3|VU}If8m5{$JXnokq27&KQyvYQlBR?R14ZcR!(l_1 z$n>^#J^nMBc$(>~kou|pi~>y=JavK*f1|f}BFRL2jP@2z0n1ss;_Lf9;DAJT38Qa6 zcqMH0HRK!zK6a_vNm7ACG(XDvxf%}*L2>NVlpU_N#(&a zhmo-7xwZJOsxjP!Nc~T{9n3gq`uy8u8{1+M}#cze4R$a#GcSZT*mo z=RSvaxR3RcHCF|sv&P90cITzcoSlK1E-Jf`dg zBTE@8WOQW8F1M!Gy;o{cQ7AI#n~jeKVbM}pp{U}}Yw*y;{4EbCBlS_i)QoH8Oq&R# zGtY()+iMx|q`nUz1et#oS~>>O*{X@sG7kiu($DO;TJmr;*`s5`VJOn+XGFxdlD0kRey>m} z&)lJPgZup90mS^6sj*D$SKO4Fu&#Ex8#nI{P2^gvI6*ubN!kMqf7dcxNPD9g3*tuc zO)TYkrvfQlzP$Z3UdHi#>kboA7vLm%Tmlu+7z+(r?Y^KZp2$LiUnHT>>Mp{^an*6; zB#`G^0~TJaLQvbuWRH$ipiM#C)3_X^=kH#;)Z#v*qDUQGdInb#rCcCc3d~c>aZhvA z?j2O{g)whxvHz@BuPBVbbr`70{n=W_x4qX@oF7WNwi}PER-}cAZiv9)9|02cJ22qR%k}iM0roko(zYV7}=Jqo!QRM|92FO7;XXcibeZ5|i zpWDo_lkN%T#WmfZ7}(Yv%~0+$MPg*J;*W|$@v)ERpqG{fEFZ)Rq^vBFF%`Z(`5UHK zSCn`(0*MOkStvCAksX1Q##jUplL*@|9)v=srdpyrgILZC5&~ZZp~RR zA!zf6quCl`LE&!PJ)N8Dk!ub&6uKcb6{}zst*BI7SfPtdd}0)3cjco0&mBS^QLEUFU(WoA#GQlz&4*G8hxvG&Lr}xWkVZwP=UUNV6y)pb24N6G)sAtMyjpOhpvIvQfG^Z4Lfog#R{)mQwuf%Y`S42-} z-Sdeh%^e6F+9y(pW!}@}@I7yrq`L}A$J%`vGtuw`l^5vgQ5LXr^W`vNbm>}JkXwEp z#ZJ(Hq<0_J8jd^}%W>H9*hQ=1RmIbr&Gx@=2pSoije7jP-!hMqSZ$geaVvq7BbGdm zS01!Y87cW2iZiF|*?+g|wW4Vv0qu58DYLudz@t1Vv44z4UYZv%xClM(DW|3;g)UAR zA5ZjtH*2w_Tyin|3UNLOiW7M}DQdrsFT|D+^-bfmF@&K?Hqr2@*q0A^?=A-kPYW}k z!k#_fwqR%ajoec@X^#(+8t8`bZ5!g?QFip)%-fcLyO$rrd&esB66U>vkyFNWcPGLk z)MTw8$Q5UF<~&4xt4aUC8Q=2A%jGDB2^>#T9Cl)gtW94VN~14As4%2Z)Q-NBjv(j7 z3}&frbr=ky=bg~PWSR(#QR+Qu|5~%J$RR9RzQIRJmZ-`>9nkB(9{!;*E-G0ZBm1$* zr-3!cCR6z7T4EX2*8=7@#ysIG_fcDM@*#IGqBTAG>dp>E#mFB&%ec)Yn?Qp^U5T=C z(IUHGGeJ{^R&mTU`6XU!TCh0gjc4~KEwSV4uc!w~ObJajO~j@=X&UkPHW)>=2xp{X z-8uPX7;qdE1#1j#9#~#BR5u}q-xgfu5|c_&z7YJRLcFfZI=}5ysLAxT%zfw2$il@D z%8wLdXzc?jzE=L5s?fJH*^{##9g)bfc_qbzuSIz=sgWVZ0KSM>w`oBum|%%Dxv{Fm zdnKi*^^c7QF!)aW2yVqi6DG1T-ba)RTG?AO@{gjC6q`ADNHq}?TvcJd$*`)(Q2GL- zbcn#$<6D#|Q%ccOlH@myf)FH&`alb+35SevK47S?XNS8FltXjM&`~lHWP`HsTVSuv zzLWYzTN7E1X*RmKHSfpP$}))3oOx*u4bBmK;%Gwg7S8c4K$Hx^u*hfn>-gw5R6rZsx`=JUPmvGexHiI9>1zDXl|SlFua@3S}Wra4*H zTYL**ggNux-9{N?so<`=7p&>Sk>JW-_nqlxaofUoMrQ?bPg1_=ZBjNKlUQZc+hJw> zjMPDB^pRwRYTvVBh*cnY7$Sb-b)m3vVMrCIHPUO(mrvju)JT*iCFr3aw~OmyBe2~y zwWUaH`jcD3;*cZ`sRL4H`qqQk97R!~ng{YKGG||!T|6u!pL*F4vda#>ex&6(z|V33 z)$ta(T{T*AJ$^Q?8i?#EOz>4c+*F=UKtgmH*|WQQ`F%C2WEgACs}d_~;_X+o-#wl) z>f!r7K+j3tc4)^pfSEogcq^U!Zqidjh8MFb<_od2naD89iD_<-$zar-@Pjx>JMOGp5m5olMs7WLDCB3wPBLm)-eiRU8aMa_P{eN^N-2PQkK0Y}d zmVfq7>wDXJ!7;epTy1>cb<6nIDe}ad3Hf#Ea!-eTibhvm#}}1p-Gz}qWRb~uah?2d z6f${&5z*KhV?eVf>!Q+viv`25-jj|jWhVaH44d>xFWAX@wqoffV`*9WIE8(1csRS$ z)V(J`9fFufFWg8c>Ot_E}|>;wFv2iIU64jH(Y&Nm$C`u%tT zYWVYaxvl#V6{Dh~HGIlXJKmuLb4xlZYWz98xmo^#dewe{6)_<<*|ota`*$K>S=N*B zf^>&;Czu=I&vrQT)FU1fF55r$5bjTIIX2t0YOV3n#v!meUA+BBU4v&K-ghK_S+iMf zrHRf!WQe^hBgKZ||BJX0ZX`O3M)YiIzWavM=Wo1f)zNrmtW41Buv~lc>vt{pw)Tdo z$VY+cnL5GHFD&dMsTx8GgrW|{>j@BLkX=|E`A}R>=JH~+5o$>lf4@H-V(Q(zXS+E& zAFJDqPnm#M-hUwYc~Yq<0^Ky9wBMHv=Xi6GN z42`c6{~dL!$EA12fcx9~D_wtY>K#~F)o)pDzWp`5cppJ~t+kkK6MEe3yYZ zH?}D|n&Mg1kwz@s=g~Cb3uk{_eDz{$}*&kLJ{&&fCFzEzL_Lp6PkPru*ArrtIUMSgb-e z4FGk9bZqHn@1FR7#ZF>ruXJLIp}$UgxgVLay>K!_IyijEW`#@OkuEm+Vyp1RtK#k! zfqj~UbvZbjE&l4#Yq+V>y!-OcY<*X6rT?wj`m2_M$&2hIR^Q;JyEm6ctw+DR{ZQ`S z|1>#DjYYM;Nw>-26JI$0I`2>3<=1pn(Qy!ep|NX^Wpt0EHQY3?CV@@oc^p}(*42HW z*Og7u#Ul3R@_IU?J8&N#2_e7RVcXs7e*8(wWsj?i)#r4S&i<^Yw^HpPBF7-e-o7)> zfBDL5k@veL$wiibAJ^})ZVesi_RKz1@h_Z4tO%(og20@?Yp8n;Yr8i)Z@gMvJWIAv z{jMEf;%*$f+oOA;`JlA2%eK0^TW9%_X+Ey0Q&k&bf`Mz_5#Z+o$cUSj82tWWtk8-a zy;u8Dsi0q-&NZ$kY*%wvUwm6$N&wM~?SJO}g#eK%{&%$*`;*4Ud$tvCs`&BQgQHZ5 z&PR_Y9yYSMuWJNC^sVqk#@6%>)lwFVN$Csw{P!np-iChL7r*cBu68BEv7n!a^wzbe z{+7Z0TQ|S5v|tR`eplh);>;%+p?OPJKHhKFO|~%e2JX*w|CIOEhZxz28C+(k^a|cm zD}B8$g)UK9r>ncKzRk)Tu7;LAhT#5r6eTtH2WNuxz1T?Pocl)15=6Q%H3|Kf;mJtP zf~D^xXU2Z2ucK!K2G+I|Hu*Nvcf3a7@75N2&2I0^U7@vhl?EMM;-4>dn~!f^7-2$z zD#Hn{{yIk}T^w++aNt%+|Ck0xtPR*Ig_2B=-u^ttp?5(*QJauabN}!$ z&UXrG$nehp;=WsK)&K5JZv$O&_Rtt2866sbqZ#)}WsnK-tqrTeIH=db@&+*QN7}U{1KO~^N;%# zuJ$vJO`@9}YNMP!3Hh+xn;+BnJ*!`r?r)+6YhrZ##IHXEH&4u~<&#KwUB7s6@oU=L zjOge}i%97zpXu#)Zb@%Uek3zOiIRzWVF)2%=yyT)`)+gD|8*g*y8df*&4g5=B!8pl zibBW}d`2rpym&R;0A5_qy%Wmz2D#3rF zg+^e7@i2&Dr7o?xhOZk(_{XKSS-cUnOpS!~4EM^(WYFsU$n(BuSx9;RaB&oxGTD zAMQ`j9#jT=mwG(>gPZDNl|WkZDO|i-YaL06eVk(h#8+W9d$!nQ!yN~TU{)8R!Q2s`y?uOQ1&99PCb z{xv*0KKe1UU4+TsT?8$B4ibR>S#iD8If2x>R~olQAK3c$toyRB1< zz+s)o+J9(iG2Adb(SHZvb`T4BxHJAPdG3=_3hx>NZzmrC%+5hea}?pWt~m>wiN6c5 z^WS6H$)7$MBrW0r>~cSip!JpxJ(oIRmSD!6p@OldprE9nL`WN*fEr$2ko#pB5=HFE zA<0w^?4g-LscD=EXg?058FDYu21RR? zoH4vAp+yHiTnCAPlBD3rW#{@q?q{v?JA&VctzXDwN;C3zBy zyhofhZqlD%VR^AvR5fG=Z%f0R_zi-HNcA5&NbZUUk`rsA(8*(7C3)0P;=k4KkVRnQ z&aOM`A_U48F?TYw^;>*IBbl7P^T!Y8ZteeZpnx$MeW+TvFAEUEg`SZ?1zi}vpXEQN z*#F{>=IcV+hDBDM*uztaIeXnp8f#ST-%fR5*VUmdy%n3`wW57`CXy%mi$TR7qMfeJ zwHGa6n98ev>B=SLAv|eC(@JJXX;WCtx^KEez_yH~)t%_NgP2w6_;J~QZf4RtXj~Bz zEy!simzy&<-a8_D^bG73Zh>V{Su`37RU7peaO}I|)D-EJ!u$AN_1$A5A)ZWg9Esf` z6TFXk{%jRR%@@~DiD0`lf40Zc`k`p-_ghslly$PC;^T1Q;~G8uSmL+*Q*`axFZ}G{ zm}4cOIT704Wb5@KbM>)u5U1-7`917%JZN+Gourn9qAUwOT58c`S4}s*G1x;m}bv1ZU{0uY)m;iWPC$;Gq1O1aG4#1+9*2q zjPLC*qlD=T%(W53GN>x;=sbuF^4Mjhas0kWajoq2iato;l!CL$kS!mnYzrL?(GcRH zxE#CvHu8+=p-T2{$!aN*@tIf}6@iQZ0pzj6FsAkiwB1jo>01$W{RkGOHM3>&0yzEv z)Pf6Ji-`}LDCA0z?1esaLz?*KE@-V8wAt!~qWl8P4dp3fhnA0POh_f`IB8%AVazNk z0r(+tlE#NeoS%zpv0gU9stn~L9zNXE?n5fC)kg~_G+Y*uyHN_0JgPy0DCuNz=wm&S zp`0%cqB5v!)z0Ctt+zgP!Q){^$out{KPNtZj(tq(;Dw69x*OcV7{*kX^*1wemW42h1`*<4P!kcS4SI4#2p4`6-y0dE zJkB`NMipvnAtq3e3>kkpCZL9v+z1Sa<_^?}>iE`*K4kqsO5 zdfGBg{`NfJ2wys5ZGDTjni**#3@USUUsT`gAeUXS;iZA5YwMexhmDodSXaXcQp zf8IwGc`R0Z89#5vYnIQPJpCe4WAx}y-(!V+NJ2W`kx;0 z%NF}yz8ZK}8hSWZc%%QcEqTGKf@`hP;OqF7o=9}7*}Bes@*!#Qmkk$3Cjsg9%BV%O z;WpQ0_lG@`mG?!(Z9%J=>;Z#Zu0+2xb*{yC6v_;8&2uZ3y-viw(Nw7Fl>drqpl*3G zH%yr$|HXXfV?h?IdX{F1?Tg9TDK=@&AgP{cyWrKzvqxV)t^awa_u>z{yW0!zz;Ilu zQyaBF0wq@Cy|dg@a8cFYPs@K8U8UQ^*l5Zj_s)V5&NwKk_Nfr$TP8?dT1%V#(>4AH zK{GarzEWKH1k;cy-y=ujpWC!;&?x*v=MXr*0Ma_ zd%lEa?o7WZAKhT8^A<7K_Q_sGKX~Md}Vd!7}(AoAp z(>PW@4@qF&|yU~pvHT-9Axnhz1+dJ0q1+Fe}vkuEoHv12CpheMxg zW-oP|8)?V|EuJXWpVDizQ41zW^GKqU%PX4sBF1lYui|s32n9r^iSFq~KIE*P>4*R5 zkCK~wZ!rqg^2}cd3E|H3S~Y`?f%Z+y25brMsM6KpJB4|W*g8@YYD(vUVT;*DQHWAC zUdByag~$woTH!Ilo{Ml1WV}7a<%<8u=h>O!4HLKfhg_>3sOqmA_GGHl%zwp!R$;ubxq9<{DWlX$me$%Dx2{PuC3-BjtAQC zh3{_DA4Y-(>;mM8$yY7=ZyhP19cJZODmIQ@XG-bV&sx{~OL&Et_nLvd&>nuAT?an# z^M#OPi@%wB{lpHbezl(BXQiUsc%jsdWOj-w2IX#hV=tM0@{VZzrhk8Lw)DH27qNxT zsWDtv@ycadX_Km)zdX?8S);MuR&7GY=x-b+$|27B5P?q|^JFV>1|!K4$jp5ZgR|nf zj%}0UE8V9rs`dN-W)9}keTk%Sqm`@gF#EV~Eql# z$+Ar|TpAp?ckxbvuaO6V$!k1N?0xq6xw{0$PJ(!L{)b1kumo{2F^u`dX#&WuGsy)u zhRF0UMemkkg~nQ$!eK|9nYo;C;${r{yM$@i1i{%R;e9qJ*H=p;sgn&M2HfN6U*;&w zN&iBXy%feTyP|*ZIXxMESM)K8T>^EjH~R~l;g>R$^tbgE!iEy;^@3!C z$ncArD&BFFx9I2&r|T^7Zfk%0W~`{;r+S!>k+;;5k?^Wr*(t(&hkSfwIa3<#?#AA} z4Aq7b88H;Xt#i*xiwlR*<7!btD^0Hh+f%$=kZ+lMsB|V*ToFlB4V{AqX+1m~etIqo z%CsB@{J^CBbA+2`h1|h^6!==)iA4ZOqe?t}2}O-b;9*1fipEc*8E2Qc;?i0hNJ{>? zSNEz_@#wV_hfUTsYlbKq;zj=EWn`35V!dv9bMo|0%Jgjqf%pggew*Qhy8}>r=nu35FUx-HujG=(yO3}{Ga`iZx=n8t^LMA9|r-L18$*D8g}dGgW4L%?-{}Gh;;-3Hzf2P`*ZRr#JSTD>9%5sqcgVnx)+rRi zY1Es1{)%SbvPnr+F^7!c(6LURhxR~W_{wwdO zrx$2t1F+~n=+R(9tSXpE@tY3_+Fr(#n=TUw3&m{5K+HO~F`GiPr)gL%R4o4MWrwV6 zUc^U5iH_%}rj^Ctr{_iLET3F9dUm#j_;IA1iPQ;J>*eR{exme37e*e8v0+T1q??Pv z=bpbq5lA%onVd8`O=)3`-bEI~C5QBFI+^)rJ%_UJ8OJp;Gyj6X1r_N8oeU9l6zDcZ z(lCm>{&1gG{CBr$IRlW#QY$n5rMz z7XLbUZ5D!Tr1_)uF{wQ3{zI2~5u2~SRvXWU+JZjA=o`HQRg1eHD?1^rNzn;j3o+Kw zp+W}(u=;?*a=5P@`oGW^DNqReQ z^gP*E-ecw2o3gq!F}$&dLPD_W#!Uj&45V?{!{pROj!shWDE`UuGR_}xgpjfvb=sHw zNJ5Z|$1mtsPcFy(CorJ`Zo*f^T%r~2+LFdj%q}8|r7%^di2E`6Sg=QcX67r#M&0l| z{wj62cd%cR;iH^C5#kW_1dH(kngeI8jwgYY&EU3;<`AJ0f4Hf_@R0~IZJ^Qq%#W3I zp{*qp!YrjOBDA=L)`%)dg92d^hR4Juyzg0B6kxkSm01!<5CRkZHc25_qo-YEzC?zQ zS{cO<9j_y#T-&Bwxj)AD|9gpVgok%y@{-5V-}$=v>DQl zlxtNou!FXMDb3YJ0= z(AICt-+c5{;$_)_Vuf91B>d}0hc&nK&u_RLlh+%6Xb_LEDBb^hw$ESq_|*kf*1^ua zh~u5;XCcSpvCFkq8Rbw*2jase{^AE=sZ&qbDX!@4a6}6Q%?lbF(GGt;-*QIoP$~T! zq;)%^=7m|E1?zBmV~PeB-88jgeZxGRfm7!F%t2388%>B__AQ)B?b$en>Pn9SC+#!CLMDH+UQx9Az9sL6hd0Zg90HQIW~4Ew2M9md z!=FxkEJjb%UGGPBn3fW~MqenGqJ;bueBC}?V>951sNr%~5q|a0sIOv+lEjoTfw%}d zb4Y2>HHme%R+`?Yx;5IbMfXFzB#f8}>L`7*1f3NW^JpGd$?isX18mhUB}i1=G&=A2w8@HVAeKb17?B!{{RjDHIoY5Y5ud=p#eJ z9BuZBhIVmW>$j`F1W1qgC$?;ldruut->3V*{H&1%in#*K?bL@JX~I0BF@rMR_7de#qUBF85=7XH-?W=D$(MN$;ObtyhyRWgDjyZ3QVZol>;-J3 z&L|tM{VBr`D-?OY=7h@d{c42Dd?%EDwYBhz8&y6J8Wj9yfE4GywRRprO|9L+-XR1+ z@4bbh^xl;cdPjP%(tAP&K|?PJN|TN>=~4ugDxoV~P&!Bv0qICPD01UB<2~nmbMF2A zJM*v1p4odQYwfqLove55=LI@EPewX^+z|Ck%)|y4Hh(ulO$t?OPhY}TMWLpP^m?EJ zdmK2R9{%bDv09GXCr#Iwdz@YhRo@dD9BJu5KrQC)3yegHEIyp)XpK7c`DE|95*_^K zbTNu-Yt%$r!O_W4ptIN=M!I3rYMiRVG7taER~)1lzxWmL#IA!I*DHkvws%0Tb{~wZ zXrQkPlAwz&ojf}p^xF?n$q0JSt8DbRxGq`}I|e8$tl;W`VOB^vUvtERG~4e$Du7_l zB5H}`&^>ZvUb##O<^sA|Y|%k*Jr)SCyviLAdsja2Bs%$~2{3WNU;4@NB-o$C9;m^R zQ0D=oKT<@TEMg_QP)Td5Ce0?dJE#uhUp!_Ll_%0TkDy7thM9a=^N@L(UOU zwY=T+=9T@2wx{~B6q(wwcn{jlJ3RZM0uOXWb3Xi>8>$SAI%=2VHC@wP-+VIpw7ekt z_UW*EAL%<8Xn6JHqjkC1{m#~xyxmJNM7rlJ-g4fO>Z@J(WJdOKxm&YVkGxl^J7!c+Y7ItPA zn#FnIfCap~{zQa}KDdpIu9<7?|Bl(5L`$=W7B2EFxstfUPKsoQM$Gu(MD||g=RCr} zMmS1r7@wH<#LgBK-rXmqRpq(E7c2SvXW~d6f_-Ai`^vM6j0;+Uf3P-GO)r(3zqyz< z++XSPnjvp#nN^Dc2u@C`m&e_Wvx*!8ATh7NGS~9*mwQ*^aj(Ic>Fzwrmt}sk=QEZQ zat|R(x&TCGSPq!GHVzXOeFQtIi}Hc>eu~QMz%3^F&FENds1Q^yL&9eSx&BxrsMn$e zP+z$SS6)3VY{Lf%&asxiNU!B< z#&7Z%!5@4j!rr3~_^}6gGc#2yKf^zscV zK^kn0yzg-(4OUd~Uh+>%BN=d7+AQD?3_zyS7TKM_j>?gF~~V7xNm}Zq5+@FJoR=v zEXetN&65qKat9y#JoGyE7<$xVXc0Gyy$1eGX7CX6>f`ON!?R^yf7W6ZV?_l8-VEE+ z&(d{}y7jn6{-+wp_)Oh-KTvMCa`MXG4X(J> zB3>rp<6`ZN7x=W0w2vo>rFEXMFF5yFut`OzLlE5yF1I$H>GC_93d243#;b#*2kH^P zoEF(7`EJ<5LV)^UrXB_W=hQ;txM@R8Io^WSb-xO-{ zx0b*@6$JRqQ!`&XYA7g}-A1|K_O?pDs(q6xjc%zCh|L5%DBNKln~U1b=;AhD5VLl1 z04}r(Lh36hAdynw`rsdIZigV7hHP!2vo*mL1$$<^agSq6w%$oNpiJ%vwrYK-g1Jxu z(qDPt8O7=-m3PsK@w5}o=<~0H+M!Z^TzyYGqssV99XEr;cxoj++~}`kt`pK})cq~_ zlsNXFBf-#;uk8+~Q0l?;e%^QI4MA?V8Dcaj7I}+yVBkHeiSMI0ze#PvLW{Qn5IUOzoPn+ zI&rbK3S)44xO)~B$~x2#zc}LE?A{Is^mS>LYVp#CQ@7ae=EJRE=pD{{6z_hH`v;be zyc6yM3Imca4B7)}gOAwL2ysLxT>GO)3+R$kmEHI=R~S3xoD^jjpKD9E?Ma=Kah5OD zv}7-5JBpQZ^giVnkeUnzd7BfjQfNGTyR9@eU>!YGpyF1j<*nUn{`g>9>PN3xFhg78 ziSSYe1Z@mv2yue@9=2#o9zbO-&|FF)w*+HXA)T?Hy9maRPy|JX*otxR6jDUwdX}`Q zTArJPMey#kCtQ-B_1iL67Fn5t`HHy1nfl(yn+lMxuClWI5K?G{9TwY{$D!MzAG4~< z#?z!`Z1Np{C(@QxehOXE5>P86C}8Hm{M=9s*?)nUGME)fy)3ooPcM@NeD1|w6v{sE zqmU{8PME*lnZoCCU;EjjHGr6MR>R3$CB+^mdpLmxb+E$KbF)KqmmM}Eaz)pFJ~dfp zbqo$>&N7X7(+e>rtKmxKw_i zlL!n4oy}lJP~a2hNs5F13cjVxXET_@x#C3vLUrm7>lRs@y967mT;g=(H0PN}w6#;u zX991(!0xylQV)#eZt;l@XPE9zhMl6RGyIj1z?ZAdz< z>A_i&7JL+C8aF?k^IB^m7kDZzM>V{Jd0#uvn?95=V|};{kM7&1Qy4vfhV63DUlCJE z^2UfWAkQq z6Hvd78$~G1TYz1ELifxhmh!uEnVbpO@ES8mwG3!BkToQ5ODWaOd(i&{>vJ?v|Nhzx348j_jq|ogi>woCj!-Jhw1xbFfeR zq&DGj3)bXOo2gdF*+nr2T;e?eV05?RTcCnFYWABG74-$STo7S2xQ}e4wREO1?^Og~ zSB{0sS4AGPW+m&Ljfv2X0JpF!Gv4A%uv@~<@d5>J2c@t8ykx-s^s3wDF^XpObGlN! zB^|8HJ>a|oHZ$XgL9qLN!-@?2Kq@XIxw-RVsR~PvI_qRurHc%T*?f4oP^Rau$hK|d z1GqAu1C@}^*Juy}!el+~@cyvfq-nG$T%=ZCy9G%nLXk?yp45zR>6sxK$gFor)H00A zF-Y#UV?bFr(2JKcd-q&Sg-_-=GiF#F;KF+Whxwvb?z$q;6sMG1`bg0aERg!u)X``N zB5zTL2S)mi9WrPd7}rJna(Q_%KyTY;z=ZBm$C`My5#v1X(o+vK#n38$+nxDIbX-Tb zPU_3@uF!;?bURu#$s>!fa1dtIk@b7}?{P=H&g=^|4H z%zyODsG}Q8%Uv7OY`AT<+!TEzI~@d%+gkd_VuyI&!$4Kvx(a-G+5t!iG>9Dam6h}j zFtaTXC=I4bOm*{#2A!n{>^<9?iBe41d7ULM7I;B(&ojBMHt~=;T78v~LTxe<9cVXk zc)DcddC3lj>PH7DLrX)7l<;IbvI*uJzB4^JRABw6KqhrFnE|gxkBJVjo}ZlUQp0XY zd6_<_#a(F=)zhPNW=>#!K3bfv)s_2Bq6KjVpEi9@&CAlq#yCD@9QJh0AQm3wu)CVs zXyIDmQXL%OL(-H?i=JsFV!kqmsC%Nl?o8yCgv@^l!lnFiGkfOj)uai|#uGXh&hhtZ zFU{}e&J^*C0LFa;bgB_C!lnor(@}G%#Y?H~p2lf+))nh5D8+s8?xC!ZHPiGBC6dCE zS&!~qOfs^yWVn!CIxb^6{fB`|x`#lPA8(jo1GQsv4DUhAVW+~~hd&EHNUhPe__i*H zd$BHxjInzev~7;-4=eP>6CtKrZicLzE!F{xy|&R))S>7yjnW$z28z6SNK%{0Pq3l1 znZ#vo4-LZ=r?15|?Z6S01a}lp>@yStO%G?fKeG2=vDBqb9fF`Hur3RIM0gR_GcLzl z%5^M7kdUhpswVLOJHu3B#TEa$iV$YBE~_Px8_?I2L-#Pl(rJUws=42{&*0;ieEe#l z>oxbO$H&m+9`i=2#3i{y15OmLhm#&efkV|Ho>scl*s71IQ48v_*CP83 z;XZb}HjN)mm4L%(*bzy~yWpwdN%pK(D6xEP&9*vl$Q?~`hr>F~CcS8ixko3y4~!O& z2LyWP#?jIzb{Au46X@gnNj1iDJ!fHFAiS=IIrJ7?L3XfZ`F?LC%1 zRgi+BhGD%tKSf0Hg%=>V;ygJd%BPdLs&ZCOy-cVB_F5A?n0L6Zw3BLwc27^Ni-Jz3BegYFGHo) zNTBD-#WsLqm{G$3wlwjMya;lhrDYGbQ*-l*%P!~$J2m`I- zyiuzh+ad*P-}lnoB_uU^mEh%UDsuBJ3FEwZZ=AAGhM#e>=H*9_WHCp%gkK_ouRO2O zI<3r^M#3k}4?O*D{sCOz3nrr2iwwi!%FO^*@JS)Q9^J}p;X|+C3tJA4KgL^4L8u-K z6(48Fc?38v6Mh5o@@5(<+Kp)+xHN)HMyfqp%^UflyHur7>ai{3jEZtQ-Sj5Cv%XRK ze&(qiP7AJF2XyEY!$UftHczO@7=o_7_x7+pS>*_S99MovLh^pZ+EyGE)8|t`Dbq+< zm8#|S6h}YDLFV+iqFQcYcOx!4oWpX;7Csh66Q)1k=nZ;QjE_1e&xL7s;nrF`96XE{)pI3a=ivdI7V>Yn`gBethBww7gf$ZuGd`5oYsx^_p$z#6j86#W)$hn7)*%Dsm**q75 zn>ju1Gy2f+RcrFuQeaN?Sv$4YH&wc~qX@z0c?Jg+#1Q=qS!~IVUq41y6*!=hjWG?E zi8Ipfcf`YP?zNZov)1^MVnWhsgy%{j286@TWM8_;kh_9kGFKxHjDM0EcW*IuLOl{E zc(Uw0F+NP+I$aRE6i@Qt{k;;ASB{%@jUk|B?L6TMssg2VyYwAj#?oh?fWQs0KR}0v zdJoMXK!wpat~9n76nw8JW1wypa}%J7hU zT^Fx_;@Jut+_n%9S=@|Hpwn(Bjx}gdv2vQbTs*C)_j7Tr?h95>AN%fu^@P#(wS>K@ z#2phxgE~EssbHT5eDL_W&8{!s_Zx3)(h^$=+DJ`byEKZ{T2{t%Dnk5008UU&JAYK0 z57!+^poJaRR5gLW9|X9hPcBeyn=rzCOK#mSW&c%=%^SyT4F#h|LO{Gozlm!yHZ>>j zVM^`SBvP)LO>10$!Hw>k*%5VriHR`=)be(NcgV`3h?^M)p^5vIt(&Xo1&RrmW7_L+ zQ*dTU&MPb|h_t8I&$Os1732g6B{8IQa#b@&TSMc`ryKK=p946v%;we}1f1AmN@; zR>p~IL>3=a9VUGvnhly2n8I^x&wW}#e-+j{I-5PK03H7X64XI^cPIDil)g7Bm_Hd~ zdNCns`-kAMfBLir<&0~V9>K;V0 z0@a$S<+%Glh~VAtmTgfM;QVk;tpoa)C_t+{wFr{~b%d3$5ZR^enwnADTT)VVK9Grt z*_vLNVr6rrs^o3d`$+q8i42x7CZ3G83rA>?r6wa{x01tnVT$BZZ6%wug=ueq-AUvr zhs@FW#e2t-nZTr&>dfa&J#N#&Pq0_!BG2L5kFWqdkevYXEGh2vo?