diff --git a/labworks/LW3/2.png b/labworks/LW3/2.png
new file mode 100644
index 0000000..f7058b7
Binary files /dev/null and b/labworks/LW3/2.png differ
diff --git a/labworks/LW3/7.png b/labworks/LW3/7.png
new file mode 100644
index 0000000..f284c95
Binary files /dev/null and b/labworks/LW3/7.png differ
diff --git a/labworks/LW3/best_mnist_model.keras b/labworks/LW3/best_mnist_model.keras
new file mode 100644
index 0000000..1c0d657
Binary files /dev/null and b/labworks/LW3/best_mnist_model.keras differ
diff --git a/labworks/LW3/images/1.png b/labworks/LW3/images/1.png
new file mode 100644
index 0000000..2749001
Binary files /dev/null and b/labworks/LW3/images/1.png differ
diff --git a/labworks/LW3/images/2.png b/labworks/LW3/images/2.png
new file mode 100644
index 0000000..ae851e6
Binary files /dev/null and b/labworks/LW3/images/2.png differ
diff --git a/labworks/LW3/images/3.png b/labworks/LW3/images/3.png
new file mode 100644
index 0000000..6f2b698
Binary files /dev/null and b/labworks/LW3/images/3.png differ
diff --git a/labworks/LW3/images/4.png b/labworks/LW3/images/4.png
new file mode 100644
index 0000000..e32498e
Binary files /dev/null and b/labworks/LW3/images/4.png differ
diff --git a/labworks/LW3/images/5.png b/labworks/LW3/images/5.png
new file mode 100644
index 0000000..b89860d
Binary files /dev/null and b/labworks/LW3/images/5.png differ
diff --git a/labworks/LW3/images/6.png b/labworks/LW3/images/6.png
new file mode 100644
index 0000000..e232523
Binary files /dev/null and b/labworks/LW3/images/6.png differ
diff --git a/labworks/LW3/images/7.png b/labworks/LW3/images/7.png
new file mode 100644
index 0000000..8a8b00e
Binary files /dev/null and b/labworks/LW3/images/7.png differ
diff --git a/labworks/LW3/images/8.png b/labworks/LW3/images/8.png
new file mode 100644
index 0000000..8065931
Binary files /dev/null and b/labworks/LW3/images/8.png differ
diff --git a/labworks/LW3/images/9.png b/labworks/LW3/images/9.png
new file mode 100644
index 0000000..ac29a72
Binary files /dev/null and b/labworks/LW3/images/9.png differ
diff --git a/labworks/LW3/lab3_new.ipynb b/labworks/LW3/lab3_new.ipynb
new file mode 100644
index 0000000..788f754
--- /dev/null
+++ b/labworks/LW3/lab3_new.ipynb
@@ -0,0 +1,1680 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "oZs0KGcz01BY"
+      },
+      "source": [
+        "## Задание 1"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "gz18QPRz03Ec"
+      },
+      "source": [
+        "### 1) Подготовка рабочей среды и импорт библиотек\n",
+        "\n",
+        "Инициализируем рабочую среду Google Colab и подключаем необходимые библиотеки для работы с нейронными сетями и обработки данных."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 1,
+      "metadata": {
+        "id": "mr9IszuQ1ANG"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "2025-12-07 19:27:47.288122: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
+            "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Подключение необходимых библиотек и модулей\n",
+        "import os\n",
+        "\n",
+        "\n",
+        "from tensorflow import keras\n",
+        "from tensorflow.keras import layers\n",
+        "from tensorflow.keras.models import Sequential\n",
+        "import matplotlib.pyplot as plt\n",
+        "import numpy as np\n",
+        "from sklearn.metrics import classification_report, confusion_matrix\n",
+        "from sklearn.metrics import ConfusionMatrixDisplay"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "FFRtE0TN1AiA"
+      },
+      "source": [
+        "### 2) Загрузка датасета MNIST\n",
+        "\n",
+        "Загружаем стандартный набор данных MNIST, который содержит изображения рукописных цифр от 0 до 9 с соответствующими метками."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 2,
+      "metadata": {
+        "id": "Ixw5Sp0_1A-w"
+      },
+      "outputs": [],
+      "source": [
+        "# Импорт и загрузка датасета MNIST\n",
+        "from keras.datasets import mnist\n",
+        "(X_train, y_train), (X_test, y_test) = mnist.load_data()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "aCo_lUXl1BPV"
+      },
+      "source": [
+        "### 3) Разделение данных на обучающую и тестовую выборки\n",
+        "\n",
+        "Производим собственное разбиение датасета в соотношении 60 000:10 000. Для воспроизводимости результатов используем параметр random_state = 3 (вычисляется как 4k - 1, где k = 1 - номер нашей бригады)."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 3,
+      "metadata": {
+        "id": "BrSjcpEe1BeV"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Shape of X train: (60000, 28, 28)\n",
+            "Shape of y train: (60000,)\n",
+            "Shape of X test: (10000, 28, 28)\n",
+            "Shape of y test: (10000,)\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Создание собственного разбиения датасета\n",
+        "from sklearn.model_selection import train_test_split\n",
+        "\n",
+        "# Объединение исходных обучающей и тестовой выборок в единый набор\n",
+        "X = np.concatenate((X_train, X_test))\n",
+        "y = np.concatenate((y_train, y_test))\n",
+        "\n",
+        "# Разделение на обучающую и тестовую выборки согласно заданию\n",
+        "X_train, X_test, y_train, y_test = train_test_split(X, y,\n",
+        "                                                    test_size = 10000,\n",
+        "                                                    train_size = 60000,\n",
+        "                                                    random_state = 3)\n",
+        "# Вывод размерностей полученных массивов\n",
+        "print('Shape of X train:', X_train.shape)\n",
+        "print('Shape of y train:', y_train.shape)\n",
+        "print('Shape of X test:', X_test.shape)\n",
+        "print('Shape of y test:', y_test.shape)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "4hclnNaD1BuB"
+      },
+      "source": [
+        "### 4) Предобработка данных\n",
+        "\n",
+        "Выполняем нормализацию пикселей изображений (приведение к диапазону [0, 1]) и преобразование меток в формат one-hot encoding для корректной работы с категориальной функцией потерь."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 4,
+      "metadata": {
+        "id": "xJH87ISq1B9h"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Shape of transformed X train: (60000, 28, 28, 1)\n",
+            "Shape of transformed X test: (10000, 28, 28, 1)\n",
+            "Shape of transformed y train: (60000, 10)\n",
+            "Shape of transformed y test: (10000, 10)\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Определение параметров данных и