*75G5eaiLJN&L z)(YX+k!ev2QW^_|Ex03>5l~kNpvsf7AGYm}3kR?IFUAZ(9cNv#d*f0)6J$O&5PS7oxS<-h@M7fWiN{Es^<}g*9y^henr?LUOl|Pl@z$ny z#BB;yV)0TKz)?!)g9(Wd;zSXH;67sf-p8w}jB4Bu{A7FZ_93_Ahq8oaS4Bh2@Dmeh zVcmAL8sRs02vrRwWMTneldvQBg#{L`{eeCRGfYuK6;^V}98*zo^3@|$u~4Agd#U|w zi*^yD?bzSxDDo-JiwB+gh7=4h$PqSi%Ay-mf8xK`m4(n}Mj4@1RUpAsR1;DwQqHJv z+$GGr`>}uO9&j zxgzEscEP4O00S2ePBr_&fr)PFq;A$2)A#66JK$tR`UbHFkbO@(1EXevI)#u6Q7GU$ z-79G?1xpK`TN7Nei)Fv##Yq_g~<>CQmlah z&>b@juaQ8Z??n%IED)1w*}}ifS9iX>=kRVk-Sl377L3Kzvzy73A0~INX_Iy(beDJ^ ztkZ2_i*9zs-5M*i^YL*){aj-5WdC+~H|*Oq7w#@uubYI$En_mpR;DoHfatf5$B2N7 z26vM;IQ?ZTA}AP0Bi-@_UUUbN6iSk=Obz&#E(w!$hVOmq;8k2JzLWm`{aIx@SJa1r&W8x3@f2VAaa+-C9F_<uVeu_`#ID4MmqfIuJy^&l1duvzqNw)K5fKts)e zg$O3?nUu1>FS!TsrMw%zNY15UFhZmNN9vQ*af4l(QM$A+cYavyQ?wLr^hqHR0+V6- zYc3$HKBA>^BX^oC;F<9g_{z4`g1ca4Qf{*^0O0F!0iu*tthn>pIwkT5&o|m}QSrND ze6>re+&omLr>GJoU=O@fMKnG5ILpM@Qoovxp|JRTgrK68x?fckYrRU;zq?QmRiY-NWC*e@!78TY5GWv1(8)Mz0@rs? zek;`D?DdP}JU&6_K>XRkEf+b#wCLW!A}pCOomNg&9K?HZN2+R%dGgt=?^tZ$!bqp} z3Y^H}vRPwP+^Mg&QmGM}U2Va_u4hxjgUL&*HXTeXnAFJ#@mZm$L-Z^b_If3{zzjEL zsAFLr5GeQOr-N5Ru!WVvya<ba`CsuG88@5MX{Ql*4n0^C&BCt5h*NsIu5&g0Q^u)l~IuznGauv zSWTD*eSG`T2p&dA>b24JmT(2x*dPidGKQedIoS1PSzGAPM`Yx`>K6-Qa}gW=h7^J~ zHQ1)8Vb95;zr;5%0SH`y(GTtgP!#H_?Pt}=Q|x4|Fgu;q$vyWZw2 zcv=TW@u&Jhy&ASW?6Fnt0rN-sw94kuNJ4xffd?~*pr*NQSp$Oj*EjU4A^s)!@i&*G zd1j|LGZS{c5t2R|rX8Tj9Hw=#5^%V9yQ+g`;5M6Z2U)Xuo1Q0#j%}Ul6Dt!2oiB=h zo#BC}?N~uveeLdoa1$a66+y5d?CZ%;J=L(U=9)#of?NfO=#GzEoFngXczjA+GX5d5C&{< zw>3l*TAjmr(;r%_R@uQbp68PVLXF2=pDu_aCM&FNJB^NIRY8Y^R}?sVDhi$#>Q)xP&L>(;xf#^hth_Hyf0)8GLtDVswQ( z=hX`0#v8+KAKUsADUDja8nls+!T8y_KgxcThc~a+nsb`3tcS`U_qv`Ac`2sG7V@DP z&bynF1g}BA;A+^sDb0fzAgUmSS^qyHM=@*^Ku5($fnP&cRnW`c+s)a@&;Qp*c#%Gu z+ykSGIhRxEeN!0dyhd7`&0;;?69SZVzTh}fyb!)TQ(~yj*VguAV}PCejty$qwaQ8G zNTUJ_9CV%iHObM6ndCcpVF$0aBWq04oY?hDR%8xnG-TrTUyf?jzimcLA}NWa_6cf< z>?$8l0HHr$KMT^$TmoS1E^g0PXLsgv&J{C=&G1>;$5jzxc21V$i91!Pki!VY&88fV z8D1j;1^R`(ZG1h*a|8Ata-6$2XT-6&o~vKIgX=~MYf>|j-nP?dC)_&x`oRprZ?m8@ zOZ${0DWfK8FS$*+J*^v3DvU+X!5@piK9Vobjd(FZWXZDbFck5*;m8kgk6Oia^7H$@)vV|rEcyA{`v?3-JO6U;|6QuNf3V}`W$)tT zC#dM=?`4nh`#)5V8beY1_hz!kY}f1Qgc(mhe|GGUt4@$bR+TTe7-;$o8+0Lr`ofZ$Ih%-J;s066$~`#X96%NzYx{5@l|wlXAw zFoA(zhxp$VS=ce|&dtF;0MiwM!a{!$Q5C6wU5QAGN+~Fdii!!Vs0b;Fh^k76|3XRr zzQymd{S{963(WkdY-fW0hH3t-`MYrc4zK)E^SS`^KcXyuTlszezxwaLI|bXXp#Oeg o|GU;--OMlF{-;>&g8tdl>uEtS3{C(b!+er4-f3qKGn9b;01+qUhh*tU&tS9SNjeeXTpea6@SPUpuS zNyh%O*R$rc=A3K3TTT)b3>63p3JM6`KtdJh?|xtbziZnWTG-Rk{(h~9la>VKhYPxR zL+e{@TT(`f%{&thSQk1*gkoAkJRJ{^N_f7&4alJ}u!&edR2uSW9*5ni59R}9=a(31>_H_Ezc80VrmKK2unzH`%sGbih zF`a_GgsJe=LK8nU^c+NX)M=c{VG2yg7T-I-A<|p>o7~SixF5iAJ$1?$(!s=N|%L(bSd&i1(U-N0%mL!$F>XC(412r&z`!f*wq> z%*kD@gV7z4uAnrdRs>H<@SFj$IecNL?4HiR3gZ$~k%>AqNv#`EU@bj`w+qS%fC-pk zzXb7^#ga?xe9ShUOh99UgvKZ!;wX7YlkIJ^>~Q)iY54){rOy5JA+J@#p-j)Gc6Sj< z7pB;EwFbg~oQ<)nuI=Rx*)RuRq>|U!-v#&406mu6OZKwrA#)4q7v;$!O0_LQ8U<}+ z#}?na*mn&Dd1j}rRI}BhWG%*m)^ACQ>51IWL4scH)p!_o`l1UA4;x~>)Itxu0&MnS2hI{JOm(1LDoA0ES}*uM`d%a#$=3>XM# z^`9OT(%&DHu8qy_bIOd7w(g||bnpwBQh$Rq#RfV2GV@tLJuYXxdLjv}!DwS)QDh|3#f$od!A&ZTjYL*Ytnl`B{-mbn=P&V3KZ?45T)f-BBb@{6^bj~S*7X7MLB_j7Sx;D z?_zLP`*-1LN4Vx+0A4lm%p;E=#w9RiGR{z%{zXbs1QzhHsG7ZhsP>%))06 zg5V$rwt7jU4jPNa&&+=~XTnYP$Cun8N z>T3wGXKlM!e;pWuiiakp?;SwRgG$~Q*v68@hvo|S1*8it9>m`p#nPScgV`4@XF9c3 zUEb{?9;S=tXAmk;kJ*1Uwe0cd;z_^62Uh|di6!ViI~MA{9*ez&sez%iuDyeyoju)O zZs))I7sKz9`g2grsGDD8V-Ce_|-ZxmsSHKGQdTU1Whs*RpI`8I-ZSLzllAL!DL=yosO zuHSAT2zA4B(D_+Mfv15frFl-3tpus4@}SO)M+x!kRIOjz3Y;<>W4u09Dp#W)NzoXb z_ei5=5w=UsgewTj`PUx7c31|v7Cb)#Q#Z|)s4tPO5gW$4(FTw5O_znV_c-V&>30K! zij9Q{dnp!z6zcl5GzAv^BQO$6=<#RdhTx=B6oy<*^Sr-XAM;8{6b4XZ1*m^kW5WNc#{YM9&(y}ORzy(O9;l#b zYrlyh;OF9u`J((12O4782GjgTO!5hjhaP7{ISeGK&NMcgKkPi*X^vKnKrEiMX6F`T znC^(~b}}Y`oR@g4FcvBlD7(z5qbYQ2X9vxg-uALG(I9s*g5G<$k zRnOZ(8U+kP1ISRX?k4MiFLwB^n9kL$KR`8^{CP<529-$#mP!md+(i z9y~4+IS6U@Wp*kt2Kr)9&r*0d>claFE!`Fn-c}P)16OE61)|-*?Ou6NJiBIOGmv$9 zCPh7=FC7cBTz=*^R>QkL^l_He@bg#z&OufgtYLvNd<0zV*IBs1jg%9N$+n7;Na!CU8UO zUg`C1tEKKwmu?L?&(gN3F6fge6179q1-nJN1fHbCABAlr;k$;lS?!-@J9Zswvv~Y= zU1IVjI`Z~KWjcSzxlcrrh04r>Rzf2alE*0WpEQn z`l>aS8>B!OIR~DbL5!pr$oGAQE*DW)IEYXdIV4O9Wg=C;}SLK`WE$*v#n9=t1qK65O&QP}9e2{i=lGMc@-~lsB>ncmir26T#T5 z69j3r7%V1t%6&dbE{`nb7>jmbMuu(Xv5SF^a=snV=vG!?AKcG^8R9GDQ^`e^a*nO( zJP@821<^To`3n5n%=bHfw6-7*49OOQ@N?5fdqKHo3$}SYzS!xYU1-~;{flq*ta=1> zlzOjFj$Vl6E?Mnpn6TMb$b+waw@1)msi|NXk@lR>a{M6RED^To-No#w;DxkPPe8tu z>95{_vt1Ch?JYQXiac%jd%m2z{J5#2F4>szuG!dVGN|}E^f6^$daN@*=;jXOb~|oI z_T;wwW@L)|l;sGiUZoWI3OO$X1q>}7IXqd1sHcOer?C&2U?(U1Hgu(QF1(ZL`I;T@ zFu@TOKbvN`+F0Sef(Uyj@<^{wz@r$g0P&Go0u?J+0I{R8ADl*XA1p&mqX~+NK8(ba zrjDmrf-)Nj{+QDA`F0OvBgJ+E)YiT>B=LiI$WoRd&MvB9-*d`YUw3T{{XDP}ZI@&G zN4#UJODi*;k^}lDWsDiQC>8pSq;?6d>PQY~Ws7Q;VCOqy70%n`@zFBc_HZ*fZgGD# zG%Sm6ap6c4)B;L#$E6JXfu*yX`Y%O!y>pEfygZi5wL@O2AUpxT&bpo+)Cre@pKgKP z5th{1(?#O_Ig;y~;xg_icX9NMI}Q%Ql#P`gaC1h&-f#C8G>Gl<556s|4k8q7u$5iI z&uM6Bm88N8SDKEgqTV`07Ffq~KXyv&mgvpVw)31SL4LH57pc-9|rdqWTo=#6^3Of)S zEs0QshT2Sb1}0qj`)zM726|p~HNy-_(%Mz>aD86?D zuoUS;A~!OCtJCSCxxy?``zKOt+U+G!6flhzd<#q6CzsBj{lKDmL;O9K9PJz^owM~!U~07V;M6G(`c7XhZI&=#QdVc zzHk6p40&(BUK09cTN+aH7KWnNLgB0q_4I)Ac|Shhbx*_Z<(8MR+^@G`&QO-j)sFsp zym-6U42X@%mw+y#Y8`Y|+>B1o;)#(|ody-AZ%0lxh=ip`mmK;8&5Ux@X=MWLsXBou z)Jrs#nGml|@2)f1>zF+0n}8;Q#^JNTTURbczDyPGGKc+<*rP%cXeNfaL9)E#0=7=@ z2F&+7aDsQ6*NoCdA7urlu78>Kv&sEgNv}U7X1Zpu@?EE>X){9F2bdJUhOM??0lzN^ zt1?fmnAh*%ihgy>Z$TL5@MjSOz3C2l0rVZG1|)$vNQ4VI>+l!4bIT~cGuaiV?I0_i zZP9yA*sic0c8$+2A0dd;9=o)lwe*)gpYWHjhNkP7FSkB*)n6_Rn0-!hfv5bg>RB^w zR#;aqh!|6D)HTsdbl4Ey#uDgT}CiN(^+%Wsx~6+!j-i$jl7* zOTWm}+m)0K78DARx|B%*%^X&&_q-|W&6;Y)WVLf^>izC^=cec0{a5HmjH+0lUuj&D z5Lzy35I5iM;6K;H7Z%QQ1}aI+zn{`zSgMH;2h1o^%qrT`t+J6A$;Tu(ncd6e&fZK> z1|62DVH#FnYZLhwjk#tB*yzF3wR$Uaw;`UWUSQp$XIm)SuYiZ{=}s=Ti+~#2!pRWhQH8h19vumgJ^WE*E6~X0-x_`VUW=2A_xc^5qXF@rF}hOBMOa!={r< z^|CwD!r@k}LnO_ZyPQrYy+rs?D)=sF7^H-b9V_vxE~@ODGUT zN_p@gW^-t{M}q#|=9#uZ^mScbN@gP4gMvPB`4em@O*vVSjgxL5Q){itB}J`*G}E58 zHq(nr?6dIY?gxtJ*!~|@-0U(z&J6~*dOrWN>j(d@toSct1ELNx34^4CzPrBrOKOF~1i4Z0W}d;(gR?@*2PWio)F2GW zX9IBLjJ@*`%E3R3_!_2vdgyq(u*cBnug!cj^qut5m z!o0L!RqcXlMW*}{ju49RQg#-YDf;oi9G+&cvW;fy@;m$s)KSJ(ZKpP`@67fC3!49R2?)NP|?8v+I3@DNW_&+P-{~<1< zbX~0-9qj4;h>O4c@#jjfP*S&ErbqSKsH*;2VqJm)qL~u|^Rt#)5f|$Wkvz zPPXZX=THlu=}`dX^=ALU^^-?dbmse+2&&0M;(X!{=e2C4RQt8w_PuhqgRp-)VsD5T3)#`!0`B|lg*y2O2Ck5ooK&XhC`ad=N) z6ZsfrOs!A+_Ms;po)Vp`t9D1l{&vy6Dc|@MllCGuC>oJ{lzcI@60)6Q&D*Fb%)0te zIG9FD$2A#A%*b;WVA<66NYqgAuERL9fK73IpqyYB_$TZNK)dJpjFK=~R+4V%?bPAh z*r0mxVG1FWpB-55(e z9>mT*3DRw__fxWQ?Hh+{ ze@qyMzeIm}-pwYM%B5(Tub3@Qy$Cd85s-!%eUw$oU(JT{DJ!P*1?l$JlMK%XF>G%~ z;?ojV#CmShnr}+L83Xlcg>?Cr7gX~$>)XRZh2@vd@4ul*laKN zh9%OsbnA<+c8Dr?SC;JpTYPJdj2TI7m%Eswpi<+0Rs1XH@9x|&7BQCvSY>Shp7D=e zpMTYzjK7VAGiA-+-RZee6$0qad2y%&FFqyCW$98*#lzp-DUZ8S3m+)`a?$35OipuF zieO4NCa9;ajlbc3J|aCinGBAIE|snkMjM78h)U0Im=w;wGqUX9enmv>)IwGp^?7SChk}jAT!ahEkm*u89QujeDSkG#W_L1}9h- z8-p05c)4o5PTSNdrMh3BxsC7EozEa?FTpd`K?_J3I`{KKwowvQ9Rh`f)+QRq!tC%k zo`R~P-1iaD!o+cIHB_quv<4m!WNTm4TkcRaA>MH%Pw6&`NX#YDg3B@&pg zfo{NcPr!fEp18JW~p5j2|S7{=(=ln1N$y~3A=Vrz{0#t%8^ls0XrV-9;6q?ce= zPi(p#(hl6!cRC7)A=ymO9X@KBTiW9lzM%^9lEUbfpDUmJza_M zC5tov$7ygyHE**N)?4#}DRJ`9ga|Jd9M*)g%j&tP?SY=nEo=p-uXU%7oWwRlGxwVZ zeM1rUIgoR-WeI?zD5x`FoXIjTv-ptFX_chwk`z0OLtA@g*gUx_EgygEty}+wV7>A6 z05ibda|CQu{{KUe={E!ylr632=#kzht3o3B$v6)xG%3tK`J=}a%GX?hY7Ck7;jGlI zRf1n_aBxkwS{X{^A?M;FA9#3qJgn}N$*z~1<5i@vCZY-^;0%!K`Hkix-w-~#aU9*s z8&H1&-%_Y1T&DM5kNA;U?7KDad1<&H5=bBg_;YSxgPBiF4y-bYJuw;zy^4e;3T&2J z6s&Y5aO*yM}Q551f`pT$E z3t|hnb88U*2sAHDEdQ#8e6gSCkOdNV?lXVO5V|0wO``<)8-ZmmhHL-?E=Bqsk1VmF zx+2`c@SrgJ3CzN33Hi`dhESaOnXZ4G>$dxzv}!FC?%Y`nBAQlDD*vVXLF z<9~9b_v#*i(vA`{mJ7dh^V8<1mvtzoRcM->z>??)Y+)Xl4|bIiFt87kc$E|}G5yXV zIHbo|-qv)-!k2_!t=HgPM#vioJzxg-w@GbQ8AG;$-L^k&HNu*T-O=#LI=<#-<*%`y zxg)j@$S*Gy{zIn2wjEX!W2E%LDS2OQZryj|? zcgXs2*d^bQQ?QY?tYA8imC~KN<=ulR)f<)NTVO^U_1WEaFj!*2n^)S0wqGr<<}XC( z<0-X~S?{t$k1XX{!P^|&io&^#COQ=>6+8LgYmX*X?Jaz#90|m*Www@GIhk~x9&@iZ?=pwbq22}YhVg! zQhem_wsg&&Z?6NZo$Gdx&9eGcP0=0ago{)bI842YdHDfN&IGF( zNm%OcZ@6!N4EMwvN=QBc9EJWZI5Pj4h$<^qE9^*siHP()TlPdQ!SHEzQJB`Li7y6A z{291jXlYq4pKms5P57fjFf44RU)q1(#M$@GUkE-pn8`Rwtal{|(TWgG7$gq3} z2&qCUnjVi(mNjVw9!!1Cmply^Rw3-dnU@*-p#{F1Q$MTSSuh~NTOWj(-V^{PWiZO~ zpT(4=x;1E(#Noe|a!jIjfF8!}ki(&9%nA^6b-Il_OE63XhAua;L=eT}} zQmbSoSb8ckBjw9p3k^4N`gE95?gH z1A*OE+bgY5^__^7@f964r5DIAdD~v?Sd{ogs@ed)z|!+OXXR+UhJU{px?nJWLdY%T86wV(eTH;p}AkSfEP}M>VC@?Yl}U zb`hG5nvTB5ebuKV0Mc+@fF%dg7`|186aZmL(z5AB{Az}9)}hAC(g=R*bw})l>32U& zdr0G)NsrK%=3^y=oW2cu{H&WyTqBz$lu3Wg0RU!60xNzBH7FP6mjMMTT+Zw=SWB~C zg9B-S8K@a;(#=rk)FoU-7H6~$XP|5+-%>5h=ituyh(5^c%Rh`3(-^E8=UPg3K0tEq zV|cbxeV$)|6z2}^%Yc8*I#*@qz<*e&xDjVF9ay`GSH7@VwT?Ae<&}=RIywsQ5PtAQ z{fvSnA-$?@wcg>%#dguS3VeRDcyg{!#{uS^dNm%~7b>#k{L&Mya$n!&3 zucDLXpRa~ko_|QJ_$juv0RXZt@c+&PW%&)+3gs26ZTjDk4QW@{D2`{Alkg2|bZ{;x z74GoxvBe1itR5}NPt5=tOQyG=KvS_ta62M8dMYE8-GL4r9C80QgT)Jr`YVGKN95*R zV_qhwPYOw}q4@9c zOJ3Hdbc9L{17}Egu@`&T>O094jd7oDojssd|}gc_~Jr5*8%6I6Ywl2^E~{U#e&Eppt+=^GaQTmrm*jVWuh#zZ`sc+fzCIv z2-owUl)#|6fEC9Yd-XaI9Io8$e5BUTML5HyFv^Vafgk*Y5&3s$*U+q|Y6Ia-IHa8? z=8oQqPL3QN83Zw`q3u-X<2tMhjPGuOfM+A+Gptl@K;Eg_~Jdg)ys40uaRoviQcGeea&Z| zTqg4BYj{(rLD&unCyIq>W%io}lcuq@dj-~aGhP7j)XFSV*FD|VYAp#^87PoSMqm!& zdK~tK-U-h=@vVT|apMEalwe0LUO@ZMF)~b236mhD+tBt&ybS%mGgQaLeoOg85*O+q z#Ow-mx_HqO%0+Qkesu{)ws(OifqVdQmX?$*Ok#V|H_6S)Y?M7b&;r#Nh9qg&V(ayMfQvHVg*Ma!1w?0zY zE{hP$=5cgxB%Js>!mlv3S?!qnAv96PZ?}NMtk$<>NW_616v^NAnYDM!vR&K0s%5NW7wNp9(Tu#iqFZ~j|6}hHuS#(FV_D8UMOM|1%Y9+W0i^x#mT=l;u2@0 zAvR3gviC;K4yuW1Q|Tt5?cffR@uU=KMh2|}-b5;EAS!pl>CkU_T8X&Uw3xCjuTRPycgyxPLR;xq+S=u&bR^x{|nZD^n6HQcvXF?3ej40?^I?- zYP5r*0Roi{!Rsg?&I*0_)2NN#hM)h~t|RlT=5DKp>pgl5o!Tvu!}+Y{)jloprx~+! ztCga;6~w8zqk$JFG8caUKP^0UEMqRq`W0ZdAQ8!)a~iz+c}^Z=I8oiv52&{xUzQXZ zIp6l#4^-=>SjfSd1)hYp?n+RX1GaL3N2^E|q?xR9_JTlWrx`hC$zljM`5?_p^h$SD z#vj78Y!28OUmrZbzD7kVDOUY9#%TTy@hI-8p(fG%YTy1uXC?OCnnoMGRRv9KwsYQP4JD zB98^TLti|sL%&OcBT50ZUt_6M6HXm_gk79mgzum}rnn)XP z_5HSzCld_(U5RpUSu}Pfk6&WNyZPJ)xwoSTv92htBmVHBCecX%~@{&H9Uq;%gqbclHXmFhdms zl6@>-Fct3E2O~8V?g%B$JQ8`)++3V}J3`6aEoB*7(h+XHWVqnYtj##oRr?dbJX49I z9A%k7I;TZyK@1tb`BH%KB{SJa$zN=gx;PCj(G7|sBFl8We3y*~GQML`qc}cmD z?hJ@GR~3O5yjU~1Lg5RsGO{aMm70$#KV!Tfnw#O=O^lY39JePOkL)|o&Swp7gOLQ- zYRGjU%Cz-rI9mn3t;Xqy_fP=`+GdD2B=YRf>e@Wzl`;SF1;yx@?@M>}g`)~RFHTkC2*3iYS%+53qt+h?k+^W+spbDB3t zr#E#cb4z;)EYto-ba2iSJ$`_AVl} z{RFMDr7XD>rIav$@RXgMHwnHS-j+C8Zm?xAq7?zxcS0GQ>Ckl?i-=qcs?^wY<3L>+ zWtZN>cQMFnt`&}*;A8U_XCnBT34MrCl7#SKQ7Np|9AZX{-Pr*|sr#+-K5M)x2c`AW zTz0gMAr_zwWlr%TQtl}ZDgfHh^L_3icbL9}(j`LO_o%qG*}vDuR$ml-k(t1XDoOC| zWj&P|%08jex%i=hnfYi(M0qNNYMrF=GPwT^)D31?H|{YS6-`*z4u>@?EWUDX^+QB{ zIcMyeL%4%F$?0)WPuwZA;LUyjxnFm7O@-#HdIdgj!^{<>!NIA7#I5Y-?0sGlUD-y( z9>%A!$w_4VZEUjOUrYcoiJ<*F>smj6nDRvsZ| zzQ4oxd3Wulgh$mLvbjpXsx7=@l5laXUZ0jd!zACekqxh1jZL4dxFbK{7bLYt)|i>P z{VtxzA8MLd$9oX~fOIqBzcV}8{{rd1`K#58roa7F!s*_5E5md}M-xkLA90-F)02}O_bG1M?f73~aXH|LrAQu|;G&(u@MXeHbhsP2$NU~Miq*_ z9Q~3g5UPP+kn14VjO?taoLg<;+m)#@?TYZf@yzxA;+f&eW3=vs>&o_eNia#O1qk3# z=HwosjUkbdU{=CTkh>8zLj)lu;0H=iwU!>MDVV8wT*)Y_EPO|ZZG)TL)1#J;{cI3? zxCMz|wVaen_cI9P6%$9J5K)Eb&5(BNrLzGj&xGTWZ=A}0XeG~~L|zTCsBLg=X^8Bg zsO|=zC}Xq2M+zV*lK_&k5P;{|KX|?XA1f9DCirD_2L}R$@ufp41U7r{5*ERwFOPKQ zYguYpE9-!R*B!xn1V4uX_FYP5!(k#@@FrK z$zx67zKVa#!<{7yhb=G%NXnEJOtgq4VN6lypamKxRwwLMJKvMN)M4d8QxMtBSkKK- zbYWKbHD*$2|4Ro_opiAVhsPp&n(JO+MWUt4Gk7`Z^y`KX121Uo9N4b={IA$3;yk_= z{gpi-OWb?ycSjITQ*@woVQ}%Keof72Dey|@O>&sDl@?2JX+zcr<=GbL;k7e+Np-wV**2kEz$UUi97| z5gxET;ovtTz@+CD+uhsS>NvoQy?qLH?8`%&D@@W0h>OP*mXcHbsmCLreNpKKiuzen zNW2-?p)T z`dG082wyjR2{_Eng5>^RR40@(eux)zNO@^mtu~1!tIvYpZ(6K6Kl(K&VRvJTg$1xT z8T}!*1V&!I3pyw96OSg;`9|Wo3Rv8k4+~%rVdY*VmqG$)aq2>QBld4vECQg#r_ob# zr)?}@U*p}t2f-q9GH#QSV+r6f&&|e`=w<>@(H}#J#}7Hz7mB$iQp`ga$SSNB6r=Ml z?Zw#2N}`Tu2_c07x@@85oW}JUSTc4ldU_1s&&&*(D;;hYAJDUz)S{8dU3Y%CI44wC z3a#k2T4&Jm;AhBUB!5KNL^*@lsX7q4ud7YAz9$veAEx~FZ>{5!yInzsE7ZCCfAix9 z06#9aef0zN+MbC!TsQk9VWr97ugw*bHK+qp8|@~k#XGsvY`DoR3zNKqaz42|mp?4{ zF1{Q+=ljsDj#Ci%)^C8)G6-cVlIb)YQ(hTqsmb+wjZ(V5O4loVyZL>M;m;ZcsaexI zVQ-LQViW%B^+w>8@Qh}kQ>5xI0zX>f(@y~ZmRD6xb%8J!9k52h2dq&vEZ&UAuouAv zA1laE_7Y64!99$<;AK{SmLzx3Y3rW6+6P>(GMJ)z^mg>06}EN;_d>azsN*eK^7N{d z`#ncb3;ha!lxU)=KXJoqy|pDP1Ks9aBdfgqGovMH>&dtM!;Ufhvs*qQaxM6NO%Mum z(es;ChmyEc)fm{muioEueP5TNe)CitiMBOUs^02FmOH&|sXO75za&cLLE!3) zn`J!vO{-P*J09xK@rkUP6*s8UW89n)=Bb*0wPb&^j$gBPjqhkBq2jwg?JZ@qwTRWO z8z5)x`ST|kUkz*=GQ_kRfC!j|x(}VbWNl2Gz0f%Oyl~#%B%{qV)9Q04!A)5aRuYZ@I{_7h4&!+QF zsv$CZ0;ZQ9Dd2)fc(HBir+-Mw{TBg5t$ZFnu|~`4C&JKNYwIW)mrHi8x()iGdrRZ{ z+>{7s5Hl@cqM~+ID^!~cx)sp+iG7DW)qP~~c<9hX{A4gGj$k4^1r*iPX2jhvrbiP) zTE-~Me7E6P8xwpR(T(4qMl(pKKH{)zW5?g=WF0X`!sAm{v#w@m-dD*W#t`Txi&{2yuhzun#c zPcLx%zQF(ETj9?b`=48d|6VKn;phM9Mfl^N|ECw>kNNVSUW7l(&;Rry{PDa0(~IyY zmGYlng#Y23@xM1B|A$k?f0dB@Ll)0eW~~0GeHA3i=^~B<1M#G_Md4QFH6H?ov|CV< z(Aow$Vtzm#ME7G|5|9U(m-L(eE|={VGdwod2Qcw+`*$8rf^yK5TG2glghX|H$mxVs zZ=Tp6?L2>uf#$<+k?8u(*;eOpy6`@oeUD(Ih6jPF6g9x1j)*0SaIDFOpAi*M zK+sQ=gJ+mlve%%rgPn%&v&A%OCS*Z;^?1s7QjnK!A^ejE`2qT+w3PO+gggmer-2}h z5s(LIn0F==9&vV&)+EmnHjILqaUrK3pixqoN1%amz6qO6_UY#$kiN(qiAjxABviS~n5tZ&vM+0F;K!yPz8?17I7 z02@EzoKzh%7cQ~9S|B5-+EtW}>7xEQU{)%psb&%Vc@q-BYdI>lsxe6am7BeJ9KTXT z#vfZ{X8X& zQDK@FoXYnGDn`A@2?994UVhc2A6!oV2@u(u$|ITK*~=|!qbYhYhrD{jtQU2Xw;quM zES}XNAjA`va4u}w;2`Cb9$)j=;dfSDH)*02mpyWG`5PG!P-)7YhC|9H8%zO;-+X_& z+-RJpm%X*2{5++mp>4VWakZQ`#7;lc=n@QXNn4{Wx42Tbb+FtVNsAiH0yR(`@58`U zNa-a+jTob|4VY5bz7vqxqSK+e-D>)gko+xo*!VKDg$1+brAY0b$J_Vq=7e1gc`QHa z67S(lTS4hF1WeyRqB)078>p>H4a_kwf{dbcN|8gGKwD*AjwHTwt(%a>o9A|zx0nWR zc2P0DvFanAyg0IPU-LJ`v^_0*-pJetQI!e(FgG>ox5vbRxKk9#j}J7-ghADJNg7^X zwOLOyaV>N6qu`Nm_+{F9t}HbMFZFSF+xSI(Q%uyr15sJebFTU1k5Q5s!KoS0y2Z6E zSjktgH+ZLGTwFryHVAFsS$Xc>`4y@*adiCOjgj_U@%C-$B!b8~h`B%C2K6zYk-SBl zUK7I0LKe+*^=J|}igN_aB0nUhI$N(#PPF~D0vPYFJx}2gdbCWx+RtnW<(;;9UJRH! zGTr&PB_&+m`vk7|tD2uouEzg~S{lQv#d!j(fR=yi#{Wy}?k~?*eg8|i>p}$z0A|PH z^gT&^jCE2g3@C*>abBN7a&_!5RI%k6_0ukAF?rjJV$&F`bo3B>12guo865V^=$@}u z-x#f>0UI{_r!DzS_XHHchK=F6OGAdsbK@J<4IER?(eQ@CUKe&3XJt^Bp|J~<@-iq? zWsu7*T|;FM6h&=5iSpkzY~@?Zv~r|ZEPxH0Fn>%gGRCU&N!d_&6|qu+c{$-KE7B7t zQtSbE4w+xM#9tkGnsAA=094Y4%TYhE9HGkS5OQ49(tei2L~?31%+mujrz{ZUGv9<= zhS0?CS{0=TI?6HQ?K*+$%eNk~)>Oq>w940Wui6`RUq@L|1iaw}h%u;d*lBxgR82%a zu8}V~HnARSZc1Ad64E|td&evYv&8F9d5&M-+)23!8*So5M4z)a8@d~Hu78f2@wBl~ zU0s7mWNV?eKCb}Byk#ZA?{8yKsppksTPC_=HcclRlSQl~8=J!?!pZCSUsn9Sm#8AM zNlF@n)S+k(BV`9tjCqqAI79#we`UfD0L72skPc%2FWy5%QpM1c^6>Ro+<^;XFUE47 z>$;U68H@$AgIG#Bs;LQ-ZsK`TKOTAPk1bp!Ps=3wj1*tfn!nlKB9*3aYV!21Ri*)&Kr zxbFr9f45vSg6+J?s+!EXYWcHV`<(R1^f-Bj2bXR|p^`bJovrSL`y{THqKNjjA#<8p zPE~_nx4F5R2-e~mz6otC=bgy=3+y;fK^yL|=Wj2*ICdd=01eV|JIqU5qbs|JK5CY^ zIj@*Fbe4afyfpnjbN?Mv*ff}SEEx69zNUYMDV)RtMe_9{jO0|r^h21&+ea;zwSjnQ zL2a9$<09I_qw>L^h3ekvhZU~UVD7Q^om@;JX#AgAvY=^{GN8-jY zBJ#%57eAy5#pzOSWj=1KK`0nU2@}$q*`&HW-;8oL&YMA3i)wzkMIGb&bx3jT_X5IS zX~r)<{Q=>xeXu)nikvlDoCUH6K==!v=^f$s(i|z(_58OTzX161m49q({Yy;$dC(8xl3-ylc$iCpg%3Xdkrx@k9ofHR=ps`^7jqdG!Ii6UW z_9nbGB^Z(v!sWCoxe}s=iV>`Wu4}J(Z+6vYZZ8bp!0H5eC-`AltM37cRiNbl;?)>+ z;OmdusBctOu{G`q2gqv%*KPH*A}y@d)c|>$a69m|VtYC|yN{^llpjsDKGzLw$VN?} z9{}d^{yA8Forxf8J4;5xlzMJ&v7woStw^U$8j7z6pVeyGCrl(-7bl`mJA9G*7&xo$ zc<`amT67kXiIv%R=A3IU3}WFHx$AKl?y(0faJNLfGaTzX=kq5o;s!Sl{)?~*J6l$^ zW`qmIy;S+wf}$}}=8%Y&#{#8*DD{ldAQ?%45%TDCg^}aXsaYGe=5qJKv+NJ+dO%{c zo9HcE{2S^FtKv#Z1N?X`c(63#x8CjOka$YOrRt#sIIJx*iwkELtmMs94c9r5ox z10deu8>y+FKT%H;^_E_q;qON;1j)jf<7a2jjx<{f}SkZTL-VHk^#aVr`vIqC^ z=Q|UfKFJCx0?8l77Z0)tBON=l{W&5ilW3nfRSX%ubl1HOS~=E4x8wGc$42+@&Pg8p z_DPTN<>x;f@S^9mFl7fj$j_tQO=G}1Z0ye6@#hG{-VX5(Xk!@3-~>!l+$9%Yc-JQv zUT9o>UIdA)RGypwX$Dxxk!Z@m3+6ot-x(ZoFPRsyr|8Upv@S7RE~}D_hScOU-&w_ef$t zK_RVoxCg%~Nk~1dwLQ3Btn&}`2!cBE6wiCQK<-grDoK4`;65YAoy3*2WJvke_Dm(Z zLcyYxLqEOcDV`I(U=py5KI`s;iV1n}1m?4FokPR_sTV$M_ zL-AsL#6x3|(B^+*jnOwin9pe&tfHIJViWBaynI@n5}9*FV+4;$L^voVH}GtPL@_Zn zO-)IOc=1AK$}nL~h0~$$h71jG6D~y_-9Th;PDx1v;d_vfk-?KN$S*4B=vs-Ge0A7% zs!bTba6}O#4MTtYwY<3a{qpX}PHMBJJUcime$KM+R2>Wn>z(H_1UE{&Dsb4{$zF;MWzm{Mr2+HZy))^j8vOh=j1|Fd#kKX-?C!ocYi(~g ze+V|9NPBB(%QI<3Nq@$pD6B=yHKW)m?s2rR5Rsah=> zy%@~i@aTvs4^0~GF>_^tO)*|V^21SBM1ci`ot|Fr#gm(|;Kl)GRV0n%{#P9TeDD=v zkN{Hdu4@8=EkB@I&UfKY8EXF9O5|idpK9qp8w5(Vu_WVe|q9W4Q z@6%;weewo#&P!m8k0-9ra)kWLR(?iF%EK(@xiaGs0SIle3x>31cVy@3lHC(*g!V8bP; zeM)%#_R3mD#@q{e{;$3!=>Avr?rlJ&@>OWNd!ecmF;;k$#&={+f*cI=3^U>*AXks8 zXRv#7OO5ue_uD^z{si;D`&rFxXGQV_y8k2Z$H#?`jpIT>ax!+KDMXjo+$=KYqYN^k zh!-L%H_Rst)7hBl=;*L87|KwfSucXt3s*W?dCkm?iT0!_};X0CAeBgKFS!YgE#JpSq>p`j@tL$<~Srn$Ny? zu>4BW>TSa8KTNDSax6DI_xH?u&?+rM&uDugEvl~Wtgp7!YGFU!SN;@#-LNQ#T| zyZAyXK-Ozm+@RV&gA138M4L|SE_B?;yUGodE5FjPyM~a9I%Rv_o-9X1M2LQRC%4+@ zK=H|CG>seJuAkxI{;{GahKp!!77bxxALFrUo?hcmU>zzkUn5Z1N6>=5vdtBXe*A!Xasm2iaktdu zG*84%$N?52!lb=7=K}3Hpb8#H1pg~#9)nsMl|=ShY;@Sf{RiU-XB{FZ64c^$it;7s#^q!Q ze*s}EfZ$}Y0d;_ghzOvLs1Nk?^ipId4;#G^9v&WO(yrxz`*B=qba)So6f?AaCcj6e zzS8VM&&GyTfH(x6E>ZjT4gV9d>HguN25|%O?o4gb0|V!0u!4d@d>4mss3{A+WEG-eyCreP=G}p%EG8oF2!F3N9m1h;R98Ii@C9uyk_Z#6hynfBl(Sc!+=U$}wBt>OAyAW_z1(O19> z*sr1CeRh!4aR%i0$9p((Vgqgp2@$*go*vQ#vi}T^9@!hHiCX7mqhh)S?T(L*v>I%Y z&H0^nkgf>q(BJ)a)78--osTi&Vv@8Odw#qPdFo^~{Gs_iC@90K2P8ktfd6OzF(0_qw`VNgySRX%+ypsd^bh*T8zA|?lPb* z8X6iLgh-CYxUI?YsDvCyk0(n_tPBiCSNl`{h&*6Fd>F|TT;3US!112UpHPu?@9v~xGiTOtKcX2Np z5zH*Cte6-WfArcs4fnyy-skrG995kcGfJld>R*ECjidsizbPl ztTF>28_GjRtx{|_1kb&;x|f)iMkF^@Nv?@zWM-C_nYsE$w`HZxE24K}FuC%cGczhb zpOUZ_Qe!?gcsO75>8}X9E0*34lxA52y3z4`4V*A;O5E??{|Z(owtgy)lfh%`jj95) zzA-KG-h9{QB*MUeb_k*9;kHD*z@un+OuA7;ghDatG=8tl z1{y2}O!G_`J`x_i(iQOe;pX@;?@^Hi5h3W=y0!(OQW>KE2?H7L^LJf#W(Ed6zn3Q! z7BRMsjSYW36=h|hO!4vYEA_?K!J>Sb6IgSgjo=|i%Gc!$iRMbpTuTX`e6Fw$Bz_}QQ+A>AXh=(C$8!g7Sr(5RBaOzlj}luY%HuS zMGot&wax$=8ym87v~N*bR+(-h=dj2Ge?8gl3`>jfdN>6INj`{+i2)`&_aan?$L%C7 zC8gPXECVp7qJ~Q3R){qfj%tf5EA3z7s1R|PmKy9xQvPn9zuax2s$#Fh)GW%U^N_6w zVUqTM2?l}c@&SyY=KZvWezvpwX9&?c)JEpPr5%Uu5*{~rr4|1NvZpLyaKF4mku)$^ z11O4Hfw;HluR10sW-WZV=xeTcf)`$biN>?TKQu8i$}VhvOq+L*y;W4wkN>E|zV5W+)fcD_v^~J9i1a5|RT6XDZzt(jS7Ek!-3F_R(9D*7 zyPVbcGc*N!v^Q4He2E(Ur|-7FYeWOFB&0&W#Z~SPEQg@;o}#*6 zKH4-`FxbcA5K7ia|b*M&<~fZJ*8ID%Qm4^ zR3ff-%psl8gq$aUXjQOC!(ij_x`LIICK4I7V`^e!(NksM62}f;aV0bM_ygI1Otn!Y zimsqA1}KO~{CPc~^6Kj9d)?elJ6Xo;oK239KBP;%YLfxr9iI0KGpj@Xh4&hXiIw%w z;oTP2;s!_lHF&%^=3;}#7I-|W8T|E_!yZvWV6}? z2M34Me3XdY3UYtBJ5TXrX=y2i-5QqHL;_{g8<==r=RNzIqj|uN9vyCO0`Y^8D%2g3 zJ9y&Z{@yY?xa8(wy3}&2z|e{`N(u)k=Swp$eoOV_e*ZLdJE0MFr=eQARiFY!xVgD$ z;8U|8{ZSxMh?1V($?558t|Th_^JKovs>d#aB#2Izmvbeb+^fJ|h&6fMv3Rnl0l49WM`uK=GnA zWLaRrHOveb1z>2DYxSB9n{aV)mFcw+w2vwYBD-DfP0W3mh|10;gGQ?5 ziBUUPuXmyo%Cv=-n_Ku6-8YNA#F;sYmoDI%*?yP)ZBY3>V7U3avsdzp&G*nICMMcj zG}hJEbF5e~30hwP(X=dn1Z&`z1D%{?QDGt6H#ZlT1gTGRAZ_*l{QU3Dt&ya{b20lUa^ zEbCK(?pU9853yqt(ST-CnPRJ%xw(FaFFz&a2SX&f=mdL<9uD9dH&qv9gX2J&KAEec za1`497WXYPi@WGu%X-}d37Y2n^0G3tw#wyLiAz<6qeY-m zj|!;+b*M(L8LR_?MS*~cYKH&Wz9k*=VM%doFh7}8Zgh(799654N&@2K*kKfSYoP2~ zO|$E9v*}Qxg5Y!fwxY*xF6~@=o{BI|y4i9^X&6Gla01CPxGq;5xIz(bPFG(68mbDz z%2f4}+pbddXp~8iu$)v~(Va3*DW;~cFi}u1(YfzODKyk(4~zTtYjSFem6eqfJh!&H zI}0fE;QM*>NCv>qKp@WlV#vyz{w;h2I7b6Mj90kTC}W#j@-?}@6FM?W@R3PD|Miom z=iOD3zAv}WgR_jxkW#J$KmS-ER<_bEF~4V-X4OW+s_!AB7?aSBRVdPIweWV}(v@_# znE!_1`DX4dguru*W(tl`{WC|+nifL`vUa=Q>kA-+bFG=qO>gUI?saeLU`p6#X=g3~ zVTy`&1Kb5fm!}{nj)^_jUVcWX8I>4)fZEO;xDZRRA z3gy4?qJ@b;K|z7@B6p%cf-zNBMn>eEPKB4PaNm}tsNdH2HoknOpZ;m>_Z7LE?}0>B zeUZncA-YXTxJKuh_&Y%nIq=1Eby2_$^lWyyhRTe=>MnI z6le9~rSo-|G-$A{u=+HhA8IZx55kgQo(Cp2?284%l7#X%&fNKa2nmtM0?nKM=xMw8 zm7d-PWJWktVmtzZlCwvNRvLTgK%wO1WJMT4!c>IKPkJaP$@+UuL0>rE~A@*_kYs#Of3j?H{1IV6<~EznH{#>Ka$OzI7}R=vYp9`SI2k79O63 zwRNdt3}G8rrE8d&vZm&50N4d`slR^xO8KLdURaxVwI=uo;$uj4xs8?HV%G)b$dg0X zlW?S*t6i*`$=^l^7UOmyLK>NHUtVKLA486$dJ)#Dna73rDqFf&iB1jSsx(brenY&7 z%OjpE+M);>DCZiH|Nqy7D`w*p5)u*3j@H z816&h+T&Yg^C&_>vCz^=sj8|93hG_t!HmVxpr;_K_*^(VIwCu!sCkbTI+XVnwfg6< zxOx;7b0T+?fv=f^gKtxMD!2YW6|ve+S+$?9jvAp9^PWXLHsDc zX*kw<=+H$J*U+hqJ{KdbWkI%tZ2?%^5CqaM{aXzu%feZfSZnAg9Huf3P5xt~jD zYUCm9Twh$5ot;d~-xv=7LLLBw-wze4PoIdm?AJD<1!rbv{5^K36i@H|QdS z{EvPIF%HhmA6;G`AzL(?VIsKzuIK*#ezQMXvZ-#B>zlU0?pf<#pNgs~Bu6~r=6JC} zr>WXuv#&b@wE`oD??aHtk3C8tr3(lMu(D3|_V#vnzelh&GrRez+`D@j7Z!5wD_`FUCpLzp@MsGpB|5hohREOJ`M>kEY=3}*Cssct$pjZ zpa0c*@xDV}E^^@B!b4v-1I?F+FfkEGXK}#_i_IxbkBh=i>_P2VT4`{zaky zskav|PMx$q6rdoT0o=1|q`-0O*ZMzwi4Q!UuctKw+YU}s;q_1UlP$?pFytEp2BKMQE!)4UM1meN04xp; z4wbIOsyskunw(T-yT@l0EA#)nR?+byl7URX29RNYG>M@1U1e(OVwP|K5fM>_>4}Fk zg=^s6Y*>d@E0T0jiK%{OP0K0A;r_Fr!>$!***SOQUB^{pEAEgwlS*J_^pG4g!M&QK z4BY0UMqK?czf!#RLYbs!><@V@os;Wa)MLKkM>&)_JA-WZ{rSU7RvFi4@AlmmG8HD54i!2I9iZpwuG07AwrYzz|-T$fi)2>MF8zref zkmoo?8qztnNTlQRVFYRaz;xe&#n*NgBr_(yENcmw{6N+zQy=;!Hp@WmMoZ?mzrbe8 z%E`$oDC{~mjVl7WuzmnX=!X65R{XX^YbTC&wD_L`}+D$ zPEHacF}W?$FbX2Tu^$mw*x4EB=~VP_>s+Cwb$?0qru2gi~<=;-T{wB$1E3Fs;Pi6W4z=R)9;*wbzm_kby3|RNNizfsSgv-%z&n^d{EJH)@KKTI}nE z%!%mJCHAfkjdM=axpDU6Q`08p^9|!1Shn10QHK=ja%1h!PUag{QpoafFfcL-oGyFl zE+U^ZGslJ_1%=G$@?L6a+DcT_M7vmlnFc>Mu!?N`-L?dph#lCfXEOVT8E$8gL(iDB zn}=;wsv=4srkR+e_mgj=47n^#u1SaS$kF0>x~sK%l;XW=?5Mwe(~y5UpO|7{YTY_KF4mLN*Q4|jQIHu3>l}l6GUw5Q zSA@zgj^5i@h5!16k+wcYi~-PltLgYG$vK;Wuk}pX4?RKxf-b7{bmP9C4lXW1v?mL7 zh|M3vdORg1`vrU+?j9ai+I_qh>uv0(2sv%vR)npkWn@atBUYE=C-R}1vwOe$QDn!3-_ z)whq?&^7|9cU>!9F-me-p(^_eu|=4ED=gJ$*xL!&f`qWVSC5sHpYAeUW@Eq?zg8;m ziwFl6anmiASI5e$bF=;G7a@5LnC9DO~JFpgh>&HZ&fDU zoc-QfT;i*;>}*qnUrIN_iyyPr;*xGgWtf{E#sXyE$nBML1jv>NVO?wujEYG-81H)$ zQ+w~RyYt6!UO8+*!LYyQ2}C@#uP?PZAEmUwOg(L_Vv~S0-1FZ~Zd{-RLLd`Q3TNnl z3mMqiuOF@t1K?04?6s%Z$pMpqFSQg#4K0j>irV3{JGwKHy1s!*BB0aZ>*MX+nx4KK zOCfo(+`?%&nMWm;Vq#(S1v0g3o^}+J@m8F-9aaPNYvlc333OMhfu~^PmQ-1OVl9&B+UnC&~e^gLS z&1J3Sl;Rb0$J4L)_`pVZ$_ySC1bIP=@vN0LuUr9{L!y|dnm{4jm#Sj$syj_jdu+Wpo?#Wp{KEt zoC`=-yj9oraSdt}w<)I7xGOEcvnVhYno_%+J@}Y@a-7&c6x%e&APzEue#(_2lg-;3 z5oh*qIw3kqyC4U)`h_GF1vW$f=FyOes2CX$Yq$h!dU_UwR>c}x#RhTDq%VfF zug=dux=0*lo70Ge%^jZvxr;G~n$=#TIJElk>_BdKc|865%9o>FJxV_pf{)zNmkAbE z1n}EjCM+6E)MzSMZg#Pbl?p5no^P;ObbWn!1`umqZ`78ZO(yBH zuRG;k|NSYJnun+=vY~;q3`9TC_O``I^%n3VD>bB~;0ehBKK$c3_&yqP=w9+wR#t5I zdyYZ|U)ZLFrR9A=^DaB|s?B7l&#uBQ3&w@6cX`b(1QtxhtcBg1CDV@v`R@zr28U?+ z1R1P5i<`8#i-Yzwt<9%8Gi&dy2k>KKQ9ilf;T~62=VkG+ep#KI5E~QHOw(u5pc_7p ze2_(HtK40AF8w+iC6=s-uQ^9}UvGb7yAud%!nROr&dI)X_Blm-PiE^=Y^q8a(Z}cmtgNm{V4)o7(GUJ1D4i9|sbV;`LOZrjD9x}@> zO>s$A62oXtya~C)Q_%_M==k1b6@V_Ta3DE|ge~y+_xzI;`w*$sk~O{+GM^|<1t?68 zU7=Y%)?DLn{BL`r<4%*|J*F(7ybJ!nKT+}!+KEdvr)e3S1O)|!3Z5@JiXzMJr=<~% zs1s0&qKWwgw?iHBD&kaWmX(&4ehX+Is`C_)CG`gk1{z>;S!r4_GfAdY;Q5cxMhBIm zv1#e({82UbR1(I zn5-KNdq9FD%pL$U| z1hrR1iTp8=gp;@11S6D(^mkj={6bW)Z2@5>1|uX%%&?0Fwshrg|)!Ui@ z7J!Z+i8hF5A-@+plNkb65p3p};3n%ClaQ_)`T@ibmDav_M%Ag?hUSvhw7}yn)ghG1 zOq=8YiR>9r)}x~xWv+28f#*PThj+QNce5dT?P`a=bJGH!gq1g7A(-~cp%-|9#|3YU zD4e7bdmW33@M*WCMOk%w1eyI>j%ECvc*)41^eNmg0zp~0|*+Ru&S&d3bYqGJj7S7ihtr07=u$~TmRX_=d!UvI2LAk7kaaVd(+ zjFM$I{S5KD{wZodf7nlqk~*(bFRByo@aCY?xnIzbf3%swNjP zG$Kb|UCAjO`j6%x9XkHLDtL0E2#YStGrLtRC0&PxWu~f<)t{+_v;NY0otyrMB!HW~EC*Eh=xX`cn`Q6MSi>r&R=-q|NGgV1$c8PC=2Io4bCP1gR8k=MVFF zyy160%ewHcCb|wWU21dyICy&M^=8p-OcUR^0?1B)NS7=LpeU=T0Ma>POL{-%t;p&9 zH8?o9TBwbs=stm7<1E`Z@gd6zZ8BFtZDvy?`h>?}qo-!cdUWT=YNiaI&`slQ4LPtWGBPqx@@q*S?Xtswv<&27dlew4 zcso1-A0HpbGWZ{dlUdzMD=JV61A9bQ&*>8Y+V8o!xdB4n9rEAP9ugy1t|aKyWXCj%9g;Pp`q-%fBk)Z1(lU()@Or8*+5PL?DNqO z@oS*Io4NUJFfyT$oT@4ohhatYq6#YD01X_D7*(h?4BpbZuY z-LW&lkKAT?xC}D87Sy~Tx9DXL8Pk=H8kA^3lD&$59mUCKsUOEZxKOhS7?WdW`8>U8 zJ^Po$WL(JGHHTR2xqat0j#{C*1BE|*dniex!Q7h$7KP|e)8X~)a9S119o^;C6&L`U zi$jOkEz?Ek?%}r=aFAjikRR=D4>nL1UtS(h*&C|)KzV>&{OaN$$v>7UC}=tuKj=s6 zgK@_qT_*dcBt+8a8{6(4AZPK$`j6YZ!@xgLB8;RpArWrrxxF5o=2BrdfWi?xwiui>+9>Cot^mj_}tvw zvZH6=nNJ0BsaC!TRj{5ufSYBLBSm{VK7s7!=s0Q8Yikyigo@7sv(awY9UK#bw#6JP zHGUsz{%~`Qc;6=tqXeWPHv>@wSN{ zf+2VYE|F8kb0DCjzzAC*a7+4%eY|=eQeKZgzW3@8gVWzr(v@tR5h{_g1j#8%9SAP) z#CPz}3~f(HA<#Y#rG&Q)$%Zn2MFBioofB)oMc=Ai2#n~)dLvchgT`x^nEQB9KBBj- z4pmztb55<)#_Z&Wi{mX*d=K{hhlvpE?!!1l(h>O;^)Y30kWz2q``+^pOSQjsH1tyk zwXq#Pv>?4^8Wk#+#Ha-n+@RX!r_JN<%*^4y`k9%#hZsWr01>j;ZW1_fpZ_Z$bUiT} zK&{@OVBv>88n!`JakSAW*Rt+n|ElMOC`yQn zo0^&mKd)y2kUTX4IS$U0Yf}XO5Ut_K)gEVZa&jUeA%WqrV0f$x@Xi%AxVLw>8mlzy zo&!O@>ZPldetbL17naZuL?^csqEr(TBTTv)3ncM;v~jb2@CVX3eKl*fl8cF6O#tS?oqrX|Xqujy??DjnJC z9oQq9W0$Hcm?ilQ+PnM!MF*DKQxDia`sR`=C5L>j!l6Uw~eqsHhPnYNcy8!2Mj)kxO)VnsCtO>5RGsrF{ z?!H`;K$zsT-5z|wIx8r~8>gJ&9JkAceet>6k=x_VQ4dUqS9gG}5w=LNgAHsPU+p(j zf9~l^3B#lou*)ym>OSd!W6o44N=X`q(a(Od(xW1&FGx?JKK6gnZ*_M7J;+mqKD-O_ zYJ*+`F9$n2c9S7<{09msk_TpYcs=54i-7#Il18tTveQ8GxTE3~|He zr##Ba=w8O7D=b@01FY#=Ma#2YU0ufoT1_vJstt=}55KXLB@LJmU6)bWOXsK1bs!vd zwI(Oy-}JIb(@=nz0y{!w%Z6V z5yQf4#}IUKbW}#`KQA~#;nbt`x`J>I?rQ0D*)OnsA>`@G;)m(PaMp1 z%-k7O)HB1wIjGvi9{apuaek)*r+|o8qwMookdYnrPYSoIM3rV*S2dP6M5;|Mo+XqFAeNls!N$# z=U;_;i~yrF5g49oOkdCAsh!gXMkWslKlr+}zyocrZ(UZwbyfzgl@{?xQ5}c)AZKZ5 zX8dp4p?j>?yk{5r_k#4QCLBn+W>;bx$_&g$+`@%y-2oWx=f~SqAjv#j?VC68$GcOm z4O#t*AtNRxwwnr6g;AOX9L!6bVA;juAfoJQDR_F$wuT-gKy7HIw*Ivmb$VF7*T~sPt2lVq@Hn{{_sve_DEfgce9+t+m9i%HhuNSh8Mp ziq_`19Zuy1V`f`Wnot$CWr9!rAS>8_E2Mm*ic zSLJ<{<5D!Rs5RJ-T#s%I5nWpj4q-7E#&_xtb`O`AK8~o86UEY}x}Kfc?2|*bH998s z35$%*e66^m3IE8oI*tzjDx9#YoQVlM$=e@zl63wgVD;?e?7 z!cU0*x4o{metCa>elC~7mhBM&pL6qn`|CZ%St0%CVm1K+ado&&I=+uL$7g5u{QUe0 z($}lJXbKX=s%GZqxHveYY1}H=ichyFJu1Q^RMh!frz7F0Lt?8l&Mu)mVQ)LL+`H-R zw(Pp-Z=3q>!@SD-Mn=j3Cz6<$xFx|Hrb@>T#6-X~wZA+ZPEJmy{E<(G^SK*oSZ%zY zcRgN}po*LQuPLEazwIlV<^mZsnDe7?PP5~n{>`T5=9{DWsrmV2exEq^FSWI`gM)*P zPENTwIiA-C;_3IlvIH+LFE##X?QiUlSLq{$X=+WelOKHIXuG@WF7z$b_IcURvd zths^a_iw2gQgT*0phSm<*^^~T8p3qi~oovP7Ou@|O zrfLp~xPhnV=MZE4irvHib;&7qu;*~uBXg_OlRL%HTf z@7*f)wSe~h-{7|KVg0-@tjZEMdKu{$Z|%siLUH^OC)7NDLtTMU{+MK6?=hR{(*gFU zlO7o*W#wv@!I{|Mtn~{-`Rqd6PXXDwTNld!Z3JML?)dzqX zuFz{^%g@Wpdpm6$78a&bAZs~O_DznhHp4&nn~r7*UYzId;PVBGLrccUQCj)^!N8ZQ z;-bo}!g)gHYv-f%a-Lf|QzP5Jgt=zZ?e?p!qxd^#FOAgiX2^@qsAMh_p-KJ{gy`2F zG2AwKe;AGa{$dQ6Y29Y$*a4Hbcj|q4e%uA-*?CXV+S-}{JYT)t6@=(5Kr$1Ru-O;+ zxR)DUZ?ni@@(-y*H&8RipO<0}YgQjEM-MEI zJrx_pTlH!iwYDjLA+Hjt4JJ^SyRp=x(PFqqz4&R;O?eXI5)YK&qz1lpl zQ+VB&9B*!JUQSw0f$WarFg7qC>^avO64DPiUaRJbYxwr>Brcpc8)2!UR5#$sCG-0R z!hU(e%H8NqzY5+X+&#-flRATk62mwzH`T7sA5G2Pj#h3k!29~cPQ1K9^YBofI5>whTO&Cbl8y% z)c>7OI`4^orHGznot+;J?It4MR4R#tY*m^iXBZf(I^ECwrld*+b%)wG#B-T#T)i9W8IcWlROdv_P7 z$$jn?e|}(K8AUHol`fqhOm>XjRK(R+0DC7$&p$iaR~DQq26?~5lCL5<)Ri{5r(H}8 zRqswezN5?g@b|B(#+YJJl?Q{Ej=Yp_3?=`QVnD-p@Z!q6~Vkw#frxwfuOS67#vTWAD;4A}`F1c2-4 zk0yzZj&^f!XmUMX0HSxVk%PQEN=TFlGL-cf4)(OG`*D>=X3)zcSlU|xvV51A{$5gU zfA?TJCQ}7IS|MD59mIXOZ8tYF8PFi-`Ca7t>1e>&lL0@ZX*+A8u|2G|bP>%gf6% zn0k0{dykHfqdjD#rj`ys4Q6C!qWa(>1c`>Aysb@VW@e_RgJx1wQxEs|g%7RsEMX); z%mXHl=BO=UnW)DcjmHm^yhxGP3dD_RXz46=BhJ{8rDdvoUcFeFyviH@D+L&0LAKg9G;F0+oWZM7_&pFO`v=kMc>==e9t413MI!mW<~4Ue&@sj0DXuE%1o zB&tBBA3!Bcy3K#*$jQlnH*O}q{e&d-==j)h%fZ1xI{%k$RD#r}s;Vj!6ckO{se+O1 zp``T_2?>eL&Q9x@vL*m}pV|wwB<&&HA64lnllps08+BL=z_a7>a^^qv?noI%A;XF~ zRIy)8rLtq_wVV;K@b`MfKJjo?NS-&xaFp~Yp%Gc{;&&?2IpdvjnZsAS?vR_SLkuax}6){ZJ<>3rda2HcpuH@V9ka4Fv!j zNethh(K){qH`aST-8tYj0rUUk2b5lSS68#kVc(x#GgDJQZlpe8)Pd(6fr_*DboH|W z#s|Pssd{-O5ebRDs_NCtAK%l-a!a{(!_36Q58O$`WCAGimrY#jt_uD;_Y*GB zxNwI9<*?z=fx&T2LBYb$q!9C|i{*i;eR*3~Xq%U5R=2^Sp3#h(m6sC{(D58x*zpz_ znp;9RCSfw9ti0n#m{fkA=-1YdqRwN?2o(|yWS4;HoO`TsZzVD?x8 z=F46948#~K#dXk0jHXzn$&fUZuS^?wF@d)7Afu=l6BN`73`kmfI@w_ovpyh~wzGdg z9M?zlnDQhT7#KuE-UpCkcIz4Mhy6m$)g~wNytmyq^O2NoH<|+1GM(TezH^Ank*c%o=4nf;%OI~`p8!=tUP?!$`1&|2MRFqall zzD}4E8dDT|}h^O(XngULLrN`+i0ea&}RDfbIl5si5o8oO-#I&2sbf%1V2I zwMN_;ue^f7TR1;$yUi#sFPEjV9?$9oR4?md4u(P!%nF{XEGuJhhP?f1{JfvUxGg(- z9WX7_aw%IoI|Zesh(oK@rV;J!0xV(z4R))xlX+j2l+Zoh7wc`n@FRfn2S)k#uV4Nf zXXOiZ)*9|l)uuxQa;fUwx@Ywi=rXBwm#?wxWwlF1Ee?v8#HY0lPSqQi z9x@HSP2igcGDg8);il@YI3zxtgEIrtR0$og+o4ZsIV1vlYO0z>MrL;_xZTaDwh^o1 z58|Iz@L0mSIwBV*MmVGFn)X*`Mn=1Jl&RF{=!ifS@+<1*Zo%6H*!Zc3^jiotUbdc9 z`MC~HZ9n5f=2>S@SgIiX$=;Wl7K9p3p}x9Ay?S4Q<+5X>^EHcp|5lShTpg&W;EE5@ zhr0RFkIU}MlsS}vh=AY>X_W=Vdgui$pLq1!Jh26D0h`|&4H<&`a94ZNQ0Vs4 ztZYIEpR2VGPHgyqs45*QYIGd6!p`>gI^k4rZ;%Uo%h4C@5x`ufa@Y*2J3??p{Qdm_ zSDBEU9A@Y0szJX9WQsx9MN&>DC#R7I;6IiHVP&%Xr>~svnl*Ho&aa&AZKrFu=4w}Q zY)L*B$i(5aZ?IG;7ylaAQg_|4ziln+@^2+)utxl#NKTYg{p$ihg37nvzN;;DA~K-u zOq(ewwhq%SG&}_Etk!<8iM?LUX5*8D+4XNUqhpg_$*K-p@NVo4c-?h>fsR%`F-_dxH6 zF$(4)A|jq|x4!*ot~ez*8Bj)_%Ms4I88uV`m4t+Z&V4yqSv@repw3r1{9bz_aL6C5|kZ$P~sR!xqknZkO zQo6g&e%`PC&R=Kd%sI^HIF7<|@4eQ#uFt;L-aF$VSB(7on4=P>Rc?3w+_R!|Mze;n zW@DvFwUsZ-@}7rk8jo?~Av2-wHWKm542+DlhsPiIJ#cqV6j_lV>g&`@vBAhr=MiVo z;Trt>Oo}1MNT5_=l*frOtfFN7z=K;`w>I8239MB3A%DZ+pxBGWHJo?kS|Gmst)RI) zS14hq2*sp`09HhgZ`(Gltt7(go!8XZVD}qR4mTgWkyDQY*~7C`AFWHYgo{Wdxz?Kz z|5hKN6;pe&SX5$sm#vdJ*?EtnYPZ*wCIt8DorbJ?*S|Z0y1Tor=d0$LJzUJqX^Y(V z=c`fhIbPG!`cgfDApva3AlNwHp79C3vedi0X_d+8ym=>zNXGu=Y};kUi<^lFwTlb| z`Jv4h24I+XxB1TmFzlszty6_+%<5&IM@s3edgYlCv!(VzyyHdU%U(2=A+D*s#!?;m zEqQk~Avb#VlHwiSt5;+>2L5?GJKMuQIAWb5^RCYVF5sZsWoF*Sm$D2k>>y38JZ!ay zTfbpaD4E1{FJce>mtUpWZ}0-TB`t5KIu(7!!9z_VE~r81h%t(k`zTM|6)%zgIkJbe%KRv^ zW3?9iw)(-nUimk{PPkftdsBqG8q3FfeTI*+i~=11Md2+;SoGU&&Ub(3nVXvv@jG|o z}^}Jt~ zcVutR3KJ94nI7v$TxMpJS|@f2-#6%9eFtaf;l4iKI;E}Z1LA0k1M?fwr5BG6TsW52T91H|E0+vA~t7Sn{f;X}HU0NE|gtNKsL-@~!J+E7Bb5|etb?>c05r1GYqh{3=2w2XMnUAE&nVEVsB|2o&AbgADNQcyCtIQ_=G-Ng9%Vm1+ zowd-&Wqb*~+?k^1<#n5@Fvhp*k(wl#3Y?}`C@W)GkK%#7mP_F<2kzVKt+qBHvJeHu z$K@tB=+?8D_N-0m1csl`mxksOc_2)`9QWrRQC<|Z8!kTKr_FI6=pYv_j~XNfZJ5t7Lhc0ezbcnn7>Pn z=YGlTrlO-h57%2kXe7>mMt`oXw1S_}U;*sQ?*c)O3r9ys0gnqS@N`B-lrl2|14!gZ z9?)IXI1K8)c6WDw|28l)Lr@X4TZOHFI~)a=4s6vX#yD%=gKF*+UG4L{eS$LqS17ruPkP zQF_Fpm_VgL2V9ZWTm=<=mxA7B3CFQgJ>j@^1z|PVtY@m^$8oeunzc6fJtRK*g1QE$ z>HID>3pM+e0^YFy2IpyAL17_1C0?KxC{;@NA0#O;NJuhUyl%aL1`MTe0`EZZ^+kZ_ zYqn0WvG34V!@%z+BqYq%dfi?`Z=Y_Be8hF8mp)u>20;+6!=zpJ3D~)#qa%A0Z(jqq}Q8QQ2{DDEN;1^UwoGz#xoM ze)j8_=iQk@0t6zLA;9Z>=W139Fb=={L}JM0U@;`L{8mZ zKYkQvPMNP@J+`kWT-MXn3@v*)Ae+J=65Y{nz1#$C7%IbfDyFCiyix2?zru)P_>(nI zY$*>)3I48oFMB6hk0`@T1)7RkKu=)-ZWWJ_fTqKZ!y0!me(8HCEHDJCkO1Og9YsYy zPyLe3&CQIA3@co2Zteu1^F}qY3Co$1`@7STmoLf+bVwIp8~z#1O5vC~RMKnl9U`%wv8x>Rs3PSka-2lkXRq?;>m@_^Pe#O(zT($v1bOzkLTG#*pFG-{#0=8*ZY( zu1K=seKsaBJUpD@#d$h{UW)d=S%5^n?_#Uh?o+U<3h>PMd^^3OgolgE4dN&`I9N87 ztJ-!&S5L1sC#Q2}CNZ1KYM6qc#I?;&J5tS=7JnF!yMB_Tp&_}z?9!4TH8q^KwzhV7 z(X2KT&;W5M$fpfj?Kbd}k1vY*ZB5_W%e4LM-U6@4|(PBI>urNB+O_ zQlLqRz*wRx;w)+7u3#eS?G_@g$W-af=5$=i#WCSg668M3I9vQ3eG$6L$g$CRAC%S; zP)u4MZjkUn`L@jgJo@|edUUtmVG9l(9$J@KSkQ$UET-s)h|-@U;k1>-5V!Z|$s?no z=~WnoBqzV(nD6a{0EdhKN2k}^;`8XOtE&qt{_j7~{kN%SM6}SuoIo{tfk~~drL|IL zmwi&i7QF`@H3F`1+0K#z0nx|+BwQiaqtAVIR#tnyFz|##L~@xz5d!e2ga}jJzr|kL zKis)}K?mha9RURe1rd?Y(hf~dKdNmVu`}27 zV7L5=I_{0Gw)}LCzeH{M&%9|kI%8|kPvZj}i;h5}Y;0@*4F!jU_&Qfgs9qkfWG;Q2 zFEi+n6TWu@LdEZL(3uDDeI%KEYb(>|0pOM~+bd`2zLDnUW`hyJ>L$@T+ocAl^Bt9# z_Z-f9>dx@_@BG3-Lr+dmfeF#y2a{lxD^9*WQ=0jPl!cYmS4HfmP}@eLINAi%o|HM0 zBz10!Nd(^>pib|?@qFW<9p1(Qf`Y(f0BxQ5p;8D+Xygs8g4ulIVM`b}y5?YpAc5uC z*_nWV0EZmR+2y6go^L#YH1UEOg+!!wlk4%HvFvhOLl_QmrB^%+m6hyauM?PbLXG~g z1`A1*3M~e^68-P>jkuSHR*;X4%T+-0Ww-ZC?poOs$$D)~re;Udll7W~J_QYvX}AKr z`rH6R31;c6k|!-mv8w9PIy;l2ilpm3(CtgpF#~ZMZw(oVaT7*xKb5O!(_z2b!P5wK zJga(agyV!q#tXeJ#;?~#E$l9gh=?$u*LHrmgKjazMGTmpo+dnh;u0Mjt0XV~1g-lZ zO~Qf+Ckz-~A-h#^p2EmT)YP@yld0XNrl!kfkHfVn;Sz0lD@I1fw-04o!Vgz4*l%j$ zG222wwhbIO7oR|^ih?F6^xw#j)mSgkD&gukbV{leBoC~LZaZ=4xAu8nmk1Yu4l+Mw7E;7{NwR!Z4t-nsH*f36Hi;+py=4b;vMY`fAr(y2!xf}*1 z8I|pLf~K`2>e6@|I`sKn9EdYr&U^hGd{*uaC-_w(yIY3@*5@IH%f}~%k_x)XE)EjJ zQK_+t%8Kg48y1|q+e=m164ovHdK#~P_oTQHaC#5oEG{k4mu^9VB3X<}t#bFqveY48 zqfC-z8y_f>Qq<;NB1P}@Yz7*?=and4B0LwYCOmIqRZ*)H|I|c;j?@N_D{nE6D`n&UTY}au%)QJ3iQ)Bwk_rkF0efP0HUgF?f0Ew*Bpe_0C}8 zmvhst4?1Pq36NsN7Zwfg#M8x#R$b?xSDC#Go!DkHk!g%V5~{i0OeQjD zxS7A9Zb(3-K6t3fkln9#D)dUL=CKb+%YK;tlEk_HXUxU^HRW<`z{GEqnY`k?!W-_w20e)zNRc0*q-bb#>@DnWqUtj!rKzE;SLo z(tsBb9_3yp5X*ut^0F0(Lg+u8l-pYyWfwAAjdt2x#|g&Fd=>B~d|1gE9simzK)La# zYu6Lj`##88?T3mk>D>%@oD_Dgx3_qBBzs@9`En$#;|X|J&gIUH4np6x{%> z9zRAG?oo>45evtJQt!Hba{Jsu|Kjj#`uegaaSId6U`|B0E_4efIeK2Y#wPW_Zd{9k z=gszkxo-`L6pAw>1Fz)1PPw|enjKV7S+M`ld?_%GDz3~YeG+Ehv$A?;m|>~}uj7@K zfV1^5C7vsaz1XE>M0#d(;CMHpRqirT?`gAoA08LP%U0lb`bEpXlgi0TvEXA*Bo!s4 zOd3y2b9|o>@Rrt-5QQ)tMvEFlTRg6wCZMRkzCVGC)YvTXee4c*-k+axYz}@5{Hu%4 z5d+3&6(>Cu*}W*z|16kanOfwI;2y=%spP+u)xNiilZr1c)Oe#+3qSL<0^>CHdj$hg z#P3^y90kZC{=$)KlnsTd-QvCVz`z6cWvB+$y|v|I@naY+b7W}f&UA6Azdwux0VXEk zg%!uW+47c<+xy!~prK*jZ$?+w*S~B|Y+U^BUnUwy#LV%aC#wJM6)ugnU8>xL=D?=T-yMttm!Fw^zJ-wkkcsj&d zKIvonj$cga+%s5{u#^=3k;_KP%{sGQojUgXVCgmrXauNI*oa*h`wO?1hnY!9y)|_4 z{UB4r!;#C?t?~o2y{FAJHSs;r3pRPhss7Uo^=vo70yrxvNl98JCN5Ui^4i*&u`z?) z4SQZh!x6f#KYpqj4Jhb6Mn*&cqB}S^00R%fGvedreU3b}WpIyZ8X^AOT=Hy|}nIJ3~f9 zY=7qQ84FZdaWv8hKowzOVI?+Qu{h%s6Js=yk&)_cs+0YLgW=TvKVJTg-j*jWL4K!K z+G@fCmrwY)taJ+G`;Dczh7cBs_=sogeG_mb$kgWRQ!90_5OdYaDAtKXV#fRXI;W?Q z>0{ttq*N8Un#CxSmu}Pwe0PtYK_TFJvzS#67zbb2TL3FL+j#}rf57M1+ur`nURzU> z&;88o4;z<=C&04p-CcgbbWH*84(F?p$hEvSYr$|IE;Zto3~rrPR#pNfEkjywX=?IZ zY4HLK(8e|m4-Zcw?0pA3jl-u;ZA=ijDq<#7oxnE~lu7(9OyY#+R#QKwLNIBf2`n&( ziCgr4KI?=DMu~b zNBY#O&}|36y8!Nmvy%`37U*_0jQQIPw6lRptl;rW+PO;8j%|hCxzdSwIXR>qIXRR@8HtHOR^mp&ijy#kBi$gEfztD)tHQDMds?%I?6#cDNK575(}%Q>qtq z1!f=PDa^(B(b4C|%e}eS_+W&m^ko{Z&XrP7`7@Ogll3TbI+qjrTJyk2bv90%h2bl8Z9(9m|j{Ct`L zMC^CbpX2|)GEk(wac$!wtA$4+CT(&n4hnH)_`CWrNugiW($Z4qz|PJ-B*TNtq@Bd) zL^oY)J^_bgX=K!?;K8yL^Q}YGVCwkzc(&hWuIOYodvI_NaO#(UfPlb2(Vt|~^Yf1% z?4vgNx91KRe2gM%Q)F*$ZuTNo0yi@h5}sg>7G3}HN5ki`3QoJWn3b9prX)ZD0s83x zJSyP;0`lnk>Z+lY`^_iny)I`{*>_g{Y31R8-%2_9iT)@ER`ZO?YG&R5p#qm&e)gqKch zSsD6M7(dv&6A1ecD9E#$4p-SM;m34O6{-QGhC{(OGBj-8I^FF|?>T?^ca@15l5d}= zBa9ROiCK(vU2B#=iw}>U%D(0E?U8E#cIyGwW*qhM2;%mEvDptbIi{itlZ%5TkC_r3 ztGgFQ_Y-o$>AVh`e$SD@!^8DUc4ta$mz%Bugd3a!O=$u?Y_%&4mr*n4)!DOO)4l7x zF%%&4Hz0?YZ*K+?nGaW5Np!sd6*Fqp5b`@Cv-%eo7uVEqF_4G~SWYwixe6lj{3cdHdDQx+zZ2ms~e)2aKJQYte`5Yz6ng_OE{STuS^~)@K}b zwi`gu;v~ZH6x7v&jjwK)TkkN4h>A-~M>1q!nZbl}v$Ok?$ffgPuMQC6atVMT`u_bp zUB`!_;$ooP!~mNT6X7!tMzdhXwd?J>rlzKbhf&|IwE76s(9n>Q{sNUb+KMZsNvqao zb91x2ygV>4@U`L5ub+Fw^2++aJG~Awj-(hbRAZKw9vB>yNtQAB_z`ByWNkQx?0ax; z?|X)920UjQ8>Yf7IpO=xjGHfuvtPJAxf5`BYY#uXV;Fut0(thi| z-O9gYw+s(&*u|2w4<9t7#%n*ZRPK-WO(&?@+a16>s$u3!$4Omq5Pk2Tck#a7`gg%# zPWy+B)T?U8f+5^(g<3A_`A@`4ZXs$(wV9bDTK*d*7987i6Nz}Emc!YWr_X#zd3e-g zX&SMHa5zS^jDy9=xh&VmvL(2!=O-Io4p~2Tw~aAt*L@~Tx_SUUp|7ux#n#9JxN_i` z`1tuL=w}5Q=Z6rx@F^kMRPWr}uJwHGfixB{w_H6#1QK)%;!>=RcSlSAe##63=)Wbo zZ;iz)^i&*rgtxGef^l0W3yyd@rH<>z(;akMxW9?y@}@Jrsd%-0qT(x}K+3FeQ?g>T z%J}zMxxFRFdhw@WjQxyD4{OO~m3Bk!9A8p(gz%EF;+LwjgR%;GXGI?~on|PF$K_Pf zpeH?CW!5g?FU;4aOq*D4DW^=nWtVn8mmNv!h&}PdKtKc91zn8<*|oQ41p@=)$~8%t z>gnkjNys~1X(~q?m6(#k=Y8jT<+x#aEjOZ}rA5%v^05b1y`{O?3m8UVQS#(69~g$N zN2XQu^t_Jlk+8&@kpJ@9Ve|l2hVR8(-*dZ5Hopef!1OSG|5quycJ-P+Ld_AY0sQU@ z4mzC&+~-$ua7;)C_sT?#_;2hI?A8#B^V~S^M698WJP>)GMG+9jTdXG%G3-TP+dog% zc0wPSou41h5N!0gJovAps=|CC&)3)2O0f%+cwSy!pe)=}RL?P1awIXi$Tu{YOej#; zfF?jYXAT$ZV`E~Tu7R4-8WDk_gXE+2w|&}OBA|}DItoQPtUf&jKmGKVW<_OGbxb|P zFfci^?`ix6G0xk11a@7)$zCJ2_0sHY zvFQgKvA^rqk)bqRJBR=&KUg(;-CuR=#rZL1!;l>u*;{QOxhinSUKTZ}$7$_g?l7z) zh{IujC{R|GaY3lBwP$}Yy=X>Z&3C4R@WKm(*-Nv&M_x=&)HWGN+$cjxEj zUP!UgzAfwDsUnGy#76Na8rg3kaHdw(rgkvX&&tj(c6T(j&~|mdTRNjA5_E4N@T--k zzSk)ivHqf+F})m*>!64v=HaHt_sM71Z=zR1O;0_$xL{wv*zpP}yvCPg|0SY02t^Q2p z%Wa%nP8acdUL66K4!xaZMp5DWgUH{JeHkB0F^N6fkRq~3Q0vTD%{nN{Y?vB{7_S54 zpcCUf84<_qQCfRhWi`D-ziQup@zyl-q~9hZn3nc6314M;ZuT*MW=*q`z|G_Zf23{# z3Lx}Gl3J>&IH;)uWt&_9kpi}dZgA-7p%D}mBq#q4OrtSfwqikO zSXiyq9Glz8dSc&}irxLqd7)ZqBoV(VCB~Ga&JM-jiHVAi>zhZlS^QobRUIp-QBGM` zm|vWlY>R}Og{r4bN5oQ1RivXOH~J$p(>FwSamkT_T_G0rAa5E@J0n4voVlNNmq*~+ zJ^#Z`9f9hqs_?%M@4KO-p&0?Wmk@jlTt9#&e#bxZ7?=3I|K1(2 zSg{j`Dbd43;so`qNp;!d>RHm*(hFtMB+^=I%N$239luu~=4h7mfUbiH2+3*Hw&SEY zTYXkb79XD3G`Fid>Q`8S;4jW$))&wb*J=)!+HW!FwGgUUO+v@bvPUi`z0b zHfA~e_3IY_w@qhfCxd2{8NfVX`v3~)R$E}>L}S#Xq@>J)Vh+edbk}yN0UHAYs%}?a z;^LjG!h~Re5=R zD=Pu3S(YJ41~9MCzN_;s`}$(7+5!Z{b)NvCRSV5lY;qs-`vrs^A#u z?F{1I1tMI!O>(Rwy4HZiQ({J>sz=l3EcQ7?WezG`pIwI{A4}Y-kQgFfd2Vppey3vY z%n}Qdk(JHbytudkrBphB!FLII9y7}?;DGQfps!dZyZYTuH}6jdm}fTtr$So{3B@k6 z@_{wAnX916aU1<60-{*RVG}u=q&To&tPX_6lZooe%F3D=)z?_SxGy(2?E=nw1ax6#>pb1*rWyi5RSO&w<%@zb`u>r)B}iWqBS7iMN=iSx4P)YO5k(_BS! z5N}z-AA2GTxhth-(NFm%A-VDMk%J)+TB3-koNQ3sdE=T~5m;&V{KJ^0;7dp5(x^=Cxbx z1Sy36;p*xN_W5(N_dK6|J=uw((@<9*Oy|eFh~i>WFMDE4#wOC$ajGdLD(M7LF?D`A|MyQqj=6+|N#Z32Kc92kE>F(PfMFJs{k9z}2h`>BIEd!4 zw`Sh%?tH88AG*TDw2Mq(ZlZ_n~rh&+> z?Dyw6gPlOhE#C$N3?!z^VklBXL?k36q&GH%Ka7K;7L{~1I5P5`S2UTH|5X;KI_etr zo$c-I!NCZVv9E;b`UD6zBM4<=WYjk`G&FK^b0OG!gz@W;kH~vrJ0cUfOxg&kwbj+t z010YpOnkB-zW~}8R9{~Y=52qzy0E!f5V}j_chSi99*P@FLS z8nE(@9-6j}4w!hrxJb9s0eKH2I%#>WRh&ksgp^{=cv9ZY6-YkFLg+=Eb)lao_W_}`PQ0*;-Yo_=_EP%G7~ zR_kaI2Jz#k+pI8p1GF_;TH`(j&3O+v8pc+M`VarSAr@0qDq*~eD#P5_lAVJV1rXSP zyA`CuhMW1*cUFyFd?qJfb+ekA)((uVl3nbl;-g4!IJLGC*QblM4x2A^dm@Pj`unF# z^@M>CK`cy6HVgY#rlzo&guHGXjfMlWEY0RBHUMuMj4v-Q%YQ0*#c46wA5SlQ^~)bb zV3_!K>xCLg%CAg14R1E+!dpD?^XJcj%&4lW#@`dPFa1y{1Q`fa9?GWM+uK0W3HcmfcM86K z{b#qdwOi{8PP8{xs8_AcQs^uU+0PZ9$NLU6VaLN|FHlnpRf|(t4BGh*oBS>I7V9zJ ziY@ld>yiAe*K%Wu!ou-ATbssMJxkKHrz~5?bYB$13ghDylQ7bU6h@F}W9j%%EXRKn z`7o8*=-I$`cXxAfaq&CveXawEjsGk(o@EbuI{z?1TNYW1E0PWuE33#&{#Hnwjg5`G zGsESdF$y6eAx_R}2m}(^D(WD7yH{DF)7ZW=hrfq`gHvO()BwUD8};ear}FFl9Upjh zVi*Cg18*SS`_w$pTCc&;d^n94Bu>DJScT8cc7DHuNJZh&Pzoo%=M@W=NKyF!2p^C0 zo%q<;we|H*VuY@*UT0$xDEJ(oKT#xSXHyUo(M1)nbw{vsaxMYjcNY2bfs2KbaWQyJ5#|bR@Pt+t+k{xFwXxEbKtwyH@5C~5YVFd6IyG_#v{hR zNp;yQmdQjDPc=4VPtJdk>fkv`1<+f*^w+|&d> zvP6;u3(N85Y`gz-O8=F@PvG9Wx_6Z4a%d9=HlW>#DX=#;Hvl@_$C*+LxLE8}F=g4R zydLhZQ#ctk5}95umgTASmyp%t? zBP7+<-8W>Jj(0`8vLxRtiK^Dbth6LX;EIcIp=axs)9pG z&ZQ?ev&|VTYq5wwT4rLM3o`a3-wM!X_9T30qW<^dG&{Pmv~EQzq9t}`)wI|)hro&b ztv21M`2jLfsrL$X)x#E=vrmg%upE1SQdyNPy!D6R-XmoDUriVM0(v-C4V#V6iwK zkgclszAPqLH+}S1Poy_laKrPz2Gr%(YTk6<@bWZktt}dx(Ekyi#e_xp{c?@Z}{Oh8J^|42B1{``UL?WW}emVXh8{f)-(SSI_*Kg@z zoEaGyWR;XKxu4WK126GMj481&H@6FCN8YqW^`EezUyce1-TE|QI8nIun(iaZo@Y=W zP>pkI)T(v)!oVY=p#1iG`}QsLGKe31{QU56aK!x1MfPlUo2y-6a%nuQ06Req6|kip zNMvqxJzh=Yb-?Ggx!c)!aIgRLTlDea9!PC(-e*shxVSjr2T2!Rv6~3YKrgrf;d^p& z!p6zj?0H=g6@`|dXVA67^sj^|6l7NC*cQ=@ZWc@)984@q&b)z`ek)e~@#IWLOWiJ4 z2|~}pGB!2-!&N^vraoB79PU+2)(4$rBn%t(n=^9cnvC-_u(Ap28{ zfEOU(vN9Xbk<`^)MvC)${|%sujSd+X8=D{b>2;|(Qh?}h=o{ds#06h*@@wGmfKunn zWwwBVMl9$aUB~`6d^5gB(+?J*Y?2Vd8>>C8A>ZV1CicXK7)u3I6scG2O?t2uh%zZT z;^B`vIY!qh(_s&IH?qo>==SBACYO7cDQ10@+;G~+lY4Km@dL2f-Bza0ba10+mg~_k z5#Y24@_#CzEqo46Oazs{9#WivZ`Ce!mFw&Klt5QeQIQ-c37RuMIvJ?`X4{8 z`T0G|>8WUG@phsyseb#?48afH$j?Z8BuI)8 z5{>QtqVeNXYw-v5s2d(09(7I4?`*y6@e){0xI6zlFd`q4>ChJw*T)mJsp%Wjp~Gq0 z6~#rXf5Jwq71(0w+*7E`ur(P|pqZAIMj9h8#!$v7KXS3Zz^v1t91G`N4`TEw3>?c8 zTtPtr%Pl6&hqfexLP7U4JVE!W>};SD1ZadjEKWO<0OUHDV>W@^8AxCRyf#{(l#j!# z8}64UlRRE$_t8qH)`lqj0(v_I8rpyDeL*YG`}o{%7HnIKbs8@Ky?k=Pa=E`bM-#ew zk_x@urb1gsXMb-m`bj@AI(nJs^+{>V&CYE3M`-sjCvdcixeVVG^_zJ`wm1`1maW zMC2pHd=hS(#f~7ft)bMQej_6z=!zoHsnyPqPcGog5kVC74?q4%!*ECg+}{pwTVa;B zJXBhdQBgD6I&i=mm6@v3N(g?~JlPlk<^_7gkva!-`w62~O`Ba$_$&X7?TLJ`4TJGd zN=bZ#+$=2UOx_@%-i0p_07%UWOn$0U_UUPfx2%;C{UawM#|F0)KQ089ufXeZ}<VK>6vM9i09DFY9hz^G-Y6iY43q+KVksj1oQ9-pZ2 zbP{$Raj0uS#2-$?qI?`g4?8<1IjX`mMBqbwnTFSuz)_k|0Tsc{pOI67nHY)^$u3kL|+i)9nN0N6QHAiM)(b~!I%ub#@)>= zt54>be!$Vuk%KKL1s4nJ3|M*KJpQl(y(%D>4^y*RYVqPvJ2%ZS(eVhcxzUSJ1iRD?|c3l)LSIUYVP zf{gz1WtRNwjPE%tYarq5Cg&~8%|o7hcz9gkG*T3<8U6OuB0eU@ zw-)dU03Tyo;OrE*IleR-i;ZV3^4b{|dxxX@$i^J@1gkFQR00pY0V!kz_hG zNdBFF^YBcss-nVTB5w#hE3l6w#Kfre40Lp>5Q(-91}MIU5%Jm^-Cpd+U{rVj!HcMD z&reTJpBug2N$zmrv`3~Y7HERC1;Hc_OHc*lZ@OF9Tarhbvju@4FSD9kpc{{h-&>X+ zqF#?eEL)J7Q%(5FXl8RGzCGJNshtIvFp}2*A8fI=w}(DUN>zEiGvv9I?vlb`Zhx>? zf4Vi2-NWbj=Lvsl8XQ#yuY<|;$wuqliFBj$K1a+ORaG3iy@P{c%`bkVed-|r`6P8W zFaGTr!NsGj^0+$sJ*vdOVKLb|ILII@oS9v-4A{L;t@P&hmRYL?*^-T&9l#||ROcn2 zFesKBn3|e)NNf!zE4h8gmVJ=ZG$9@_sqY&Ys4Or4+1~7NnNc<1^S!h*>NPnpuSO*` zEp0(gPUq0jKH%ASB|aXWZ{>i~_pGI)5Dl^8sIU-!52>$In#xH?NJQS)==|Zd_m+|^ zvvqfGNK91oheKKRy3`Se?z#n2hEG7;fq;aR%56(rBYMpF(P}oF^$*9;-Lbqgag^G1 zjoNp%e~Z_jnd7lS*)dKAts}x2&ofWra!o~`y@y8ba3g{WF2&DQXLF@-oe#I60cy+0 z$QaPtUQ9twj);h;e+pF^uLl6%(enXFxWvNVcReINi#65+&pw+HMno*VC67@o0JK+G zT)Yo73+TSVAOai@Ghfa^XD1~G2Uay7AP`U{jPMnLYHMrj>)kf;yPuJekVs1p9336y z$)cPf+vw#&ji!Z8x(Q~&G(di3)M> zcK_>?h{;Abd%on}bGTg|G0$JRa(@|me%+2`gcGFq-b^^>_*5IWgz*iDL9GlENbW|F zN>HC&2iA^(`NT$izK3b4AB(P3$B*T_SZ4>l3>^G|kqxqh?%e8M0OJ58?Gb!v>-j1X zeCI>z8ci*&HAZ?=+n9V4N5llOiPuDLNj2LJj*pLnP>E9boam}jbA6jEN*u(pi3kZf zY-7;qMwGQ+CR+iB1BFcDcPS|>q-PVy1QntS-^{||$E(`4jg9A7E7g{`>%;~xD*TQs zTc3~yiX1(;?XlF*c$G%X+t$m@&Tg%U%z5Z zzOsx~5fv5v^a*+_&||u*GLyv7XB}#pq%!AyO-G7nAxTMiSYHqjKgdn=3j8gGAm49Y zJxhA!;o(Ny(yYS{-^`({_bv;8joED3*&`cI+wNPN{bf-1%<(H zK5$)q<@B> zhiJ;2t7DR#m6esacxUa3wk5}JzqCs0kc1E)LNHUTC)+Fpt^_t0&&tqo3ekCz6rNRDmt(n#AiC=3s zg@cG_FLj_39Q8tO2Iu70A1)gcU0>9Emqig|QuGr!>(;fI@vi)2C8CCdVJiaX;e@Pt zZ;PY{d;)?fBL0NaUeffbG9t3<5{udOsFcN~$+B82I=s<*Ig(+D+=I!2SZY??MyFlX z>*7@WiyFOF?=%S)PYaB1fmYMbs;8u!X;P5`x^r4X{`w@T!b;Xxe5pxaf3K}&ONMiC zsVMPcqH(6ws%(~a;?r}gxT~{I2&^Xe5I?{8G|dD&v4(~q>(76bnN57l6yFn_G%y6#7Iu)Bk1ix%WGHW$YCmS?3Q!e3%7{%rVEqNm zUuJgIC+wk&gy95S4reG%I0)C&2{)fE2iKo@$c632iH^Jp+7u+va0~kVv@nlkv3@VDvNpasD4;mCpg6T&`R-7hN?o4L+!aQo$@w4r;{9TxP27Lj z5Zz1X0G}iD6|o9ofw=1NFXDjyt{(Tn+Vqa$uE5MFG=AKknazl_sj;6E-}5YG(a(|o zeHH;B=9>@m*E%jV%=<+11F6ivs1`?fDz^E!XgB(+>ICQ|)?QZf8Bt^Bh@0AGc&zqk zy}yOqhs zPx}J%Bvzz`240&eSq@N!es;DI8cG*ya9a{0icL<^)1>5L!*qk2?+M69`XZpER=Gg) z@k3(JcE}e2YnykX50E(`dSd-`W2!{@;Rhlc6U*5jnJdylLGh%)`Tv&D5CXK#Z7e5^ z9rfK}#0wP7?Ixqj<7lmx28XPWZ@p&8@2s@lMKio$!Y;(_)(-Nz7A4s^aDQ&|(tYq0 zhf7-??(n?Ag|0m)(?zCc4n=QUhn@6?Gm1xfrD2Mom!khGgs;~T`CS}nYzzAx+RRky z8ZmB$(=rAo6kjWk+b0)h7E54yalfjD4Ye(UVqF4{rbSjlxd?4DUgO7xRc_OK1bfRO5^nImA{73kvBH|?t} z8uDNjeEt@MXRD|bYQ-6LERagmRFR0Yi03|xLLD1+L0-$ty`osoTo;)rGp!4}ifVf5 z6P#4RH~-FF#O^glbn(%$)@e~JP^49v9?p9{M@4wvb@kQn1nz0g(N#(bg~+7kzx01I z9Wi+B(K@3YR+w8%5glxtF2SktI<5guG+BJ?fyf6tD7#o}{tY_k3$JJY`yU_!h%=F7 zw;+?Y@jpTpSg|`y>A1FEZ9`^M84BU2z;pi_nV;8cJt|RdTIJftf+FagPye4sw^JH= z3HvbrVNz=9<@SW>*`KlhPMbvVbG3w?-cuU^sFhi#p~`lp#d4+u+Ls4~dK7{h{h!a^ z;o%d0jps@O<_iu9>Fe*OV_>L~pD6#>1E^e7P3_VxjGmsJ-{(P~6M8maX#?$Jwnn81 zfK?=9Wa!rLi9Fft>})Yn(TD3T5;3uk8MW8{eaio?ytNjii?4Vtl*t_PzN}59L-k;( z@iIF~Nb19f>71a(p6+hldVAxA8f$LbWlvUYRfyYmgD^lCNb>W{lXY8zY*Z43Oy4&%6Z$VcWD0q04km95Bxa+c z5er;x=Vw0L?3ViZJ)=ZIKzK8pEfK}V$=Rh9*QHB#(B0jgvcbJn40{|pD{yHftMFuP(m3RLSfqnUx$Z>dwa+sL1HBQ&d8;J;sWE?EobU%qlg8$tmi3X z`$#9&x+DDSG6g+wR&C79>m9Z}MBZQs>0a*5!L0ST8wJaL@^TfBU_96h^{}xyNaJ<* zwTgc5?c2BT@bHCN+y6B!ZUvEezwujNH!AY=%;r&VjE|3(e=Q>=C0h0Acl&c?9`MD) zbO53x)AQtoVPKT-!)#mIlWObbruyPyM!N7bD9&&HS0u!np-cYDbflvs1qQaJ1qK7Z zfB!CXW^i?L+uq)`dwBF${&l(c!&6-Ri(;Pbwv)TQedZ+#sVkn4Cr_u>Pzq-Xo%KRZvAu`nN(dk9hbu5E(;=}5 zJ!^oc2PP(BpYKM za_wpX4}T8R($qApZ)zfZ7*bdvp3jpdh`eF*+!{{Ln!&e=3JVL1hSxF1Mlw(Jz7pnzUf4L!8b!}1$7KflN6<`B9&4NX3kbVA$J+;sk|TRJ+apZ+fy zR0IUbwx$65M7|ufA3iKhT2^hH_nUr(Ya_Id+&r61TXND=S69D(r&wf!JD+tP?{iUx z#E^E6vx~$spg0H9+Pa0 zjk^jc53HU^VPeLPj*j+UCnhE?%*~lEr~EMB{iuhA7KK8<1wA9C%=x{Kg@uKu=Ut1} zEh-_;Xr<}M%oqWzh(Nkn5Nb%*s?Wo<>E3L)6STVRL${IbAeC9Gsm%bo+Mg}|2+~@o z(RoUS9|o0Jpf_z{WTdpBPL235^1Rn~iMN0Hm&^UmR3Tw?-GC7rgSwWM!z6_f918wf zINJY3+FQlNv2|a72@)*026va>?hqij1b26L4Q|1LdvFNu?(PsgxLf1yQ+>aC|MN6Y z!y7+}u0D14UVE)NTa?66!#)8DBMIF1p&1Z~fX=|f!UAKoot>SlO%`z&CtL?LfDV%x z+{)_e=C-!FGjk-cx!`%e#n@G-G=VV408%MwYW^+C_J!B6oz8_`$sF9a0jzGbCs?Tj zBk6fgP464lG-at300>;1oxgtly0yK% z^axx8T5ZL?B*4I!Emq7sTB0N;CzdrdBxSLiFG~e%uCMx3e*_kVgP0vJB{*TQTqbWb54iEl+EQp3KHbu!L{Q~>{}fkwgBQ==gaLf0I|^@ z0Y_2Tn$D953=AZ)I_Vl8w*jns-RF+r_GGD+>UpiF$MEIx2KF;J!K~~AO1Mo5iGW8% zU41ACfKM*l6)y0@Lj>ZBn9uvu3KiPBlVvvmU;*Ia;^x+1OfD)cgolCvtOk|r3%zDt z*b7JP=#{-^QToA!d5w&u@p(8}SkPej0oYNhR%vST@bK_* zza2C6hig=Blo16Pv*8X=c?~w!==jlvhOMLKG)cdSBP+FTdlD_CU4#i?k`$dkz zZXPt66_+o}%3g0eFDdCAibU|+ELshL{o7Y#1YgT66^a1jwU2;EpZ^hW00II>PY}3Y zhvGJPo7%TE2u-i)N99L0t1SpyrLK~)D#cLhL$?QGtOkZZGio=*rl(7DW}t{g{mb9` zI>Y+608>`+`?3sdli=e)-Vw46CG3{0$bT2$3wSpHk?Qf0)ocP>Ru8^g?Y}(if(e{n zF+N`2@p^x;jp8}ZrVOwf*#G)`D`;wZzCL@DsA&4c%TV$Hc*c9yc9kH(5Bz6KP7Kli zXyr<~*TVF4+~{u4#OUZKz?WoYWPqny{e>|(G&mR$pDPwX5=)fFd(?lpuYXL;BmfiO z_d|T7q~IPYtF3NtpxVjno0}yg{6(jEfECgFC-mH{1N;Gj-PDvax(jIY*5>AF7(%O` z(|qMnipu}4Lh+^=&iadFfs%%arDTn3K(X-uUy82Mzq0Zih%5jL3;U~i8Sek>Q~oAe zptSV&??fuA#8r~1KY7^S|CGU73~id+yc|mX|3@G7UxeXahK5`7#bEkFx7z@*p3}6A%yxWXr!iTuPFoHEUN?R?25STy7xL=H~~E?jIe&TD35` z5HUn+S5!6BqqT5wAQ|0ZlKtW)6r%0+Ugr>8q`HfE-j;?!thmJ023!B>msAEVpz&MrqBnwt2|d00iZeS-H;6qJ;> zpFU}*s}ti?93BDIBmDBON~JlPH@OAAoV?&m9S{)FDP^W+W=j9L=0An8#qmU5I! zA3rRY!Q}v2WEObS@T;$m%+6l#@`sH2@#D80zvIqOiE>dEiyo~vIGlQTc(Anq{)72H z)4V90+8Q#6h{EGkU5AJ|oz8>lYk#B+7IApE+y$8P@xcL~5%=!yE?CdZmoFRjzl$Pa8fk!Iu;Y6?M-r`Zvm+=Ha7NK`+PvF%p0T&;r~u#9bkWEov(-sgAyRAPtk<@ zEkIPH{FD?NNJz-An$S=Qz?k2z)PnfO z%mG%mX!0jt%CP$0BXWLzUchN@44fx-3pzYL)_PzhorPZmvvqHY<-dPB2s&R5m>0)@ z5Uk;JeSJMOhh+Kcr1y-0S>mzFti%A*`M6N2Hz@U|NS??!q40#u1Mu!p)&@X_PWn=u z33i|H3&3)HIsZYsPtJA?>?IeKL=>)xg@s0FGz1$pGxG&-{oUERMC3W6fg zlV_*2NHLQ(GlfwlC5Noor9H{brD^+&a_wG;gk8AU>`d`+fSiX2e;N%qV+jLG@6nNw z=jZ1%4x5PVWkCNW(y0NF88H3}!A{WmOjcrIaP}MalV%V(V6*)L1FC#U&u@A^ghSB6 z$pJ;h;H6Gaz6TI00M*1Gr%Jg==7t1JK;?a5&YPO7TMWyVKR)(&pyo z^mM$-ZqIxfSy}O~2WkLeYHW1dHTbZIpxNc9Klww`R)KUy#G*?K2( zcw=3i7z;mtyIN^gJ&ZFhUag`Dznu8^p+EFI9*7Cl($Zp${GA=JbVy&*OF4cQv z?7(8D#_a9ybK0%7T_(D?!m5*pFRZMrR8@VZC$15v>g*|fvip$1aIdm8!)ZK(lHq>v zr4Bkz(_ zAVwfQ`F24B$kYGwdwzLIv^?)N%fqlwslrZ*W0=szIN99n=E?T45c~3FOZj_?!|aTm z*6)%fL34A5diWaYo{0e${I&cSl(z`!27I}mmoYNG6?b>Kx!(2Y-rfBEy|g79&BbpY zdnRO=M&uak95@n(F}O>D8*EMS6X#bW9X?RD5Wc_l{LnMIh+pgxScnut*+P5ca@@MV zZJQbKiKRU%SX3N|j}SY1?Q~l-h{v10>-Xg8$K}RH#~3$JdV(laR(lU^1g~u^Dk2_t z&0H1;7Z&1V8W!}ceiJ1d0{{B>aG@bH0<8?a+jG;e9woCY6H~*ZU)A*RJUpQmq1`Up zTzCdb41UTN$7_E@ejtV28W-MrRjML+Z#J0)PMa<&qSDs5y6em0_u9I1Ia_O&S;3}P z|Lm~U2YC8cGCMW(!B{G*R)e+wyr|DP-KcMR9G-8{+v9O15R;l(TClLOjmD>PL`fc{ zq@)(11#dsUGcz+IBFY||hY{QX*aV;$zJ#ODgCJ%+{iS_eAg(#CgRc&hreKJ?!3jZ; z*i1d4#Oj@IuTKDr+>}=+5PW-f>I*|v?{L`u0D{S^1>)e;l!B;eFE{}@G*ooSKxYVX z7=^yQJRVDs!*{07YwBx8{JFH3*OSlFEy-Rgt66knB4z^+aImOot?~Ht%FCmi(9C#w zJf?HSfi;U}8&z`H{N-y{w}0IHI7lLdkBNfvy^jm(^C}NEad(jeqj2>+5d-++cbWn^9Y+>w7$)L_CInQXHHB8iF)JNDWZ>;F^M=Ay|S| zs|7}Uq!gZ9GH*!z8Q|ZoT6(`nOzy+GKtFuy=kW}P*e>5{dRUHjJqil6GbZ7wWqrQS z=9CyadZ!k$E3qoWeX)O{i!QnX35vFD$w6YrEP&qi_|#7j3e(j*`ASe>m+$1>8|4~^ z2jgO^um9Wa&M=^)-aoVTQ}WZnifYM^F)Yk@9fyUk=kWF( z&0m34_+=6qPRH{M*O+23Ay7o(oVmTc&IjMugg85%PFoLb+?9-t$suU*1%@d8fR<{_ z009Y7n0MY@qn@(jBFdfexZ3N6p0=Y?Efd;++2(Tv;fVaY*d7#xDNv%3I|Slh*Wt8y zm}}McP#oF#_&5=7Mj+zlKy?@DCr(@Z$4^P&wu2UWjh;A&gA$hjPV%~)itQR~9w6fI zJ0&3_Xnde9@A0Hj4nAsra|&D`a#xZ$aXT z!RF92m=sQk(X0JQH^AE4+VES+m|IZwU=O6cxi|8n0_$ClRWts;qD^G-CTC?upJ6F0 zE1UCQ3P0=N6`m{8elu+dDSvtr&U6ujI5U-Ns4xxLBQ-+s@uq>V^VseG`8bWK_Fu6aPVxMqL@&eI? zqnahGDFbJ5T2xC-;lWyo=3CnGmowe4-qP~)PiP6rUkeO9?6D_-NU0cK`X;tlh?%pJ zZ~z*=l0!=q~H$dE4$JDKjea#Z3y|2(y$Jw^5YL5?wUJREdqqV> zCXb8fpAG$+>n^2(07>$i`g-v_Q&ZF1+uP5tobx2?0j69qZ750Mkp2SNb>^s^?<}Ef zg?D-@lntYVh`S)Jsd;#C=c8@ymTC%QC`I*z8Doe9qZ2?pKC+z0Q9XsV`ewv(g3hy4 z+1|MYA0F3!2}54P@n8)H1O#9)YQy$R;9iqYyrGhP*~sX7(BRiHEw%nbKEwzi=jP_t zm$cY!`$7UsWadrib#nxr{Uwyx**|kTK2n}dGiW~NK~+^XCnx7Vb}K>Dt-BoHc~g+M zp#v;pZQ_ww0|NsfdO*p*LP(2>idy%Z;d>dpZx6a@v|9&D76}4O^8kDp7#Ju>^)9#{ z${2)`#sp7WIGfusd>2cV3`Seu@N+7Nn@Y_TeJ!p}W?6ZAf;}N7iJt~+R=m@B55wx} zn=;;XX!eUwbofmb`b$3B(McFh62s6@#ea9T-FFSvYE1!v4mUL%Fe6=V3>wD7q-06Q z8Th#Gl`ox6aH}upt`Rc1)HKwS(o8KLEPu=6qP&c>jKlvp;PWEc_3T=F`&hZKr7ujP zmM^wX+&ECKJ!bvmql|Otgmow`hH6)MV*2k$Q=o%e7c$p#r`N&HG2CBdaJq=f{;(`JblWfzOwQpHM3xk89;}!j5%gf8@ z3>#`IJv|T_n{J#83=GW74ztCIAhb+%`T@qJr6pHaS9gTN+4*_X5S16gw+PhK{(gI= zXUAG&d$Q+PDsuAK^3rss4}Vj#vQ7a~;<*RN4e*=Gp&|WK*zv96o5%vFSX%3T#&fxG zXvL<5WgfY`!(Vf-`lqX{;Do<;D)}vpqrdfZa*EE<8Ms`eypbD`^kO^35-}TNeHJ0{ zC)M$CpTG6PyzeF#KtTB#m&SvQT}|>7YvldQes80$d)R5KFB}G$P$rEHjv2WDkd!%5 zlHn754>85SI{_Bm=$e zF%~mBtZPuqLI<}ONfseU9?2m79dkN2C@JdbV7F0O=)!YyNG9^MI?H#&67 zFJV(9M3vIqejCCdCDUPyu{txzDPjr|wK~3BV#tABDo_ev>C?;}d|J=OzV|)_PrZI3 z*$vry%xU-5yuO%bo$bc8$q`bb4D3)NOpM-5D7G#Osvk9hC(J$u*^Fb`qm$LvpdqlJ zT?xYKhjUXhJucB7Rw{_lL*=3 zu2Qid|AU2aGT=xJcI#QPchvxov6w0BE|`Keo5=7t`X&`K;cr?^cbfKzaO0Okc4Orr zEuScPz>zSg!DL!GE-H2Ii2{F_KSDGfCnFmh8w(3s5%}Dy<6#dkB{w=+4xs7mtdmvJ zLo(~ZA~~k&(=(w)n?c5I-6D*~8TC%HP?em<&d5!Z;?b6NYL!=iG$Kg@I zgr9>SA=GDY7kaJDSIovOIzsXCd+M|1iFmtg8Py58V#| z{|`jDdJf3hY85#Yf`V!-;r^O)7(&P51%@tVA{}IEDitkh9VHFz$>IuXm3|VPac?P> z+lCOG*su@uvNsF3!YF;xfQQrRyE8H}HbtyAI~AFXCPi!!;^BE+?hFHw(#ff5d97^Q zPD~8yshyXBp}eXpRwvP+eqw0o{OZa`&0ISqN2BU3B_-vJ{^K7!9Gos%K;WmKSKak= zX1^li8X|&ELBN@N{-#y`sd(i;U$hAZ;owjY$T#^ z`)y;5mR+2kt4J2{u?C*a1j@_^%+`u+vf4sFK5?+H)Bpj2zCj|3i(939`S;ne@J%jG zu@;s>_5OmD$W5-mWv%q9uBQMApV2^}u%>5P8~lUn%$Mh4iXlmQx)G*cHoZW-3_Vp; z_#`Pb(m%CImwKjGZH|)ctajF|IjKlcJe+w7lyNll-zv=VbuQ@X)xKGaAj;lmtZ7jH ztWG3!7UV7O>RFSdWRQ;l6&)OyfeoKPsxdb=52FL8zyhLZxlE(Budk0xBAUqtEN~o*ja3Bt#6A<`3T zwb7yru+B&xF-j&qCQCdUfBkp7JHZIpC@WvY#Yy%Td_s#I4OaE5Fu4d7FOdj%h%bfP z+MV_%0687Bv^b#lkQm}(+Ul_V1w13kDI_Fh%YC05&x_{w0ubonJ~QigeSZVh_dhOg zumy+`K6SRXKFyV?#UXobd~5DXcR{Sjg=$^p$7{dZzY<<;FUOaKcZ^^>^}%s`mRjz< zvC)=ciN6G`j5+>6pfEhVG1|Oi8aRip#C0V0;}^AoR9^iN#1}3`sWST#A#gdVk)KQV z{5B!Mf|%1)S#Uh(*CS`!F=L}`gt3=Br&hqC5sK;Wk=oyytOMJMs@Lg~nO_W+>yq+A z1g*bn>m$R}$%n^jMV1c_EIO8+l;ObVINw^l=Y`b$R8tT^W_PEn;5LI-lY$PZjSUT_ zz%Tm~nV5c`_P^Kr0eY|LC^3{Pf;Wvr7McdP9AExpL9OO0(0_P&zIyxa)W5c-~Lu z)hdJD-O=R9Y(c;IG7VV&DJ2DbD3!b~QAhLTcpt3+s0TNW^9re#&lVUezjbkvsMlY| z;rC?A$M?eS!3+v0C@28WkdL-BHVz_2zo4?T0ALH29(CaU>50?kFUDw(vOy@g(4e(5 z8lO@wqc|_GcVORRn!v3;l*Px#2X)4FwFUEA)T5bC>o@c4K8`K@k4Q)Z;H;Xex3Ha! z`yRRAa<6`Z55rYv6Iw!uW63T)hcyumK$Cs8w+ZPqSH9- zvJw)mN0~a!t*ijoDgbaVVzQ48)ukg%3$m4vh_l&;(z1UaCIVq3A1yVpb{Lz^^QqC5 zfd=D~*3oiv8iBTx%W=RAM@8X;!2A-;gVJ#6z!-}~vm5$g)8Jt)TD8KlV5Ny=#AKDD z{u_J46dqSSdv4^UR>hw`sGb~kq1_1V&?>dZPuR7fXYXSKuh_^ zNvZG&#}YFew2R^jeP1342??97Nnxa=rO89|H*vv9(X|pUw2AN6MdkrR zin)$u1%iff;D_#o3slATuQhV0v^!EvFqqH}fj*8SLyzdSx__-HMKS@brJ$n1!^Q3S zdNJ1%jF37*h82Kzv2zsDiSjiJ+Y;|4L{${2s`T%jqg_LuA;fJs0TUS+WP13oesH~5 zkJqi-O)9IIAxy6T=N)&dZv~)VE?EPefLnJOoEOX7z<}vcED2Xr6V#$S%-*vssOeXzIj*bPW> z$X?(!9K1WI6|g({aXmdf>wa$piP->s-P!4*&GzR<@dvk!pj0rD85sm*jVov|AOIrc zJ&G*}`aA;CD5F-x$I>q$T%q2t5Bs1@1<7XL7tf4ZF6-$dLq0_xU zzx}{6FOD3g&FHXbY?0;hf@8ZkjO#1SS?#W0xMGRC`B5ic@H4#gb|djBi|&By*OSjQ z#DjDnW~#Y4e2APKue)V>#3|#ErOUH*;p>_p9R*Q;gDXcrKDlVxqsz7}fJOKvy{5i6G9)PMP+VZ(P<^s&E)yuQAENdXU( zXIO;9_{-3xXMlT?X4dtjxcIBRaA6$2E#;X;dv(_b%Y*&>!Bu**djFUzDM0+Kz0ISQ z&V6Mca4c<+ubEh(?^7CkezAUC?jqP`I@;__$V}g(u@-OxvHrR*X6xfKhp=7d?)~!h zbm$vjDfQ`<_Skpt#hHDtTM)O`E&a;s`uZC{g2F_=RZD#%1Q9u=2pfxr?TX*D8X;VgNHPS9y;ddEwxi!LS?<$0cs^+2_PlBQR`P(t z_3r7rP1OD`;Dpg;DAu z(e5mTE+C#)4zUZ$j3eHhx{FzKXhQcVN`7Z0vQ$F18H?)bwol6H)mK=`ihnEkbIYQu z0$Is2ZlYzu@J-OsN>Mu#{@cxHht|&d!|-q zi)a74b2X%pVdGtfr&;BvN*7cpVHIBT)45?f-A=u%%Q;q$C&O9Q-vuO=6qSr$Dp_7W z9KWE@J~Vgb1HppbXcKUb%Hf9&fFF=y>iL6*zR=~`0s7o~6zznFx& zxUMeY9o;8k6e6yFP9e|mhw2&|InygVMOP;$W2lHI_Uh?lpO~Yf5a0sM=#=10DjD@B zZC@K!2HhlBg73}|= z-YOumaG&2aH8lmoAloihk+8AVDdFxSp@^Y~Mg3bpzeWdofTDf^=MmbiwYi*CPTk(#0+Dx{Fk69~m)Gm&Xbzy= zGxPJl6xVpLuE-+oOBNFs&)9(%#8XDRcXn8?X z_z@%c>JC7ur>AFSJ+W}{zopg}x~Q;F_Oh(3wJR$tfBq=B;ysCn&&|!r%F1TbfoF6w zdA5H~OT$(5ht^YD2^QImyo9p1GBPrXlf2gCY*the7460LiJUVKRhV5sbE{Sd0WhG8 z0_-!=f6M#c-_qR7ZZ&VcHZr+h_v?ByrN zt*tFaot8R}t39w`Bqk;%B-r0=ywzUajjm6ghXi};BH5*$v<}tcvXa0iZuFwLm8fWv zN-RMu+RV<)Iq1ZV*Zy1RMhuROSNXm;=*>w z*^`q)lE|GDh0AGkz!s>`%l?omuQQ{?<&%RT+c7pXBjR&MS)f-|RyIjUBo0d(f+0WU zd$8A}p`o$k;KRy+6{{^SmL!c+@KHsezZW)%5F7|NT-rx4LF4cYP3G`V(GhK{*jad8T^+yx@$m3O`@qqB3|M`UqoJW8 z;PQO@{QP`;_uc|euH6C9OG`_KLByv0L$w*6&0urHj z!)ONs0|N>Qia0ef5mVk)JmnKEZrabEhNh-iI5=T04r)GSTc@k7>fpJJ&BgB`-VUbo zsN}PV>Cn;9v9Xf@q3R1OxG-W|Q&m+J76z-Qr#JhvCh|(RI}3%7Fj`@1bQIwUj`Iy5 zgRpD*bCrab4#c7=G%ViDCWgpF%!Yjc&5O$0M*OO4CB(36yqB1iK(OS&$VkiG|YFr8drmG zqbuBZ32sv+kil#;kiu-#^l%@BV>;lU=pUY9@jl1m) z-rJ7PC$P zRT6yS>r>g(wAkAV*-gGjCSc}vT8&p!SlH6qD(HUxZn|)*rApagFghYagatGQ3%3}_ z>T!vrxk!Z?MjG975x_sy*BA7{5xhlGX2PxSKrUJ~>Z|%28n&K|5g!5A6fz=tJZlrP ztoT&3ZR2Kx;nPtnJK+)JGd?`@tj|V=R1CcAKR!lfvilyM37#DZ9{&S2~gw1%ORtOINvT?(fo)$MgL~)0PNbqE5!0n5ik{Sd-ag7OUBW%-645 z6@?xi9$;EDQz)Cs<8nSdIyyQ#o8YB*{BX6OL(|dG;o#sf<(wbXq3;_{DN8JKZk^R8 zIyzUXrlYO>i|IV&6RRolC)No-w+eL_hyw`ecDf=g*_E+8GlQPfQ=(i%DUgsC1lkKK+`xo^5 z##y=re!RFI&6d=ES*d;qe(1ktoJZ7}Luf}&#Yx9qVHlc{Wa1^0Cr4Z6)01Wx!e=%%&_X5A7@FEO@_tJ| z`RQUcnf&R}icAwikkLw$!%BwLLPd_;uMJRW7yt=?N5DC0>*%CPB#m-=T;>B-=p1?{ zvl91l=Dnl3Dl|0o(6}~|2s9MrP?AsZ7y&*$JOTnL?_8x`r^#s2Pt^$v))zZ@H-7p(Z3+@9PTq3_wzZ(Kn3X1?R?&u9OfOOhhfkGE8NLC`Ogx|%y zxB^az@)&}MCjj;YvOWS{s@$b=t;V$$7psv38dwERo4<9|ODuhLVQSTl*}QH@ahSK? zK-IvF?X4|fZFxmSp+*u*!H>w~XmCi2zw^jY`mBVvbjiAcgP20^NLCShRGt2mNfzL` zU+xT7cE0ksg47q$(A{8pj&mwfKmg8E6{ND7)gVFk_xGdZ0q#Io2#6_u&ui*gGr)qv z;_{VfNMT&n(1L|SOifKuJY_m}R{jF=0`N6(`baXOJEr!0e}6Edt+LiK(lWHwzKt<>JXypxQi?*d@weXZr>X8FyNnvNIQT}K0Q2iQ?<3XGspAOZ|>{_^-WoU zXi*-;Be4u}wHj?BypV2(0h?YZQ8qi@=)Q+tmB0bFJE~O{1F#R}06<`Uo$l`L#Z*)t z!%#8GLJH@-NQ!M*fVCX# zzj5(MEDuShSFh><80*eZob^V(JNI}x=RVfp1MWzT`-NfgpQ#*CkIUsG`(zNvntZ#4 zmWHOA-C{I}L8sOAWGV-;k-k;F*Rgl^NYHLMgWn6wcO8%@^750a<(j|;oejm2*|YHT zr7y0K(;ZtAzv__gcunS*lCY_jPb;C3mb z@L}LtNDlebrw{e-+#7_b517kz20Iv#PPm>vlk@S=zfL4h4M|j!37jGFCQB9gMzurX z20X9y6v?A861FnCiy)`AJWA+YGk>19n`=h>lfmuoa(iDOBHP`AsN7E)Uej0-@-sEn z15jmYOsDt&z5}Q{!q`vekMN1}Gp=wOGx@35ICL(!dk!`>_Vez*MquV_sagyoQ*Zu$ zUpPAWWp0KkKwCgbZIa|Xey`qu={P$(tK+I8jhO;b9o%i`3$(#1GRg;gD}Yq?yBt%< ze=svKc>yd9AXkrj>2`}5J3Bi!#|wazUR_>(()k9Z82_CDJmN!GAHb#NZsV#z3Q9^+ zXEt{B;A{QLbUu%os3=6~@7w?ow0qq#%2(_$`Ca3W(o*WHd5#T_kL$EJ(}-w*tkWEW zdi-GML%U%mFz!$iL)CE&D9VLXKsAw^Z$c=H*KdJvs6W#*Yir{0-Ss ztqsqTl^BeRfF5ZQoB1jFY(s~Fa+Bj)Qq-txdFg8nF8uftCk3+Ky=wEiIq_D{}YUknK~coi7{;lQ2%kZd6z z5C}xP3u+WE+*od~K|n$>O-468Q7)7@oXY8KT=yL<0FYv})9+1!$fxfQK#%lG)_tFB zfFM2gSA&HM3*!6GkYvNa^Ye3YF}1X`w5{aG@4mjiu33lIL`iZG_B-wUy83#na1zwY z0kdqM+l_9>AbSf73u`-OX6A0Tpbp>x(#t9-p|v_76%8^T9UWQf>wk9I8v{rUIQ`l@ zwEGn5X}!^o=qZf<0pPxiGqD0>WZ;pvT+{+@&raSh04oA){Q7L2Xa-iHS{JIN^Z63N zb^AT!hx6YnqiQ;{v`B{a+g^;gIES|cb5a6aF|$LnLedflGa|B4Hl=UOY=6#}%gU1) zOf9O1M%TZQ{~XHy6DCouXi$9q3tS{uMgQ3BOE!e93Q!HWaN;ga#K3@ru~$I|wp51F>-qo{e|lx5O}E{1gz+yg zR5~2;{LIYg$jF~Re|j21Lqj=jR~DzH?3ZdxfOzTic-`#t#PPI>HTS2@<4QW25pc}N zR+kIFUVBqLuMZqSBHIcR1XBuY760{sblx!4@O3-2q?(h;A1khyoHXT#VG6 z^;R9^xTy6bDhfO^i>Z476~=b0?b8!7gtS;V;0_v{zAwOXf<#DIz`be#SRuJ)WMrgw znzaNFP_0be;b4jcq(4194X8v3L)#NB#f>o(R8;>=lu$TFIa%5InwsBwodSQ1bVNgu z5aHps=F*l=F5wsE7C2bu1nb`%z&!;Pura@%q8R~Dq zczgy|Y8BdW>_Jmro}Q=coqF_GwS6h!omKk)=u%R|5DA9J_wFrJ=#1XTF3is2f7X8| zmVww1RUjfN>QEB0F1nt&LN?jeB^2fbE*}*xw&7ARHYWdJwpe8#`OYEqb7Dfm&i=k= zzrG?$VS~+b^#7*!h&`^hx{--;#aC7`;ZFdr`RV2^OQ;|xWNvOwaTq*93IUtQhMms&ZmAFESnL+?)Lj!nud5zBcb>Vpjq1ycZZNn&22<0tt(()uL%>X|(krzdwO|mBQX2OujE4-lG6%RW2 zGx#jK+WjV(P2uvQMqA}_J3EXI3V0?hQRe|fA2MnOBL9czW7pS%o0oKX+*DlCf?a2V zDO@yptQ%UDkFykcSuA!@KYqs+n@X-&!8Z*c{csuxARosinC&u=o&uH+5Cj z5Ds_%Wcf)gWhaAv>$fU!TRNgFR#o!~7B74jjvS1_8(gF9if?3h{Q-!OzVUIy3zAkc zB(D%B332gi9i%4?X>d1jaGG+{0GqY%2-L2YQ;AuaM%XO+@G15Y9%)}H?d~*IDn{Y;hzL>Ws5793Zu5G;G=xZ=2>*(lM6gImLT!0?bz**4i8N`;c*@D0 zSB)t&C&v+(O(RgcM60%OgwZrT)7KaidO~#bHq`wV}*=Xx!)X*!wBG ziyN(0o+ow1DRC4)nL}ih|M>AE1exe{rPA`4{X_b^nA}{Y0$v296?M;=YH!e*&V*C^ z!_!kOn}UrW_awu+BP?5}V5N-Vc!c(?v9Gg<%Iu}s1!^u;7Gbod6P2r4? zHJL`*Q~O9jmiSF}GQRc6X&;>(zNAiDS`FuknjPe+oi~vsvjPp_dmXs*3^J5?E4_re z=>DfVGp7d!53dp6&wwZcLJ=!7b4lX-M7M3bbha#AGn<%8FZQ|9A9!WM8kw?dlRu9B zv2W*&^jmBaOwuI@?e>9cDOzc9De0N2zjA0kH?p(UvSXN6xnW@fyTGuTj52r`ejDFIF2#oRS^V=j0BN zvv|2V72o!b$}c0rp(ILO)<(wX)jEG}hwoiypb{*tYpd1@BxRM$19 zH%v%JlmRh^_{PH3)>N^U&_%ZFM&7nL%5U6rz92C8|9y$fvRmV*BWfe+v}zUFvqe-Sw9NrqzEhQvk6rhLqXOGiP%>hq&WTm+^cJlz{l8bxmvK_Y zj2)_tlVV~kvTTiZix$<{I6a7D#_Kx3RJ#yg^12DLvGWcdnjx zh-U3}TVu?cH@_n*PPMu0d0^VUKZCuVeG$rw1g(bDib@6J-}myC?{@hRbGoLNSwNmE zOFDNl^f&wvCqMIQ^MAsx&$G_5G|Wy)F*%po9dB4L#wYF z($8KU&~1De?mgF;(gfS^y{`Juc{#79uXuF1ad2HYrKBaO8A}g*m8VPgKFPV17iWLW zY5DhTA%`s=Ewv!;BG$qtNE6qj-!Kce(!_3kAd^3e;reT2AFy`=44_=I{BNqXSg;R> zyq-;SjdUYS-}Xx_EG<%&aP>u^isozV+C3R`s+t1mO(Cu72+C+LPH-P691>2EUeDfo|N$q%R zbuU*>##7O@xL0Yozo4>Q9c1 zjPSS|Vf?NGqzDFVJ3pWgBuq{BYD`8C_J1`>NJ}3NMB&|NEqr>Ga>IWbVT`s63~aB< zrzc}hw){yP#tA;q()Ndfc<-zXY>CIn$ViY#PD2C7tCEi@LQ^c>u%fH5C1guWi?_f! zKwg)iODJI3*y`%oqkbUHKIEs{K{JX{^izoBgw!C2fN>OfWQtK$*~BHBt}fjzgWyxb z-Q8W4__Oa3r*M`k)pk(;vp`J0-e>uiXlZFlNV*3H2Uk{} zbrQ-4{%ekvW4yJ!EuY2L?0TYkPzG+l+y1gM91jTzIX6qI`zbvm10NfE0ia`++D694 zJKz>T$MfZc{GLB>n0xyAb}+HUJXTg$6;r_Dm=4+=Q1~@8G={*lCG3b#|JRf7E>}GA zzry^45E<31)C4spbyMStBCFzl#69&^fDFqT8s=tYWhEzLlfMFw6b6|vgVk)}`f!G1 zgif_gz0=pn)z#Id-1hDD1t8d~0C$e_09dx(adYHcsha2txFpbesiqOtt*oW4PK6w; z-S81S=wr4-IZq-62^N-`j&6H6JIimozLfgkVt;BPc9;49&=%fc&m^cR9Y4N*zjQR8 zWtnGhb&wOu{FrhUZFZ|eGxjn+pPc$B9G&X=<|b!48~E4`pC`F2J~)R?_Y2ZK9eS+Z zy#gihA}G*?0oDje5>^%#aB1<;QG^UUJUpQ5;F8(d*1oSP z=gYM^Umr^=D=RZISEutN^cuZ3e^N+O%BBTDRBF_k9?h1BB?8hQxpa4TvvhaY_wK#d&p#+CyL;Z5 zC(fBOvq{AH%|C`gz5MC8eoaG!K;y#|-oRq574Vn@!ya#rmmo(1S9h)(ox!n>6>Tqs zbq|1xI9_Wb+mDQh$Y~VxkcM>Rmi4i|xV#kk$zj}!90mfb17_&)Vu%HU^k|``Z*8s3 zQrU2$pLhaF4dpSk!$}KSHJi_A2hv}uRI>);cIxTwZoSfkQ}y;joWS$;1SMDM^ZAsL z$VjFDUL2y5@6^ijvJN@=)$tM@5-O=6K1bXuxF<5M3&@+s>nvd(PvGSwB_TDe-}9}$ zeIqCTR9#ugy5%p651HS5zKMd2ynbFp94GxNKR>Vmvcs+cLm__6om&0hHUM9sTm4_} z^uE}-1}OQbN4IhdZE2=|R^3d*iOgD-9adyQI5=5xk>C8;$v&JAQ7}77^U%APr*b|p zQg26$VmC2Y?=jXSx#PKeyWH28@wYR4$W!|Lqz^;63C*?-XMNWc{0F5IYEK^5Z>vMJ zU`3Vg%#^m`pP3XkIJ6^=` z^~+P4L4Pd8T!ns{{RY%KbbS1<{e*-BTqZ3>&Fb!+9@Cqn1r3;JREKO1^XR&|n-CNt zVq)Tx<;K9&FmvITJ7XrMvI85gf{$Rmnvdw-kWIK@dGQiGJ-yofzRcu6v1h=_Cq}De zd`8ct+PS&DhP31|vLPRQ-3TQEyq^Pq`@kW9AQKZ4ylh<7^RaTO8X93KF91HGkqK`N zkbVDDy}Z24ZL{d^?ruMTzF>^Q#>U1w0RscGcX-Hvi%d)I9a&pLMo9pyBVY)63DY?( zz`MJ*8=P2*+vcMg6Vub(1)=~m2seDBp{5QB4mO-~adYcqD=8^?e2EOG^}86L1wt^Q z{6&Yh>yqMW;lj~@JTj_s&}5Yy2K)?sey4WHCYYKU(oczgqGh@Y$$q}=ldE()hJ!PKxk#wdqIFzT zD>gPZg^HO?#(mLKg-W5Jp~k2%*o!EA5&;zFg7>-}_#9jKvS}$P^SNU1bRTx8qU!5; z6h|}pT||vC(wv>0Tc0lEyl3*I(2AT)OsEyHS-t9*b+q*-^H&B25c6VcglPWwrY9r> zG3;M#4^eb9=m1aM3BZ00B=OX3oeDi69v&WfAV$crylzi2IV~{D3GJPAnAz6G$I;)y zSw8|;VegjpH29tMOL3Rbn5NxUFp^-3(#Dv`?-;6 z@?j>lo#d4_xErx3CJ&WT7&53{9-)vUXeTV)p?W=p?{z2L4qb1vA(O|B^oZCT-q_ff zxQKNhAdPBH`>O)EWH5}nj?ND(c$$ORvPfvsL_x=`{%qfea9T1Y%fW;X_AKJZ2x}w5 z!}gs
  • z*)~n4P0Gx7jbMs<3zy=Od*(VQ}L=o|<*4Yq^O7&|hA6mWVwa6{Q`;irO8;D?dNBSBuYNGm?3TcBz+amd#R{@9{srFuGjVoLe59-AO)$@b$&Kx(qG!P1ZL$ zt5goPg&FPw7yZ>7RWN#lI%{p{bTc+9BSx9HS zzahjoID!Dz0rW?x|Id={!rW3LO9vcZB-4pPg*qBbyl$BVUcGBN?RT`Rl8sewHX z-~B%WQ4qi4#=^wZgNU>Xl>wErp3ZI4X@TJavR&bypDE1twY0QsGU|y43JMxZX4&Kn zMSmS)Y164OL&D1(8rDVekpunuo4}Cw^1Mi_1 zkq`s>kn=tu@EewTa7s!FAnjcD#^2&04lmSN;eG!&{{&>HL!*lOif7OJ)Df>LquDtO#iPjIAbGE#X-_CsVsiRZtbPk zZYmrj-jkNPuvDwkrJr_16NNTp*7W0 zq_i=8yX*+=9%>;a`mj)PYcJf*c`&lHl&(!xzA1Yzai#kaa;DvZi-U6wh#D@-=`h;1 zU^r8>ckli@P`fihOOZU`A3$IB@Q z9Ub2TQ1e0+2|hl8f;l`r%=6RzCGS@X*U-4Gx=<9N5R+0{$VQn(H?(x1w}ueF>6C29 z*MM3zGc$vRN3~vVIL{LDLT9B*f=tIqN=iB!aFhSI=CE3NeJg;$l1mDtlk(CabL7QB zqZ6%bR!<~e&6>votnt;}L`wUMeLmL4Z5IFfJlwN>2C_` zgro$YmuG1g7cNZ?50h`!;HiXY2AZnNKq(=-SMt-?NVLc3197mPaa=^f?^(E&UY`AX zi>c1$$wMhJvgb`nBNNHtz-c&}6?y`l9#H9@E1ymx zrUwkkUha%IIXeDU%8o9{O-jP$eoKpTR6j=YIx#vLO-p79i}xYxsrh_s0E>u-NJk}g z2TZk$XV_k*U4Mq5uHs#i0_~+tGTXTVyS@tDuORlFL54%o$HBqjD5!TNL&Wa|a9l}D z+P&S7Mp6x~HKjk}<2C!f=y({;14MZ`Uq!_pdX{qr7?Fsi7%Un^pQ5+i2k#F_u71)% z=~FOng#7G20GOFI{yJ7hM&=V+N@>?SJ*&zgLey~3MaL>!+tFHE8<<#Ug^G%bfyAN~ zt$`g?SDFG3_i`&;Umx{U^Djz_h}+3>k#fEU)ZJjtr$E*TSyZ{e&DY5Sxz8T1Cx9CR zIc&mh5luw|2q@ZD-T;PgT~I|u1!}|9-riur(#lHgSFDGvZFx+L58@j_&zLyIVJe4L%Qsq@iT=2SI(M*DR2?)o8>#KJk{9A`J>m+pRBCY zQIU$}Tx}_H?9=y_`{y1QZa>*Eok&`ki2gcFO-}5wmIfropJ&&^tO*jVR_22~_J@X< z7vdz~M*g^eaLBdjfh}0%e|=1tczybE-C7SuX)o?tS|Y-@`}4H~(lUxlD%u$k+EA5l zr+P;v!zio)fkGcX>@V51^#nsMd;uBD8_%fMjIYk{3();b5jYI?rribTi1vgX-}SxE zd!k7MD&6C06mqa4+$tO-`Q6WQa&sw0MFj6#-Y`bry~ZmE2YiJLA>7RdRN#hZp~f_T zd(MCopB#CtXnlBh(h3lfJh|W2k3%X&P2NuwC6K#DExZ#Ql@7gKji~0V8KSv!BmhD& zGJY9;C%k@mh;)5Akq>{Bb8zhHyn(u100~Ev1xSaq(CMGH?cZ`9etCn` z#|K-8=9(9fQ`4tPZ;C@gLV}42bs8YI@g;Z)7v$1$kh(%?sthygGseVy(GBLM`Z((l z{jpBrOB|!u0WR{(%Vd;U+YMPzb2!UunF|6kwLkqRnKmHt4mb0s)1+TWDtM;u)(kCr~ z(sxa9j~|a!w5+3*V)L`&>I(9!stWYFkAK~r{pju=UNWaFFiK5}Sd%kNu9S$x{Zs)d z03}|LOJdv@I8{-~` zFD#@rLnUDE?(c`pasx!|hf}|bGU-OA-O+>u>{3`fqm0OCGLh(v1=U|?#u>BpElYcl zK4~W_Vu+2ovmRr}j{2G=lv{og{nU&!$?AIwjobC4St_!w#$VfRi? zUr9jM&!E!t*WhB`;?LTi(XP0?`f@UTxIBrfG_;{XBV4l?o_jXLpHRYl;eW0v&x-y` z=ig2C{P;9q3_Y_o!|=G4US7HhqYNDvRuHX+3ozb*B2!pX^3!4W^Rl|)YSM_;J$~9g zzq{5vZ@O?3A;|P)|9b{l?fD>ah;qpEqOvrD7QG*>$Q=In$NapCzBbK-OUsKx^@Za6 zdg83*KLI5r3~qfeyAfl;kI2don|(10ny?`dp{nnD|DK&4lY6+4Mah?;V{{ZX?u~V8 zPLBVxbH!e13Mu(+V{lb{z2C(aV)D;N2{u+%Rw1F5NpCgB9*Qz_Zk+~u3O+u9LYi5; zY}!-naE@I_3!afQm&Y3$lOFbf;u>Z@MlpA9Bcr_JWIV;(uSg-BncMbGIq=QqV^3F! zUBX5!XI=?!FZ`u*R8qf^VBV&|c52cY08|7B;n5O_#|0vPn?3?~UmXpNuzA8=7BwQk z17Dn5n>duQ(iA41sDk+^Gr{|rv?U@o zGQvE-@WsACY+!hhgYwsc{_}I(uLWDHXqGdJ!ebSX}Xs}`7rXFVmga8V`iH6^r~ zI-5(u(fNe9yU7OxYRejlGDBWIvw#0eckhKY7zH3tiY!A=4j(f5`OZ|AKO5_g@px> z>#>nYc+Kc7i{`%Ej?;`Ol~fgeMZ*jiG3G5{_w(beIe-QZ;BH}`hw19)BNS;_Swar8 zAKBSt&cwSkC-{=i9tRvSKd(_52Wmrw4t!AEJw=15+l7IJH53s~K?yWR z`&Dh%)zwv1BXnDNeRHEIBeRk50$3_Wxb*=6Q-_!&%AFtK;Ts`j!KKFLW-WE|@g_hN ze0J4|!oOy&X10EsP?> zOx&oarSm%<>61EEVtCxg%rUp~O7-26LR#}h0d>JrxU#X);j+zb4Q48birg>Dt{>tx z<_TO$rlb{~*QgNBE95D~#MGC+W5_8mUTfW!^krFza_+2jV+>Fmr3QpY=NH1JOS0qX zAM*QhH%Hv^44_ZDkdaaAf9jFiw%zo*(9wQY3f}&?fgNy|W~4rL`vEg?RdKLPiw>J! z7*Wk7pH$F&yhOcXa&i(N9GBz8Yo=4J*1YqQ|6EGA84s_jVk;G5V2BLZiBvh1D zdQlw7kY@=piv~+mWA zRQe;wot+(pG|mnwm?BnRVx;+*u}p!`mBed@-GLsGuOp^mK15Tz?@XtR%_+2qeJ`!+TV_aQc5@ti?K|;kqMbyXw>XxT z%e;PLqXUDz6O-c_6}J9Mt!;`89s(UucKROJt~{RW4Yre7D`nC%aoO1(Dl*&LDtjsC zZEJ%K4(hLdz7p)3jbCnF154CKD%JAx^ErSu4#%QxeOQl~K^ip}Yl$*7HgvHm+F=}z@gOLaCN`wdaZ!>l$Lz{Ka)frC#x+bxw!!crY@1CpT{&F+XJA*L`Cg?8(UqSpBrXwt$($D-r6$BfTu7Id%OPi z`QiHU;jn^MDXZE29IPbrJiWB^xIa~>aR3nb@>p6e=CmQ8MJ9-?iwrW-CM*H(3kac@ z>d4oa%!77tbc{d%BMaC>l;1BsF zOtX8Zr}`E%M}-nP_WCN+!#u342=Cr`$`F1J>~4{KqYdfw0T@i|Y}R;b>XVZbS;iT! zH}Ld2;#Ep~ojBjdtBm_3f8giNOQb-wk6r71$RiF^ub|tBPERg10R{#Jhxw?<6#5C^ z`Pv<^4{;+aDgNr}>hjnuIs<|7^yFPy<@@|_w903Mn1I6eM*Ypri@7wW4%u%Y56kA~ z=YjOdLAHE-M9tJ1Snmj2n9Z%4&$%8W!Ppb?=i<51PPvVx5UW($+1i?4Tzq(Va6CQR z9O$EWad0raZHXn_*oNGGxt`~X`thSutXc?NAFv2W;W5N~-#1$TT?H8wI%cS?p<#ft zY;V!;eRH%h?#B6+u}!bVa~>Llh?k##XM20b2uMOdkMY?lLLQg)%J7!xBf)g^cy^DD z6#ms{Uitz|O$X)uewGK%hbF-&>FJrdELNwY+7^$V0Wr1JU$w7)p4Vu5e_?OoxROoQ zd4zsU@nI5H3`ZjRlEc%U8; z6V>)jGs~O2`Bffzh)l%yNe;8Uect{wf`e$tV?qT`-Vr_JJvQB+)zz~(EmUUiYkj-) zVQ$ajLBu^34YULEC2-ig>vlwZiI~Qn^YiZWm4?i1yB+>{A|&# zI{33TLBo(7aZJCPh@EUl)1#Kc5LH$E|udme^E;h-+e(iS@n9+Kx0_Pj8&$Vy*F zN&3@cbLA*JkFxtbSC;>WCsxnR+-rcTNS#^k?Dig8VEMKf|(wR!)mwnr|d74 ztS@ZmZI=b?Cql)wtq}-`l9X{<+uH!sRTfsFk0HT)N3$uEc;7z9#mV^@y54S0uf|M) zjBk)x9p%3_!#|^JZAM5_gZ&Bj9|3h4HB%`Y@Gv?ePRPz`pIzrw-!>|{_D78d4XSpaj}j^_kOQx6Ud zY=rc^$FH?)`)_Sd0fwc8ju3X8!--6r8gpQqg|V3xUHq$ACteY4pNUPG`?bozJedIK zPHXOWl@TQs6+ANj_+N`Dz?UGvK1L&oqNFV-@!uqesM1hi^vYhP}~c&}4UYG`Qz zF$v)#OG|(=v@|rx6+Atg>gwte6TcQbVIX7U;^JasCnP4G-+Kd=<-~%Gn-}K^--D9D zDf#8)s9lmq-r{utq=cqFbK;gdw_`)~pY_>}jWAtRAf~3=-;BNEhJF@*kMq4e z_X7eSuB}NfOJY(ouFz{R;vYE;4GmGzPL{T(&LiSMq~I2pL)>l>vO@V(@X=ATK?hWp zH$o_%)6PdFJSHvYi>(2%P}G?c_0xlyl8%m!-riooH-?T^Xx08vd6AKk0W>|xX%`h+ zUra0?KsJEfzwdMgR^c+D;>}G=1VrnDTy2I^*rEuz28V~2L2h==&YU{Ld9P<1Vnz3M zZO%Jo>F#X8BWrX8ewuKm6h3i{MZh>4_d{H zu;Dn!w!qK$_$pjCw_C`pYf?d8Zf@7j-YE2QfQAgFD*M4bkr?e)fXC%mWjRGT@;dZ!vZj)_p17`(P!u9U!u51*)2k~# zSc;?x$g-Hx-<4+ku&O>J!pHwOLPbfrz72rOo%{2H%T1y@%~F*ye2r_?)9QVT*X=yM zn)I;V3JEqgr^O`o-UYz9kin$sJa*M@iJ(-+T zVMKTp_P%2^A(?<$CbSBYyxpII=SHR#c0Ec74hCe?O=(iSshxKy;pCx zz<+qTlYA;JtCx!u&r^|6lWmPE+3!E)=k?TF6~c4zo*EfTzID8)8yuq73z|DsKP;XQvvZU<(M3hwHA2P>6*X&`deI^X=qN&qpYvbDwXW zOS-%B>a;Z*?a>yuwhF5~^x)~QGg#A521&(`LTK7?&;}S1b?DCU5ojzY!4T+YMQ10!;zNt zaMv{dOWlEY*rTeLDv20J%T)t0HTt)+Q~j`zWIxVA-Dhr>JhETIN39z-#nMjNU8Gpw zvqFB8YPh&ODe2xax0x83j=wT!fIOo}*b}4Jn=NPlOcg4fF4d#WMj}c{^;NV!433N( zZx4yuEC{Ust*zCrvqm&G^%=*RYHtn<49soI&CP|J;X^jY2L}hmDV4>=VGaemRGJzZ zQqt3nDc`@BV8qS;a)c|;?yVSRxH^)?)uoO=l(Sj6dvtVUI+RpJC^?|)ODz^;U~Jrl z`^m`9@0CHloE)Xpr!xqVZVWPQpby_9D>1Pr{x#(DIdapJ4F$4Ebm!LARzp)$HQ&L( zL5eV7Xw1xL9Bcmmy*fKH(5En?qoXrmxi2sKHxs2MR@Nk14;N8sGCq%ENVa>s`l7pE zer;z|lqo=*a=Z9kHnL}h7UsQ#sU-qz?%ZbVM^z~*hWGlEN#U|RfrWLcOMV=~FB?6D z`sFA2dV#04m)LqJ6N-u|DwAc}^^!59x*8e{K^J0yqi}3j(ICN1WjB2%8V2B8WuY0M+4NNS-oJlO zN0EoHT7##SlA4-pQ0a|(X>Vo7rOMG&bMU{WFJWo0LGt8VUR<#0|cEMsGIO{OrSsdJ-t5}O7;+d}>SL^3d(qpPz4GMz6!^5-HJYll?x<)nH zKRhhx_tM(0jI2bE;I6gPWF#X#|3cxPZrW%5<%4>Wd*0TU$Wq1irp*C+iD8>`{5UyuSG0 zdlwi4LYLcFT1LQPjGbjgL8qsrgbkyhpeQf?VbQEu`4<|S=*cn%+z!nil$0OXIX=)& zEzhIunu?s3hRjBEZFV>wg{wR|m(`TT?Lh|YoFsJm`ZjR_vo0%bha4_8eT?Z#K zp_wY1fqA4Usni3P}}MMY8(qj=|@Oq zW?*1oVp?f*+Pyx6%oRhy32{B#+XFymsIL#+Vl+!=sHwC)d}U?D*vP1}tLymqI9t#I zlZ-4T^()71Y@p!E@-i|qa?Uj5JOWMR3BY_jeEbI2N8gKaUzg5a45TW8*+f zHL>Fo?Eh7~P4oW3Of5@K4Nvq12vpWO+D{J8Nq|Z@^_LFK=H-m4aSav~N60XO?Mely ze*2j0KxlWQ?E9nD!&H)6aOV_F#XiXLq5~`4_Bwl6MO8bW2J~2}StI;%w+UF+g^7u{ zv0X89!}*l1RuX>az26GyNzA&IMn+KPrawpGZ$+vqD`N*t(2bm&oXQ_%-wzKDug?OY zDTZ2U7{)>wkdcugzxHxk$aFOG#I1x4-JWNYJ#%0+|kzG^j2l2rA34hgc7{agT()!l)t~f zzvo5Tgh@_FxXR;TIX?&r2{EcSp#uJzkv`_Lm_}?i8Cj<1^RN)tb=sj2{ae(V>Ywbj zjxK)&)PI#~=+qZ)wm40|(eU47;ePfAm!^*w9UHv<@pt2Vg97mmNk6hVy>wK4Y>YnI z?`O2#*hw>zX8kSe>)sNoRjcscYI~ib)EMF6FnPFyQW8=+byWGI4Hjx)LBR|eK9jVTF`T63LM5^=F6OTq=x{NK zl<#ko3SOQcjc0hbLxO|vA09-(+O?L!6;Dr3==94iUh34CE+@+bkvV5{Zj9% z?8b6JUOY|;Bq$E4-BjbkJD(!rg0pl1cVoVnC!5@B5~ZI=hIB2MQ93!cy5*w9pUpRKE`N3h?_imj+B7F1}wx&SII z+;LuB$8SL$WveoET%Q{7s+zxxWLWHp#KQ<6hu2y6eLUuw=Z?1) z;nVT7vVt_ht=Z3VoWFXyzig~^cXtOILcYrO-O$EtnD=}`EZ0+9Nc~12y4$R8d3ibH z=3DiUirhyZ$IP(=-K51g2Uz9dgVYnHxRp+MvKG6g%=Ep6#iXVr-Vecdwh%RW~uw6$ zIIp}iOESp!^Kq4!x0waTm1s7g(XFQ)>aD3#3xXx zRqq@ev;;J&$)*NAt1>B zeXHBz$;Hh*J3Z|TU?Z>=;2FyjUo>wzJnlug|0oM+-9)`09TbKv`E8++_dOPSyjSUu zbn!_;SCTU_H+ErgYPSw4JJoRN6pFdscV9PkJv z4#PXo4g22=p!x6YyLb9IS;Uzh&wIusJ_;N}&`ttksaI=k zPQhB=vC{1dd$yg4Cjiy4)mzn9tW@*pH#0ddUyqAcQj(j?=JVjpO?+8mU}}0jK%1?h zt!?}Oyf|Rj=O!n=H}F_Y!f=p~kgQb;0tV;5y}&~E{I)AsT()X@>}zr_Okrn2j(Vb|NHgkECSQgeq(w+?d0e7i zF*-7Wf{NPedaPw_Z4KEGzS_^P&_C}j0&u|Xc=38?Bz<9FLEra&I~WP?tn3T@N{zXa ztn52%Vy82lB*+lZ)fTV({QRnQz>xu}pBhs}TwMIqr%xUp_czB&HPzLqB>X03W^g_t zz-I+KD!{W8nBj}{c6KHv-Ll`UhXw~BW2{Zlh-_4$^^eD1_YB04G5TAkNP0 zwY9Zj@NE|Vo?l*qoOU7S%N*?NfC#O!fV4w`MLcH<;OAw@FvF+`96F=ln|9opcfG#lYq2a9?cXG za5;d9mz4IO_b=0Rki(ED)v7%JSlix*(dFhy^G>J}S)6emh7P2`Sgu+a{rfkKTHh*a zq!D1d(fD5$%(FW-lbLlRP^@SEBo-9B$Cb)pwVGvQ3?7l1$)%s1oJ7wX&fvpJ4}9|m zR*cPdr72jRzU8CrzYBhOy?HeBiM)?m1-;wU5~fWv^>JaQP!zAdRkW>487T6983grI z?Hx|Lwo*zrY8Df5!(E>pbCu0wLX)!&2o4cAr%Kc*WejrO z|Gp1|wP{ZTPFE;e1@L%}Zh%l;^S(rZ(V~p|26&Kf-#$vll~z`wcRmAB4;Kd~ikR;r zM&u<$$jcRQ{9}MZL~IandOkX`Mna0r$symGsxp=;vUhiHNJ_$GcXo8tsxcb@etK#H z#CChY!NYfjP#o&}Jzq1mwym3y-%z1E{RDS zyt$cs7e&Y=>U`dXCX!>VA@DBH1tl>s91I3;4<+C3P2~GL-ptO6_8r=etYA&O2YR+Gv*^PU_I1-n`_w8Nkw0drlvn!*kVpEQ;kRKj{;irN5$2T zzthtU)W~K@gy~p#X*9Sp8&^uU@53a1dNO5X`jp1!SHEa_&9p_81h*(FaC0Xx(b7WN zp#i?|ZoW_{8^D<^tA(XV@@hycoB@GU`+c@4K-U4zu(!8YQ&W?fnHexOEuPn%_2Uje zOp8Mu=@wdd+3&@d01qFq;qsam} zm3x7**f*<6Y(A1z05s=MLVzJ#Qt)kY`yRAY`c)%;eXf0JK}i_SIX zlEF{acV|}_e?>V2x?(psqV;gixL;pO>Gg9)=hivc@V=e7Apc!NMP=ojaw)L`xoo6@ zcJug=t#W!R8o{WKLFt`k3sPukXklE#i9W+aL&jJURq%p1tw>h816+hd^uiSHgo5YP zv|vvcw3P|Qbr>h}Bq9NVm6DQ*6weC_3mb^1r7D@8{_+$Yso(65@gZ(_VgiFBjuy+U zoI1sTudtwCgXK2sOSm}rn$RCz;n&6tY%;O&!0a5 zp=pGXK(K?d(l;Cw6vSmUyS1}3Tc(Y79VpwLR2^kBfxv(~O?KXXvI1IArxQ{2vu6Ka zsC9!x>Er&CTpn<{G*|9^vec=IZZ&h4_I@-(`N4Q)QoSn}NuN0AoyPyJ7KYVVNQ^1q zqmV#ZPKZd&yG_UxnA^Iy*sAB+?v^uG_6xEc3@FDJ*=?+>#Q@JiHj_+DOul?cNKGXm zBA*-|XQibz2lUWgzbvfW?A+Yk%#4q+En|?sjZ-QuEv-5|v^)u4SC8RFgr>vJ$X_C@ zn1p`&r@QmJp?CfjFZc~2U( zf3yW+{_W935{frBmugH_=J$mw)T?CeB;ba-pwu#vL@_j#iP)hN3{UaQ8V=r-vM7wS z%#D7u0DcN&=0T|@2k#*eqJ~v~AU)C+5^DLg&&9=sW0##gKFI5add4Y-|Aey1KfW5?D=<(ot<;Ve#2xwhR6e2=6dY zRh^`c2avG@fH~lS&`v?uBaITZGD}CdV)dZ^eQd%ALL~*7IGJu0EIA)9w9BiY46oh% zWhiTk6OU907L6cvJqOy>kTsK6G86XHiV7A^Y$)B&pDP`=2Z5k*0^5H3rr{DdN3x90 zT%MbIept~a+6SZ~jw#+rRz?Of7S>dO99~^HfWhBR;;5u#<>cZn09_y~ z1_%ZZ+l6WgI&4%@!Mpqz;RkRe3;AJEQqsi4gqfMyW^a_Ld`u|-1msg#H1ha-j?`XN zR={^Rz(=jT&;hbzSwg_`<+hqF^?iQG{Tqr(VmbM10Lc6}OtLv3u5ocgzP2wK-uK0j zxjFx*4b-gNN)yfyr62c!qP2CYIAwis<^8rV+HYiqA#Da3MKn$2|wG}*%*-PU~I_mgD{=#Bw*(*~%IEx*eh1@BtJ zE;wc}+fM-7-Cyof5i}EX0PDc-eP^G5r7RmmDuhm3**%iR1t|!sGVW8%7LKwz#|ZBB z=6->UmIb8W1{nO^6p#}^T)jO#6rUU%*gCU*{rUw2Cg4IycfZ0_}s>C=Ev2nLl6SX%q~yX=rFl zhj{(l9NqG{I62LRQ#zV!Y?kUEMJJaV3-Po{J0~3i{QQu~)>5O84pyFv?Lz>5qeRY3 zPm80T=`xf0zoinIg<9GX;;{bqobxG=-+Awio#Yn-3yXnAeJk@RNT*YJdN6J(tD&f( zA}aS=hOfg>w5O|q7Tu%>^uOg(e`aPz@zLR^7~!;d-S+C=M7*cQLVzhk(bw0<#*f7B z-w*;3#1x59V;Z97;n~^Q{o1&^zYCM4`|ts9oV}Dlvc19!WXsDK-X{1;?$ zs(3A4{F?ha+VS-zQ2NlXZI$BT>E1W8Ar`IvyXzLpaSa(l6$4IJ)pW_vk4L{PJ+$9% z`<$q=q`m$u)f0Y|;Ao~ z7s34R&oG~g{*6#aAB6N>ZES2jUaSSVpBn+7vGAz6MxG6jYYYL$PJbNLG9BAYCjic; z%Z)XrL%3dEkLvoeH1R794x59C3}sri^mKFxC@7hn7Ajmx2VFt`Ja8(beyex8|63rW z*8q0_F!UNPS(QdcVq(D^?xba|-C8SvZ(EXD-Yzgu9YEH)0a~)5p#c^NH@2Zo=-+B% z2Wfqmdn6@!dG<_Zi_*HTprwy%c2KiCu*9gxdT&%Lm9vt*((Y@gp`psBwCD#vOa9In3w_*GZOJp|am zi%6DMUe}|PkK~oLwE#^X01qNabmKcL667u|G2n|{0qHEC${x%h`RUW!3&?fgIAIXy{CwmeYH{GuGJejUs}KB^v(33aRPStTCGb z_(7r**AT?X$(hDw^*dv*O!warg({uje*J-W?N!k7-s6X2hXv1&#VCFc{FFVlX555| zmn-GL{F17I9!hvLq}SQ<#iq1S-!@za1{oSAk%pHvbusvgUQ{mE;UqxH7Z33{fMEpF zX;0Nw0sQ52+Q}IPxgHY=d2&)WK+3>PP2EmcgO^!!o2CJ6hXfgABfRokA-$-mC;%G% zV*6KNp;%_NuunucATs`2&JY)GDQM4y>(_=|>UqFbfJo&Kt&U6%Y_fOqX#>?Y^tq^bcBA@@;MR{rv3y4B~=^>lTlq)6!c#}8!hNQtvD4b&k` z<=;|2@Esnm_ARD<+pILr3=Ne40R=huUaTcaMI+(Ip6FB(dG7YyG0+8kED67}0V2IG zw6c1ieAweqeS*90KO*0kE?^g;I<69N#Z8pHsEJfHssX6{l08x`QXI@K)5fCxBD`7j zF=y4+ogCARUqd}28&CtjQ00|{&$b5i8t+iXrS`miF z4Daskm88b}Iv7+!Fkkok_YUw3l`{Ej{a#tL}7s)v9t+%i|jL3~j^H zn$n8CO#Z2;Xpr~5yCo{DTe(K~duhEl-drD(*HD6M-QXO?LL?98=>7WQ!@y7?I{@Ur zj*d=eSJy%xfQkTceJqaC12lo09Fp+!as67go#DNuW4968p_2Xwo}IOnJ!)D&=}Sc7 zLb~Ej6~-8Wf!NIw^tgDszeGTGA8RJ-(EzM5kXVL*HYtmK=xs*&;*$NJok?)lC-Nr8_>A$nc!vvm)4r1gL zGx@mygT=v70~uDmF5r5^Yut-mM_$ylx=>?oZfdHKExfX@u+Z7rIX;elF_KyjJRv~v z?GC3pt+&60Oi6uH8rl4r&U*3hK44BOld^=ojCKIT^nJdb1Aw;Ha^_%VZZ;J5A90{? z;Wvs?CWaLp)%=d`-He5i{^fVb4xL0m zv%!7?kmLqu1Ox;?V#*~mOVBv?wJiSs-Xq2IR;XU2T+~>S48zHOQes3P{GAu)U(nN7 z*CSqj6yp`?T9{Jg7{<3VDqC97h4{Q0mJ^?8Uw8lfWK$iD#ck*qHJumvq5{MHZ^U?6 zFMF1yzw0Dd$fZo}W*It2@sH;!$gJ;x&k%dP0&)}UUSNoAayuP`Gcg%JMnU1RS%mRX zl9vx^O;1nPZ}G%|i-B=Dm|m!}A^Ii4&%daI%Hc~Lq9?$wR*td-moDOc$IAHH8H@yQ zRPbSq;$;{?rAiPC9pX?1G)CNKy6$S#LdEC3{221{6fA7)t)nCP*%#)2^+vs)^_bBw zT##!cE@Y9Mx>7w_s5e1)CnyY_aD%mtf@eQt&&V85R<7`sO7Pdw`On{2Aoyhw? z*4_fDs_kL>MQM;OC8VUKL6GilP!Q>oknZjl5CKW0ySp1Cq)Sq|yBqIZ{?B>8amN?; z{l>VAaR%eKH+!u;<9UA1oO8_yf7|ONr1xsEM4Q)Y208&@TYsnB4wt?BYp1b_S+>Rt zp~w{qKzoLy^`SDcKZPCI{lW?oaVGz(MFv)2+s2gW47y*_MzWy8#)f2F&vds8rtH?1 zbgHg`{E{$kESG7unk+@dmZ&y$TbXm;Vmh=}x$?eTjXtjaX2zqrh1AZc>!;B=y-`eD z|8+(b^K8oIkhkmiQtnMW^<4V%BD?=$2cIM1MsbzCLJxT3o>uSNZnnF^*cPp~OiVGf zL#|xe)FGGuD_(_5fBgSG7vw*0z@&%yZ@rmbPFPsj#3UyvDJdr>Co~k<{hQ-{CXfLx zE_I6Nt=(NBBBH{=LWV}Mzkdz~gg=8vT1KY7x7Wv@p`oFppt7POOC>)%9MvJl{TLD3 z=)(sB0)kG-<+RVrAi#n@XJjn?rLC=@G2P&}`&mCzB0?g9(D=g#^#!6c;xb3TFzct+ zdz93m1HnD^Rifo!VNseN+o7PRHwWpnudmM^0c{7UWlfFK`Nki5GCL_bIkCcY&hqBt za{Ua*kObcm0taz*v?F-OhL39Ln2jYeQQ)iOx4VY^!L$j*pFAm);K(G9A35<|Ah4Lt*q-`aTf$i5VF$+XoHoNRZwtoc6JsZof8v& z_;U5YEcg-W_rvl4fRfFSrE+1gNNt0qcQWyilg^9~>;nSyJBAloCjl$0VmN``t7LSS#gD z9r!whvH0-NqKBUpR;*lHJ&S&ReiaoJ=5$Y>Hq(Rx8j|w2t*tGV;8(9+O^uIta&(d4 z@VV-uI{n12;5O(AAr}qO(A1Q9@dZ#Q@Cj%)9FWQ_W~nwo@SF^(O!Ha2=IsLc{bK5}IgM%d4uh01+q z-@ScQhP3vy9fksvhK5EURkwN(m8)lB!U5V+g=Ayt5U?Po*L<_P^XBdbG=DA&7zh-bNJpb0tE%de6i&oB$N`J<}gOh3M1sIr(sdp0v=bGZ_$_~ z(Hyo$9z{RNaXpb#uKhRzJ&gqXHUk=^hmPNH?ons?sP3FqAbiBM5!p<(wxf;vl zSC$5>7>WlopKk%}EjBu1L`GU(US0x>c}QS29Ki=d(xLhJiAj`7`e)$85;Y*tiYZf4 zP>>xUX)BdM1AAZINjj||8%qYZ&B8x^l$`hbJ{UrBoyR~L)o9+neG7|-4&=sXY-V=0 z<7N8lk)8)M2lmxuZ@Rq9$q6z3KCr*j)_HIGQFoK$?pu()SV&D=K}X`Wm=X$1Ur>^u zrQ^e!LXX;;$d3`D@SQa=Hm1X08c7$0CZlw8bfE9SW&Ln%=`EJOlxGEHqfS5dNAyUs>`xg$`?+@&%q3gauk z+nK4A6_es`+29#_0|Sa0tiVmHKZe=J$jDmlpyz?a7z96xGA*aJ0qQ^*ulJ&d)H@)& zbHeqcMM6VdNC*ZWI`|GC+~>^@EK1Cywkxc@rPAnVWzeGs1_wtadi!T)p6!aIpF_7U)XJIEcG$aYpwy`o`^(i7HJ$7tq$Wwb66Ng&Oaa3FpE{%DJ3H5_l zHpsSic77CdL{hHncD^0O+-XQSP)+gA9zED z3*XvTM2idAL_v2CB`&!vpQRUBo^t+l5X3X+j90t%I9re^b3YtgyGBQ3fq2{CuokIplKtqfS6A%ORc+BP| zzzQ-%>LeDguS6v_0M`~hP6xkG$sR1Wwz{60(0YC2aRiz=TVu%!6hA)1%+~f8RFCUP z@;w3q0#FP;OTfn89N&B>V-%T5xD`3pA4up6O8LM}-lMen_%*Z>U{_2q*L(q%dRfUu zQQ?#L*E`GN-5?aL^dVjglT=jK_3AIa-O=qbuK-?wEM-un6^kP)ke zhS%KOO?lWzelCK&S!|;03oW`FwYIFq+|n|x{DR2zbhu`wC|$;5K_vW5gs-=rntuJ+ zI~zLt}RoqrCVLst(OU|cCNJNAz*417FQoIHT~%{uJp|FVSkk zQT57D4wlu#ZHEU3*9Cyz(s~6};Njtw6S~ADuSogpnE>_x)-3|#FXPi592E4N*Xc<% zA-Fg14I#`fUUB;9cZ~_t^~thSDv4Doxj#;ks(RWP@uj;dq(Me=bmUY{gj-Q)^~b7y z3gpm5vC3@R1fVbG7PQ&R?83sg!C);Bd{BF6uEn~o0$rEunf@P)jQCtn0_1yodhExK z4As@u5fOt!LXeDx>T}f4#~rdhgcZ5$5)#z!?R`u*0H4T$1#X#$=ypz(r4%A!RREKW+qe?!RGewBAM)Nru#~Ulj_L7pykCyQr5P1f5hF1{P8U5n6WVxIcO0R5as>ZYX27X=7M%B z^@zw&9{}DkX z+Rw{+}hb<^5NJ2t_i(M$e;xF@-*yQ`)*$xP7nlPq4w*Y#t9)f@YxZZhpigEgb zva)hb3UH~5!^Ku0qRg5p!<)6B!-Dt)NXt*j>$sah<14>*zmCA@c<)(1q#I!NGveC4M}=us~jfIbT=D{q)t;$yXwDaTCQJHoct~YWXBV zZ-IcT#={nSTU%*a*+%a_0yU3!4O&Io>aoPNnp|u_=4QFd%F6Qbfq{D+V$)^7`hsYD zS`Ll~s5twFp(4=8TSScP_NG}C=I3gy1wpP@^!@|wCI_q!B*JN0xEIhC8&iHBn9E0N zg1RrY>M&oYd3`3@*&of6+(SeW$C7A=+FJxxMJBUFmdV^ zFTf4xMFOsImc9z6tZDnqb(!Ivt7-4B<_S$410fZoE-qi{;MZ#Af zkvNBc{_Nz~9fo!sBmdp%Ettjr^5v83w)37ylJGVk8VndhE~`%;XM|2ZqEEE|cm{kP z7!aUUXY=UVP~0C8<2jFmX|vmT0ear!TWAs_Cx@CSXk9!ZF;QPzLPBCb=35bgfSp7> zX0sjx1qH=)0=sC?lkJIo98`KG!Ot!rQ-2hZmPY1l1h{g2ea&Ka#nCtjbw9(x<(2yS z`Vcg)`*ZadK}Gh0kI9A-`CQpv;b1-1B2FFwG*o0Tk*Cb(%Iz!vRM3l$0$)sCgSiH` z-5;o8UkuGVs;5t%LVG$y1HC*kotA_KZ4jW5fEAw54%)Nx983)l4(=&9|DGwl@wXNE+dI>|4%4+Y@14MBdcVyug;`+8okOC;>NtQE<`V@*vH^|mlK_ZRruGP-n;gx4pQFa91r{MRSdA5QrN-rzK? z6{-*1iR<2f7-FUyuwcx)Iz}^J5|n*&J{MoQzFvK-Sx6Z&8_c82zbq~BF0{^ILVU$i zX%(5{Do@%t-an83pMjubW+XK#gYES5Uz{BJNG}eoQFYrCBfVyhwBU|4|LZ+?o_$75 zS4R%U1Oa3roBwWk4-Z^we0+RW=eJgjvbwsWq9PXy%Lpi64-30NxHsk&7C*li(}dLi zv$8b=MH5I4h?Gfxv=C*GY*SecB`yyZb|(va5IT#OL%IzvG)?_n{%a}KQ=!lZ`a*2r zb8EoHk=O}2&9n9@nh-c-@Sua({m5XJl&B^{5 z^~!sS=^QQPaCNi{@=YOBi|_y0K0JW_82Holw1%9RgakDW4G1vwjbd=e-~XMrimP)u zHqg{0Om6YGw(AT+J3l{uG<#Dz(`r<&WIQX9z@yLL)Ffkp}>jKeYWMv&4 z8am)72qpW?{%>i*KqoAQg}IH^(kOmEJ3dZNJ`mfNIPba#0W6W%xgRT%ugi1`ZP3gdaVeMa$KdiySu(PX<<;HpL33P+@5I%BN3HfSG~hFqPEL zgnLYw)|X|6&!jcoJjg5_nh=_*hB%bt=W@g)gs01h|D6!g@#*aBai1z5k5B2wV zY$G)~fL!nB=xEILtc0zfiA=c39$eYhh>(yFIORCM;N)X88cZYy2Zy6y{rxhsvf-La z)joBU^VleXrUKZ`lYXJRNyBYzFf_wS4oEAV|1Kl< zBU6L>B0`}_OfzJ05g z6;R$Uqt-KATP_%mKc)W#`$80UeQoWZM%XT8sgVVQmZA9O_vbHPz5pKtUUPMGqp};` zZBT`kY|=SfX5vO9%fQI!d3BT?85!ByDj0t~=l?^*4F0ij_KJQKxEDrRTA}xb!Q`!X zieeO4vw?3WeiD&5hUdJbr$>46L~M{Dv#jeD0dY27*9%-#ni3U<6xYE|m96gO0<@1^ zcR27cXM20qGj=U68#<7e=Ar~2^1866tP+YqdW0TCC@aIHMpSlkaQHZp7wzxg>0N;5 z?oOIyHk9)E=H|wl>_91|JDk8U8(LqSQ-7_&83668lMob+OF+>eT@Jc+ zIibWGqG!*R{TAwM4dq&ok7AYD0s;a+q+OL2k0LZlpn+bAMRaZOjP21kzI8Lw>C@U%eX!P{(c!Gw;&^>t-lGEh0 zPe?(5eTngiSRNMPiPD`^G|REC5Dhn#zdah$ACQCrbI>MtdbRZEUQm)f(ANf3KYe@6 zatKgJ1kSea*VtJ9(9n;RpYhC5L|o|hbxJ4^er8TI-|rPVeiKE#Hl;jBYUd=q?bA8d z^h?~UvC4kMNJqE)DOJ7O@s`baE;Q6uh( z1W0IJ@1#A}*48dAD#EP8$HOx=HfGdqAs&pN{q_6zcMGu#97*6I?w1D~8-vzs1A5lz6Asri0Wu~c;#XT+o1agI4J~l%1fq+3YaJp+ z!EJ13$9CeI@lI27XKPDOpbFN&w-pNu3#9pnhljYZ_7a|=IXM#(6E`=WCd`7cHzTEO z#Cbc=a0f3hJ5hLe`0ZZhFbJwB+M<)|>+9_ahx7IRJl4}?lFyW-6$58A5XEJ#Q~+7rkz`Mk!5B_ae%;i;m(N z8+myTrj`BQN_?Y4m;7dkC>I=^-e7H^+)qnb4hre_&lyg#>efKI|36h zBK!*n6Ifwi9sOi(6?$y592FW`UQ=T-7Q{iotTSJ0O-f8s0@JxQmi-fWKk!{>(Ah(* zVry&5^>m{-k5A*MD-_x?U3>@=^{E6%9a{=tC`PO}Fi}m&Mxc}Kkde113kitbZqBxW zvSXm45^2b3+`hE5EF*IJU5kzvwq zh>eSbcLsWfih&Uwr0 z+pS5Z3V8mABtd-#(L###SeCSDCJ~p_-h4fd$PX(RXy1v!WUf?hTM7@Xf?%h!%gbPo zPSnvzVPWCb9}%MH%=9X@w#W7Mo3pjnf4T{D*ETi`vXx>E78+Z7H1oI95=YF7exewW zJmvzuYRWtGTt11{nMsqKl@%-OV&+3?`ed5b=RIf*5g^9NNj$3$A3s75@lvFTa0GIP zEx=1sv+(6?6d~dYR46EcU?s{2l%xm&jW9*P^X9PikA!J(%JL0J{FxFFBCB?{Q51VK z#xEnQ_1+ml&_*?pMBf=H$-&b_BL69;>lrqAc}Yph*RO9se*6f&09HY|v2B$BdU!Lz zOATntCuqb^7;suBrET9ADx@mCeQpG3Ttq{oPQNoKf{2R(Tdx&pwkU8>9b{<#uf|*= zJiI`L&*9;yP4bG07zA;^&}bV?tMQQ@?V>;8{D48owq!Z-*cAU0u?dEujk9yGV&?=o z3R2$AWMOW8e(0-?yS=@SNL)`2Z_al$3Vm$g-kh=SLbM>^n`l0sH!Cw8oi8g;z%;xOCUkUk7C|`{ z!M1f$eLSX%D^wc8vDBCtLlcu41IytM+A%% zR)&20UpwtReaNu0N#HP_$dip{0ur6#=<4oXS%-vXZe|u*Iuep+Ii(_3jb)`TAWMyl z60NAIIU7C|VjPXf=37mvCYlf|K0<3s$T8*ZRNl4{vH}2aTJmBS?cCGNabzkahLVCp zX@AdVYcvyaAerCelSy?&g+V(@Ez+({2r?xl*C0J3|Z;-uspE*&dAz6OFV%3E}u?`^pJ( zx1jyIXd8tFYJA_%_8pDSZSEI^gy(r)j!OGLRY^I`aS2HnLnJ=jFC538ldJKuyw+G7 zKu{O|mid6CR(W8KH0E`8EZI#-61>Uq%2*y|F1!~fH?Y9iWUfOzG>$grv_TzbEi~j( z+`7ueA4}~W4TrY&LV<2ML`7+7XzgUx@9}ZfnM2l3doW1!A1y5xmC^Eun#EOx}x${%S z@73?|b9rMi5qmAPPn0kKYEPSsigv2&hLNce{#CcycAjcdemC3?T&9Q)kA6M<>nkNw z2L}i8NgVYy%Od*v=|E=8i5l>+vHc2rB}qNAxMwL9y{P`Pxl7v{77h+v0xxc2raKL| z$Lg(Ko*@D+(Z+ibEiZ2mTLEPr)T`%jKAtn3$QuRBSB?^k7|-gKe3|TvB4+WN*Qp2y zqG>zmsO|Ol_xGF}9E4m}9@i)97Qaf$ZN;aVX-waB4FEd|iKU>R7)c8mTw*xFemR{c z3=3!13%4JRFPRnW>F*dX!eFQ}kbw;6*VfinSI3QnSi-{{-A<<;JU?d?VjgfG6v^<6N#S!1 zXkot_oukghN>y`*La-=DXtT4kG1=VR-CvPo??jAhc->R4DM3@PrLhCi= z=e6AI&^h%C3_89edmWU+ShUxEBzzMlY6ddR`&ugyWhe0jg4*n z^N@_T@&Kqo)F0DAnBX(4NZtC+ct9Pmd3g;>g$Q5znW6+Q$1$i6ed7(rJKq|GH4o_o z22s-!McOkOcCr4$)u0 z)PCdkqN4T$+I}K8tw7s&T0?_b6Cdm4#n>-wtqY$X+1~^~;}g3Ge6(w_PF}cdWMoU# z2XE70n4jf_lbKUN$(M#s3Z`WJ2sULx4+8|V?O&y&0!c|p6~goBBw-}3p*w!}m{H-+ zn4SVG^r7hd37B=+Xzv>@pX<8lW0wAQ~`bl22^%iMx%Wddc-gm%UKSO zK-rfm{zMK<+Ko;c@WdHAE^HdN(CGu4X1!vbuM-m!6Rb^4e#C0O=~`Lv?Khh))BpLE zk$mJ$6#NwBI$(GPTG}PIm`uivp>IDWJ_*ZLj0zqT#9srGVdLc`B7JvCzS+?sLTmwi zSOcDom5t4y^;af*IqVWcFApJB-w~cb#q-YM_q9K){5dNGdWzI&pTFgT9_#ImITQq) zQ&MJ68ws}TWllLwlaUR%pReZrbUEeF=eX}idT9@~dKGz}ND_)Qt3`;)GoMHvL(3AT zBp+9?+3BH82%&GF28s;yR1`Sckl#J!FvWD5AfR$NPzX%?Zuk3RHgkWaVyeJo9Qnpl zuR*C&UI`y^MvcfyiQKfOmJ{M*0_=-4yvFMu1V-N+8~<>I#?S#t5?=KSQd>BsGeh+ivoV0s*zW#VcZB<6LVZ(if>kMOsp?^4QWW884 zQ52SK;Q8C-gfa>3edt+&1X)jWZ5;ZJy6+?lm-hLb3}TK3`r%BJnJK5!=ZXXAz1MJG zVcg+!-ihYDWhkRPrb5;^tNcAzR5g3*VP(-`Yc&}ME1ZV{Bp@K>Nq1D!2lSrIm{Iq1 z^XgL_&C~h3A!b zjBZPX(Po5vN=88eKU54U8QKF={;G_tB90=zqfgS0h!DA%OlaCR#ZEbCxoN7u-t!-?|D#AsZkb%;E z7)nVWrxx~D%KKioiZTJ>`)H;lLnB#NGPKP>!eW+m47lQoJh$yiKx8C3iDv;)z~Mx` zYT3o{Y7ax>%dTV}{DIzHp}V3zKTNoEIZ%Kxw(0FclVK?aeSZ) zzZeBpmM}#$E%fAJh0N`;qK|d4vv76M3FY?K+igjDP4(98D_5B(dQF00miI0wt zuLZ&le7OCwzn|Z1wMFnM`YAg*`+qj{i5nRi0T!aGyI9)#FuS;@JF`wjpR@r&4={;F zjh911LkB=SX2of7E37eqYq}h-Jc32=xY^E~mt(Qt7$_@a$#S-{+xgw>{^#L7PR#?{ zqPY)5fEHs-uBT~Opy7c_DLdJQ{G#OJ^H}Z(JZk#<`7_8)ZpGGUh?ZxQ>bk;%x$?;- zhK7~ai!Ir*@gjTk3k%Fzb!BenJ6`uU?W>s?88zS~xyaDa&G)2iI2&(pPacZ@CEp&7kfU-ntj@8CrX>DJ|CnDnYy1Vj) zMJ(vMez-lv@=V7X^|NVg@w}a#o9jdeH|Pn90H8m%^Sa&1T3=Sy3u1%l=x9^AE@(@_ z-KpYJKw5Wxety=i_fE$v-5vs|@3XS9j6gL=jTfpjTUzb|^vvX3K8A+x^O&HK0}Bfa zEyw=;{-N*u;TE5uCjQ*l*B5%exS~Sk3?btOAmU2TTPIMX_%C*McV)B3#>P}ADKkNR zY63c+pC>*=Lqm&*hyWDi;OH2qy@o-;fs!ODD;rL8_|-&D59Ca0j%U<*`vBppD=Xhu z7=^A-I3NCIeEIV1{M`7^tfs0;EC~QR{Rdc3WCgGw2@1Qq)+^Bhy^vB+c9nqwdO(MfyP1;>M}h+AE;wefo7ln9*Is$Ty{Yv`}N~O!}(fY4Cv8KA1)m_J39sM z&s0HN1vS7L%EiKRbaCNiXQyHpD)=!=g>Du;=V4q?kUZ&2pV92>Y<+!wPEHQ9tG>wU zkBDO~Hns|@g$6!XoX&^2YKzNQL0~%n3bCg63p4@u%M#6yu6Hz-&i|5 zkp1G~;$#l93irzc5SH~Q$jKj#gBWr4n0mcq^HlQFQ&zLFEFdSKt_cYVAC5XPKsO`e zG>=V8oB=*!Gj9fpDowN>=-={FG5?y6Gj?r0k4*c4{8a5VgHK&-75p$nY;27N#d^^d z9ZDrtRapgvAC6Lgr>pd!bAfGb3E7O0LK|g{Ko$~oJ%MXb0*ww@%7=y$!|id!La00= zPD!u&_H9txQyfH$=VCNL#28hdZpsIjGPEJlrMiRfaSrT%)(g-ah z6>V_X_RHJ`tv`xX@JG8b=*vcZUx4Oq@9a=02-vM5X=1LRS4ytRWRIDR=d>?yS^s|0 z-9vs2?kbXm7kUixIH7_ogbZnAFuhPY0;UE1_10Xib#@Q2G=W;6uJ?WQ*qq9-GPDkQ zom|IHiFOYFV$^ef+z%Pe(@*a}%>n>MLPrm}TXyM^b9y?K$JX!Xen&=2oA<8L1nCQC zCaqU1s8fL?pNP1u&bCJ5zKRs;V-fpD zmOm<3rRYQicjCIrr<1fg>o)##Z_Jgxyp?s-C(^6M;YF_`PS7>7>S!l4H+IV}kO~fF=W7OGfnXq--8S*sMH? z0%jK?(*OU@TKw~7ADJ+BWo2arg-AJSRVU$)UqYaDii&0y6+J-<-oObJQ^{AG38c`$ z<4DzbDN&eTSV(}6pF#3>x^(FjSSMh`Nr$4 z)9Uq&iLU_O{q;4p!2~o1_Sn})Q>j|WeWAe-6@}aF%oL?T4VBo9+g`xsh>VgFHljDC zHm3qrya~ZnxJw%tBJctgr*{A=s0(NgxV=Gx-`U>Uxd5h9koe`5UJ}xf(IzWUxQh@lA!&yGQ-28SGRF9#&CH$m7leb1Q;~-2*;4s6U z(C_;~)z5@JQJQ84xW7)ExP%Rc3s>k|e_@Z(ZQ**Miq_W3=@3;Vnaom|WK0Ots3=_)$ z_M%RBZuf9^-A_*=j45_50v*n{PV?f0(Z$}3o0}VeWYCa42PLw97@I486IfTrO$)8d zcRrZkK0GWfDJj{*#$u-A;v#U%EiFCV-lmH)w6t8Pv3$?;NbR9Fni5!bRb^#NpOMT< zdSyH@t#AV7$gr@A+S=itKj$ipQD;q8xj*sW8ThWk^$|@>WGT4mcMQ>S-H&eY=++t#5&yIUYfhk0HX}{ri*c#a$&uzWlNRFXVJn6+V=VNh%h4m-( zx=Q!{^AL7AYBO-r(b0kaoUd1htu%`bn|bX=1d;ds`LBUQ_D%5L$LTWQ1Rnuxoo);| z?oNrvkbnGLWfl_^g%SVV;w`}DH^E)a5^bBOz&F?c3~2BQpahDbqM~j!R{Ruiq6a|g zd3#Y^Q!@pv-hCJHM}j&A4;QygU3p_;LuBP$Cu=jHeNi$?E31d=&D6uWTE&?_D*Bkz z)Tpl~r>CoZ?i@37a~LE%XV5bw+#c}qT&%2p0|No=yW893XA~ARNEuT?DT8&b_FQAkmq$e$o25>P;0#yR99;@{$p;Amyb_thKh>bKPD?M5Zdt*DwWgCVak!&GX2gt z2KC3Vuz-WUfB*g*7uVtL%2r!jyUQdYY8$+iQ6Hym^G_l`Adi!Q&W-@B0u(gh+1o#} zjfZ75g(QNkcodn`hgAq(g)XepJ>(Ulg!Jq_6lmO_j^BTF6tz3iYz3{ zCHzMdy8Ni<5gsYSiN~Uco%!m!c=KK^xVlfVlDc^1KE%JsK3_jx zWu}^vgnYE9XKC4N(2YVwL_|YF(VRg)7E_CRJj{M}5WkYf6)WZ?^yEHMd5eti{@17Klmb4L>L!X%=zw& zf`T3JZeYN*Wo2ove+&(w=aVe83fEm_+5}4K$DZ%a#9({a*_{CCr~TvV>gsvBUrQfH z&&0&U$r(2_W>?>I5l7)$3}q2P!B&tcaA&rAr4a`}OLKR32W)z=JucZu53__OVnYz#6GaM`1o*id(1n|u8>xMmKUC(`N!Y(bacKPf{ohxcQ;pt@U>i4n zCF~5w_%%71eJrKlTvwN1rw{z?-MeSI1_Lf#m64K4`#oA&B#Qv!!j3sFo?cZ2s2~$ai>;$0zs-`5N`BUo!ZCDiOEt4O zLWbkj(J}~_l$4Y*$m%6pbj-{^e{#-zbldRAHL>GkW5pySHa0gA;oy|yT0u`xP#6OF z+;qLkWRP(Gd&l-D=;iK*&6j0g@nO?ZNj;9*L7nBP0uCz5%L@q!Nec$*>TGY%>w2;V ziXMa<%_`G$CEAwj4I)IuU<|>#&yLG5FxCiujkHRi>fA3&a&tHR5it%H8fiuc6k}=0o zZVZ|-Sxxozh0)@c7AQZkW&QK0@B!IbNC94bf3k{XFv0;&CujYEIuAZEj$XbB3^8u7%M1g=3`~5L zG)XiG3?VX{GYs^DP6que42%s-I|=mL2Nvnz?+e62Hi*#gB^W>6zyE-UIsXXy&F!;< z4*hO>^zTO;5zH4B7XB)J@4PjVo|Hs5RV{Q6oeV)o7fJ|16;1rlQQ;P)R*;vC+1?2W z`DhlKXroWSqVZ>|B!UF-<%`|fmh#lPkZ7J-fuG-FU0vO&Ko}U=J9uXqjQ!c_)3Y z^6(+mGe?d7T|DeLSy7^)fx*_&GkkpNQIE$R4YYZPJYM(&1o2 z%3~;4@zy1v#1?ITm4pWg2K%4VWcoi>h}C?PE4!fJ;(wly1GfNUP2|_8p{A}5Gzug` zPAaPI_I7=sQsAd3-0ucjz7mGVfPhBGq1Ofj90HWpH>F=9A|f{z`@Gh_`SfL7U9Tb5 zxjNDZJdl<~W^HYagM(8e?w6L8wYD=^2#^MJ2#0_oPqUvuT`U({w9}PbTwNocvWmE` zca7&N>b7{iCMRzn9%d#ZYa1PnR^A8q1q1F&$()qUD5SV}zwxle9Ryg=L5Gq!Ejj{_ zi+24Qc;V`Cuk zwf)AY=ln^Q-#(|NLhC7rc^rCy^#1-0lG{1xDe+9&=Hoeot36RY5k!&^nUm@O$!u(F zVyNY58-E(H&3aFl=|jsYngiY%a7bllW@f9u_pn!hey=7Qk z$=0roy9IX$?(XjH9^5Tha1HM6?he7-0|a+>2=0*J&bN|&&+bm|?mqpVpWn)Lk;Q_# zsT#9t%rR>`RduriNJo(YDY%JfEFvI@%Z|NEQ$cLYFLXBW4gh%mi#Saab3)jifn3}6 zWa-nF^Y;~)PHEci{PFq73|=tMQzcAz;Cr%K115~uhg;Cd(<~BI;v<=kPQ8>>r@rWA zdxN6sTj{F1y9Fax?WtTMK9ZMxB5!U@-k(Ps-7Rg%Yw^-q+Ij=FNnr4M6cg>Ms)YY& zw#nVaO{;$1il_2<>D@c2yLXn~Lqi=!vx)hIj*aD^j?`QH`^A^{dIG# zQ2r9cbHIf2y|LBkp}XO%_vCjQ`Wr?Su6(l^e)OAtOUe7S`NpP846dU_5+8m|5JNLp zEn}`M|Nc~9ipO1P&Sa378GyK+Zv+wf`CRk;nh0m{*jr?FW$6*aiRy6q&CN6ZZGD5Z zfLMf^(SEO}i(B>9UO{e9Bs7g>`Eewnqj}Lh$*key(J#I6Bzd6jfhVLSu=opTySt6v z_-Cb!iz?-#(7>LK1VAqh8C5uVSob`I@ZsNf4ld%tcF=Obhy~t_uxCKm7aM3EfolJp+b;>GpOpTs-iI=IOz!chd~I!vm;KVEW=iu4ls1h z_@w(gsvB`xGAJA1uDnmW5tFqJc+-goM(5LL6tTu$)&dg>4Q)KP*&`>%s;JnTll-QA zmJsLKM=`;hGw42tEN((xw1H30mVjGpteIO+ue2`#HX8{apU%|V2j!sl{Auq^g!JcW z1rX`zpdbi#dU{HL6kZ)F$vH!hSv+8j=?4cli$G9n@>k>sl*5Z6nC+FGj`xaW$Ug8r7rV zLa5-U`zNB+%gyp8fZ!!06g-_bjJHhlIRwOFkqLnacM zPrw=~Hf6~zv9T#cXlQdUGA3tzP~+gtfW_YjQuk0|oFLVYxDXlrAWC>AJ zHJUzWc;hX=-O%n%LjH>|SXnD&MOe_oFWE#yWpiT=o2P&rhT4qzjaMgRdKhCO}IkGc#Eo=WErp>87H zT{6O`50ZpE>{zd$K^#~iN$!!0PYwOKN($<7<9z{vA=4A@^b9btBtwD-(uZ8!_-3fbWFSb$(}*-pAthu* z7Aa&z#EDIWwBAWEKg#3jh9R%UhE*&$D$ec>IaHYYREL=v?B{hY<5rZtsOjm!K3dY& z3YeV@JUwayM=LeZ(@>qkTbhSMoCI}+Nq(M;$3{PB1Tmw)7P&bpUr{|Ug~v(>Y_}+F zLotK5R{=*VIWY$)$XSb}z>`Gq4Y3@p%bo`Q**r#dY5NdW-3Uc6MU_CszTb1*zc~Ol zHVU@K@Jh4&By$sD*X0r$crN=^EfOut%!((gO>(>)eks0aG}=|{x{=?j z3iR81@92yQz#Ogz6H+ZT9fir zo*Ek)3-ADAVJxr8hpPDlb*jjU_8}-B)<D(ZOg*59T)?g}+{)%qa$MM1WmIZ^^PLC@v<(V=1&w^regR16d_z zVL@F68JlP0jiix-C;_jlO`2c3r`IV7XV-f%H8a?R;CM^UjuaqfjIWwJYGTc*hoO@T z$*>l^fbXv&P`>;W30M;bJ@xe~Aq2EM8iRF?Q6xM(*O<*dZ^1fX^HxQ&6?853QQjeFV76MA*k)Lgaz0`HQlXAf4c7#qutO$X@X0ear2`g*0APDwhCnNMj- z4h{}M^gYXMz0d-7Wct%_apyRg0PXS(JO!B}WB)B7E)X~w30V>~-SEUfYsfis*#+(x z68lN0h_I;1{@1|=e95N=*v@xFA_3hM2(g|udms2 z{-A+!4oF*NMNISW`Z8H86wt(Z0r#@05M9c9z{YC=Mm$17437dj4QuPIIDl1IVDLsx zPS-ljJSCP9&Zo%}_iJSW`H&!aM-bQ;1C=_OLoVwSK3;S=i-N?!aD7$!!ZB^7(6#Sq zDNKaewd<|Duu@;lr6l?I_y&_RvgL^2L8E7^>=r82dm_R61`>Bf;@y)i&Ol+{cu&8~SMHY$=tl;|XJ z;}!OS-x%vLP3N5wBZ*M0ywOl)tO?BrX{UG^|5{UpxQ0n2cW?EH$CvAuOP<@OeGc3K z?MrU&{oBX;tOch`vmvI%_VV)bQ4-jrdwUj42sk3US7IhNHq~Jno|%xP{xCQGsuT_C z(QOU*79Y1uz|P?tl%95CDB_IC_yNwAYiaa>=X3B()!N_lFYQDX626VcTk-b=e zpVQSaW7V2vxjQO5Ff(XZwe&Vc+b%9Gp&iKCI@He0WNvOBS|nG!5R=i3f2ua;pyhoY zdEyzE5SfXqCiONzE!&lNRz3l$-DfDx0s4s66%&(%HCFBi zNvqWgpTe48;HprNW5p!X`t1NrqA4J${PGbtKImBJPK((@+F4Xu=iwJl<2Va0-hQju zi3RpO;}HRc#RaUJrOA2|{mt{E^>N7ys)a>D>-nwPy4qohf`k(*Q2t8xVb7jUI;{2&mZKKG@pYwy~Id3r43bFGsQ8aR%#D4_;gg=NQx@Z@{QOUF?o1iH6WAHDDf> zg`aO}@Cmv2;j#FYC|7x)RZ@dbZ*`+BX6^t zr>#p;i6zZ{kj%(Xi6FCavocgbfO zd26fR03^7TK6aOsROh1=&F~gQhqOQv*Dj2`{e2w_*MMB(>#FG|IS*G?{hXjl7Uk1$ z0qfXwox2udGo8miaZ90jXzvNiNmUSxBmXW=**V3qZjB2AFj=Wl3kYHl`);*;R%Vs}(SWrgv7Nd^_X%dc|R)ClGwnIt;98YU=goIiKP zjhc_ZXjit2h--RVgwosiXdGgVO5j|Qfu;KWeXiUm zs`$eogj;^S#7BOfmZ@vtjxOv*wejITaAVx$W85Rf&Aq zZ-&Zs^lwfQnCg!9A>oomR@o(>6JWk4*5YB;mfd-gvawPT%!jYh*e|`&0H-rfJTYug z480fe*UQ*dEos#sYymkNA0MZM$UpK9r3DwFcAbss>obinZ$y1ZD2(pzGKfkqGk z3ElUkdR|X1wbok zdQ!plpH9*NGT_dSYe)}kD1%~v{hbLViRz9;ymRcXZg+=>URzSJG=15~p&nvH9O1^N zzkqQ%dzqR-Tx{f>Vj(2l^?vl0J=eQ?fa;MH&&~s|t`p$`et{ak<+%M7 z7$GXQ-zVPkDHxGhi?0t4=@`RPJE8Q1l!~f#B^a$-j9n!jB(jajnLSfg?)r6ZDmU;F z%ryJtP*TPtei>Yb2&1(|-|^$j8+|`wHDYrl?e@Dj0mda0z*7#?<>hQ+ZySb&+CiOU zN_1@-Ngtkr5pe`4nwolnTx)q&;z+M%BJ@fjMvMCg9Oi}QVv6~5>Lci7%}mWqBiek@ zNI4%5FLO+F4pU$^wGxJC1m>6 zwdeXYK__#(&Ajh*fCmeE)?CguKR+t)iiJg|rUSFJ2-e-HL>*xVA&`8zlF_#*<;k@S zIL_|UD;($u0@fH*x3*MwLy?kF9TOsuzbv`y&yX_5&Zr?H zjX_O7X0&vkX7BObT!@6_m^bbQ`z=*eR1PZyya)&gxVfh^DnD7&zlG;a7ekkc!p1R? zC*&!uP`w)LKyq6bmXZo<3WoZo2!E(YG;6^YR>*_MnCl805ua~#t|zf3v>5qd_q|Ak zmcR2NbHk1dba#+$RTT-8o9jzq0berWsd&YKPnzYn9r19yL0Y<-fRSj>(rfK*5&Q$E zIxH?1Y7L(`c;?*nG{~IX?)6YXXFhEW<+$&;@=5Z8#0MuQC*Rz&Ggy#2h8#9xulu$L zz-Dk5q11qkk6q#-i6ZPFw$v3oq8bkuD$wAXsPjW^v&IC-|-Hd>2dG6+qjH(p`07(*~f{o zWN+PPY~kn+LhfK*ADrsx2vHj{U^l{6u?NR6v1v^S$c8DY;!xn;wapk;T^Un-#JaWw z-+zUp4cI5p($-EG%qVo_-zG9_s%HG6N9{hIXkr5US)f!qt|8HRtQmz;fVC+##mv?h z8Mpj)WMZUlV{Kh#R|>VwXB=CyAUE@K&}TQL;NIRIIn3QbE9?fTPA$(@mT{M|#Q;Eb z(#kgA@hxZ1vGzT)qVYbt`Pji^45^W{RHTy5>Go+R#wjjRwXP+*hCz8*(L4|HHWU!A zNw@d_uDmyWnC-Mg}$TGWEunnSZK0~10nD5Pdh zzeZ>=NegqkYV=y^)J}5Rw{JJk*4l&T8j1x0#{p}++>pKH{01*Mx8Nq* zXeGeYZ;IY_H$OXDUpGG$K3YhOrTj@u%vMwdbc~i}5mEZErK@TRVQz1Slh6u6LJH=g z^U-LDLaR>*Nxr?)2!*v-mqR%?iwfze8x!JNS$s+(eJN_{;^WD;rluxl(opN~4xX2GIws6@_0(@~!j{jD zu0|$v2pi`;lI8Nmx5NNPlxLp=6jjA=*sQlp8ceY>5K?!J#5e$L`n@eRTnhgr*zbTN zK|Ok76W#=Vn>~V3;eulCLQ{o^jy=CL z$^e-13m;$Ea*G3FE%s>Q2t`GFml`}Ut41%?O-@MH>!s7@SI7{!aneaYM>yTX*-{== zeEC?m%P@j3nr#mvRv{gYWM{28{%zb@m-T{$M#B{`86b> z!fYIqs0vv_8JUC;CrL67$r#Yl(7-Jzx5Yx5c<&!V3ezkv-N$DLn^n1rZ@AC@t0t}tnx*gS+QBuZ+V+%a!s%lO#4v8oWQa&<@ zHD1Vkw**HZ;TZN*C?7$-lq%`Vw?EmhGYDHNQx^@{WMe``7}{25U} z8WKwSAd#)c2pGu45g1588Wao_2nq@c2;Nxg9nc?MZvgM@>>TOs>`fdL>}~BF=v=L> zB4aubdKlmWT(a_p5h&rHRO@w+hz)%pt2L9wEHefXyltZ?3A_7uG?E`3_no{hg`EY_ zF-ll2-`Ow~kkv=Y6>{P?&Lj?z05>pz^-QlS%IL_>*Mvm)B@mW6Or+XNz);9$E75#P zED}d#f_)>|Kvx}HK%H$hVyD_ka6&YZJnv6@d2`ACGU-&=Eo(lm3))T?(VHUTj}R>8o+YGl7Me#Eb?8DGi<*c!=YZ!8-y zR(nXB+6BRdR6?@VQ##)aWTh75TBxEU5yr;$-+bSn7}((4&Zy9u(awWzk~5#{t+)mw zu1pyZz`p*r**(5YCN{2|s91`2zKP{>r}i36JLAM8T&>YUce+_~TO=J~WR#`cR0*%n zz&^`Xj34AI!XzE-%aTkY8;3Ads?e-!TUocRan^h>>z`jPapEL#Wk-SeDqTkeSsC`l z9!qjf1C!mQbguJrGOTICjBwYuqptFs{OA@|O_4^mTF=V)$XDSwy4al)F36EmUaC=S zJ8X_o0f|ks@2<^=xrUESuocXZNtct#@gDOh={hUo*Yuz9<(z4ev37vP!uF7CWcWa5 zN0LS@YZ>QY`Y1A7KSuq)1bgK9_ie>&Rgv+fT0Xsr>2cgw1 zA7&M1+GsTJi{2sP3`|kEim`e??GD2*zM$k}65upq{*Y|4Nj%6zP7y-q5UBg~ zWnD48d!$AHW}5MXPD4jw*|lSlp6qxar1fsg#y8ASJ?z`2*}-v3*Ld=b5vjerE|`W4 zwV0(MlyU+lPtYIsJF(smx&T;51LXhBe$4-|Uq^hm6euIo5Brhcl34>Xu#FgKFz&Gr zKt|5a>{e_jOm&Udhmo{~u)Gu!(Wx?>_b5<%%(s!`a_IaF|{xrrsSfN zj~c2bQdN|dQq8x8bQvh?aEIR-2jlx^U>PjSNUaCb3BP5ffTTHZsdQkvfucOIp!v2p zCEys;Cd^3|5lX|fZ*rjK{yOGXL`BVK1+ENKHF{4|2WwJ_5`Z{>SD#Q(j%$ci_U6me z*zJ)}XZo0!_aQuV;9y1WM$^GNK$>ZQ{Z;^*EVxd^A6JuSg zaLayU!X1f;8oJpakw^g-%TzHeL0N`8tMom=A}t64H5unwUU~73?Gw+)I}B%r zeo7v&hEa*#acB!_N;h%kKq2fJPa}ujG_b3~3otekWNcCzQp^A%@7gLs^~incZjrAo z?dK~m#kMzc+t8H;lwJKot2gSoYO^SFRx?9$-a+&~ti>Vd8Tu8_E9d|5zWevqV)~D@ zBBRG({27n}0KKyMmh31u!SG56Mo7MoPe4Kqo|T`ZX12*T3WoQJgS&2n!Re8a+07_q z`5gi&txry(oCUYN>rK-(*7BYLza2WI_*Q`z$^C#Bg0j?0k46Ft_M6g6^E&2ceUV|N zh*8*kWa8Q~3OO1ptjUv<0%csGVkyGsqwfTbI1seok9G_viwMwY#x*x}KG{fOQTK#| zf)=hk(y2^o1Jg{Hv~`SO{AhXC)uG@gK+DPgA6w4ypO#m|PW zl)+t*I5OY(SY-0a7vFhN@%7^aU+H?5sh;QCn($mj9Gx-{|Hw>O;e0*>`eX41YxQgH zAEPl}T6-t;b0oItbUdS#c!sZ&45ocyG>iXE3`G2AIfk6l;fwA`THyv*)0%T^(MOy7 zuwKqhy~UdI%s!p^0WAO3cLx13MCUyTxdNcWXd@Vv=^3ARK>}eT!TETxCC0xw%98hp zOc6Qo1>u|A@MknFnuJaO4jTQM*m7R@AWwgy6I^8%C4TRee{KD7O69Pfh+1pT^n2S4 z&JWw^Dw2?#0c>~i|7g3PG2733!T&#d&0oc-KS#>0#+uTK7-}bA>g9lH6wH`A2nixS z*FuszzDVXHzCLYL4TMEu<^0nX)#u52c=#C$`U|iQ=jqlDz}!{U?ty%gU3GB`{Q=V~ zB!>xD#;0dX=NZrAE>Xr48>>*DjbtiY; ziVkd|TdUf4fDJID{UCZxQ%Aq2%}jG;Ijnto9~wihtVlyLpB4+#(yvg_4Bxi^j16H^ zsFLV2r_GWCplkeyED`$IFZoLeq^-#a#6xm>I2_| zHk1{SS+5f?0yz_K#E*sgVr{2}3#S|si#Naum_U%-F*r2_=PbFiz)T~0XP-YWq~ExX zgDCB;A!1*penCt?awK1{DZl&J{s~>At!o$Vtv+nWC0`X`YG4K>+ zSPZQ#mRX;xgvzyM_pJSA-tX^FSpp1A2+d5F5D)^mW1G?#uzaz%Fd)vL!y`l| z`+;UV5kzc~XB1t|M zU$@b>78xrgl@Iyh>FytT3m3u^wzAYXUktUxW%;8igljifU=`e|iGpvgZ*=Obc%DX? z+CJCX@PcJTlNv%f1MD|u2a5EZ*U%?vCYe{0>DHu!PuoeCeK zy^!jj7(ESi<%(2JRpp%$R3M$iF;`vfQf#D2-t6Vq-k8s?tC)0)3&nYYi4`w%%APD3%L)I+)P>zD-Rhi{k%Q|rzRH}>89YL zyC1SKgSK9M!a7}Irk}x^VHH1&cTD!-DzbfGNqHDp`?`T|-Hy}lz{(5yKIf}To{w)R zoph5!7Q(2HZ(|fE6R$?0{HR0BJ=7(yGkC-q6I|(PWkwuBMSGep=q2URQs*(WF`UJa#^NiPCv<3xCbCv+_BqJ=C1i|=}C z$PbmT-r->LqrxBB2b{{!oR_SZpjOsi)l^LGm#DMU3OjxtI0)k{W2G95s+c}?3^q6h zI&0mFfjNeP>xoFiEAlS9g92-#?Snc1r7CwJuQnrm2FE-?=g9%hqOO6jXcpg4CSOnf zyqF#Rkx$F082gA|b_J0zwwOa3l5FKFBmkjJ;;Ga6@bIJa!`2ROdD{oEkWhTe5o{NE z9+q)cdcWlcqLXNT>(u+edg!_3%afStntFEBOY{u*LFw-__h0cBw&~B$zO8gN_q|2Tk_Fv3?3a;-a>d}T-zG^JJ!Kd?N%VuGTMxsG)^-b3uNGaHSOmj4k za1td``P|W{!pO#;n*Ooi8<2p}nYw?@*+2?(l}5IpF;n=)9Zpt5a<*MFa7acR}2>u_|rOmf2j^l{%W?n;qX^ zuqa=f3I0L_8XT1H7X^m%Y;efLv6*>0qwXxB30Yz91AZt$@xf5Vw`-BPb_^!ko@BqX4Tx$l z-LPcih}ZW$~e$Q=kS`9PvznXW76;XsUT|%>Zo|V#xR0 z!FHYvzRyKXC764n0;)FsGzI*{Rka;g9M8nO5*tO3OLdO?k8gQLKMEakr{ES<3ASgy zz*AM7NX1vVw1=9bQMWsL*dWmEbbruJ#Jx8;>na90il@qPM?uT+sEDLS6NGZpLlwjv z=POx|Q||^fj}%k{pSf<)Z+&^C0(I~uy0beH@NV9mrKHO*L|gaUG_ZJvA$1CvCf0ff z*u#9_3aHRqh_)l}F#P0k;tgun%pG3Xv25vwq9ddde)(6BZpL9}!HeZQV~?eg;X>Gp z`Md{7qzCB-1TGRs@f?V|S{BWI1cPc=Pae=T``X)DjHDy>#_%%?_?tody%Z;SHr~-v zzMvss8o*3E4gq534Jr6aBO1e-c^!5`B0n98XSri z*YhrFp+l;c%z;6P^wI`mPmkzqTh76VD>~ENQpyaPv`)zaTVG_)v)4A-Yh3ocn$LQj zJT_bF=<3#YQK>(aVR%k@UI&flx3NUv;Ejv8gfdTymKiiz9eqH}x*7UtXn7ks)H*8D za@R~ReRpU3;b=SHo(oY6oUads6nYL@qta1kl>Skmn^yD!1l&b-<`&!qEo+r8(a_;= zv?~fUp1I3icdrQzom<9*@uXUm)GvWPx8GAj{NssaFh38;^@zhv3JZEmu9UuTG9mlh(@eWkVys;A$d3y~VLg!U z$n&co%6;YXQhAjjY<*G2c0UW-HNCl3`3^H2nN87i1vceNZ#1rz;yBTl4#l9oD}S}@ zGH?U?`g7*R53NOb1q=x22IGJ09e+NfoW2{g2YAQMTS^9=0ix2$+KQTt4{Kaw5#*}0 zX6``(gI}P)G=SnNG7JgQgb@#MuE54>4D&MHYE>8kdU%3X@miMBQ3BgM7jm@_~>Wri4~uGB3k^F;_(~Q!bb83MAiA> zG)vG#{*jyJ4V93-uaNl29D&<6Pgp_rOQ`Rz1_GR3nK-IXCZu|=d!wZ2`=dbS1%lxb zwe9q^yxo)Vu|>ixm~691HgsFENc6f04A!!h2G0x&YU{KN>ZV++O3Y1+OUVWTJ`%gh zpuqsQ=u@ld3oOPKl*h*CW53ta|3=OML%eL3M94h&(=!gK%RIz$9Uw1R@UpGjBY_JV z_*yoge_EE6!?dNU}VUQ$iFw!6jCIBl{_F6v(k#Nm<08m zS2)Iy=xI}KIKbWIovBMT>s#Dv%BO!prJ1rHGB{jON`Zdv3^uA>&})W`1EZD|GhrerUg@*qUkV8SROAgq> z!gu;Hy<$WmEElTxa&>;+<8wgJK;q^l%_A^9Hh>Q}>&1I9dW}{zQ=P3!tJfQCr?Ps4R$%$tOXv7`WH3H9LCs9?%2btY zXRFvz`^U+V6P>IReMZN3Nc?*4u7TNK?74@YTpT#_l$}vTc`q=CYT8K2#w&W=WAyWP zqzW&vng@3$1Yho?Yxy6VGvXe|*j1pd4$w0gIHHMdNaNmwIUgeOLeOn|t7Wj?D2N=( z9R27DX}!s+>EQPDZA!%euWIKz^d5Rl16SXJ&rOW`)-tWA&Itt&RHiNF~{c(gX#8!#A&jP@x)3ZpcUC?m#F~4#l*1V zF!V|gB|0CH9VS@m>LbS)&d)LijPU)boOUCyYTc^F!2^x6hn0D=^CIU_QZCc94Ph4p zWoyUQlwS-NA%T%OfL$OrQ-t_F5u$g*>XT-GK_<@MSvAkwmhF5xCHV|-v;ey32YG{s z?qRV(xYagP|8e@|f`1_XW*>w2QvkUC2~`jo_d1VR?@dU_@I_21Z4~sg&_xOrDsZj( zB{2h`mDKcXe6|RLQwMUpBgG-Ws|ga zVl(c~z)Mn^zDRUUbwCcN+jK5m(N9`6V{tT+XB~}OZv>;9BWY{Z@EoCT+0xYp=NRJI zg~xDL?&+Tnz0K%K`}*u&U4*&EnL8|Rq*j+*YvwP$RDR;-oRMid;2*p+(;a_e{Ec|{ zC8%HW2IPx;9lzjiDxSrY>puG%TZ6!+jjYssEFR8rbzy?Q-VW~%U^EFuq5}6^L;@~2 zZR#+7P&Z!qV2bti8Wj=Z-9)&Ng*-4ef+;i!8x;%;$#5VNu<3S4BC1WAWQK$8)$;bf zh!IX_&-t40C3*OJ42d50QsXMvYni>YhV4=&Zx^aB;3-F2p0CkJ+#kV#4-@KZrcudg z5uFwUx4v9|&Cb{Jhc#kKt6l0=T|kMC;s}Gs4sU9*1gdP{M9Fj-mw+ryi^O+gkLtIs zThbM7wWXp#NIlcHYe|W=2yw{oUEm{GPiPtv-ojfqW^S#-jaMEDtE85HQIZHEH@5@B z`Yw8dQ!qjR5e*^EW*|f`tSdNCI+nD==!Y;^yhQ299~2e5B%Q5O$FIvTwQfhsK77jD z`jV|^NJwq&eO;HVrXsG_)wPhg7f*4uJ204BpN$T#-53V4Iw|VAu3SXnrhxx`gQ_t< zx_{ZhC`1MoT;vFvd7wXm;O-**`wl?kRtMDy{3hf1WO`ZJ_0?tu(>}R}*yZ){{kIO< zsju9F78BNC!rKdC5)ttWVpqP;3KbX1mi?zH-WDChiOleZ=zmogL?I{pi3)0 z2q*e1Y6Q=L#L6|VCq*dK%igxiC^l;qEA~B8IxQtGjl#_LJj*z=5D*?z^yz50FL4v{ zR6&g~i$Heza@y)ow;Jf+i z$JC?wO<5-831UAcqH~Z?rg4xYmQ2w?ty(|!`@#y6m6&pp%o{Qc>*A^@SeWkt7BF6c z*YQ_q?gct4`Z9#R&e&ugHRZeySBD4jQ7`lOC*;3{foT|Y5cofxUPs!kV^fX7_ZT*f zFDcT3@7I5Xfe&WyEmSWt1r2h`oA;a|B+CI|U@(~m{cr-z;Ctyreb6q;iD)!@Ad8-7 z3aH_aeyQKWE)Hdfi^D9b;Ml$%@E|*(ATQZ43$c?Pl73E%rx?a7J*!rc*M^Z!zC->I z23n(s%D_f>OBtaQ`GYYk&@ zv+CknQ%zV-IN(>YU|v&mQsF_rDd?&|3OB%YmG}zQx^ig-Bw7P&I+cw*|1`1#w3J=} zuYS8imHNaDTjG239v`k7^znv%Z`#XWB1?8}H-4P`oJO8YPj9SxPtR_!(eR{phIYZ~ zi670Ab>B9rwAtSb4Kq4GA3pt$)P0)iG{T?s-9=u)GnYUdsaIuGoCrEMK>7c z^}003X`gq+NZzoynP6zgLQPOhj+El>%@8&p%HccoF6YsD+mTdu=Th8n9GqF)Q!`Bl zmKeClg7Wv5HV+0)jIU6JU;MJ;PvBD!>-d4}jiuAt+m8W0&nVF4g_pgUb!V%&6Zyr;z9YBS{vcR&{Tco`S2SNryDQ9@CS7 zqqor`Wb1-doe_timL^m5pex-xfLq(N|Pl5*v4%8q+o& zTx~DTm-AX{sC{Zj){2u6OQ)oY({l6ycl^QF1sl5&+rL}C#!HU4Nz3iL<*-pnCo7$A ztH!HCy9saYR;uziX`iCKeD@wh!`FLaz2^}w>G18R-e3xk%#McROGeMm;8o@(E&?+% z^D7XaVuZVX5c-O*2bZ!^$||D_C2oW=!Ini`B7nRnM)*0NNR#zJ2R2&v&OCgdZz^L67vM=zK!4(zGc+gPjnEzK!e{*|6A0T~m|8bkhZbulw8_O%u+C}2z+v4)P z>Q0EPt$k$?FU_R02JK4(V@l$_)X-RI@!G^-_?W9XH(C$uLFT1!j zU3&x4zmQN6kJtN`0I&H7m_7XabTA4anWZIcYvX8QxC?1Fm;x zPjqjJ8Iu}@H4;*iKT4oblx)@zo1f3~JaRjIX=!Iu)j!g%vUbMxFym!7MK+nRMHU7g zMtmpQ=sUEDU*<&SP!g5`Z_J~1CnI=$Qa2;f&_hk@(w zBRAQI{138lZo{QBK4(2iXoFF_Q@Y=4;96&{&*rgD)^66Hm zOo+l97?Dnq023U*)q*xOILqKvn`=^eYIGgS>{XZ&wJ1F5TSaB^w>Y`(9_iRR!t>~b zC|CQ|B79F3w7gp%*kLFXxW4}oF+IlDXNfpJAu-8&*R2*%Rq zew^c6*9VErQfYWcvlNGzjiq(5N$@?%^UioR4)Il^X{UGp)pSwb>Gp--F&^1b`IaLn zC!K155o~Ri#nP`1n))?3_4%i;V|6u<`Yk}z(fIR`1Q-Mr=y%l~JOuC86+qt-0Q?-F zoB7l0H+9U7uLZL|Fu_^^va73rf={_er;+9{BJl%bHK57^gqA;C1KEs zfAaxQ5h*~!2!5#$Bp{pdqe8#6?uS_Ok0$%e%f4+U_Ff)vh1vjZ`L%U&02lS2*2x&S z**ZD?kFe^uZ~n4w^=21+EPzYq1n7Ie;!XhW@Sku$v(rR|8B|EI;k9aWd5ZjN|w# zTvPXJvp%49L4Zp^^h+lup#NtaBg0?u7I%f!$^o1o;u_jtcoR&2!TZDb|DWFcyRPWZ zhWvZ<9P__I|GjVdBl=gbwZigm(40R^@t5QJkK|wN@s;)8kbkrC{}ugfrT5tW4f!Ee%YQja1pdDu|GdHf<}`i|zdxG* zDb?-lhohVWL{Yzbwq>Dz#xefoEp?uiv$6`n%R>J%uTt#)!u!j6>K|8p{2G8%Nd3mM zHZis^ptrC#Ff(DKv$HY#a|{2tbmG@gF{cvv=Z)%@bs~Q*{zqN+4~50=uhgpFivKk6 zA6Hxa8SzJHGvV*$0q?l$enI@f&;R{!_@jXLSDHut-_ibl2>emt`YUa_;qPdF@3{Ub z8vT{_rs?l!e?N%+D5Lz92HX61w7+*ye_uE8mcOI@{V@EaB4`zaqMa{u}X^q2M3q5Pn6pkN-E~FT=k- nJ|zB_Z20|kF$4TpQx6K#;DBcUARv0cKPdztAeA2qZJ_@TZd_xY literal 0 HcmV?d00001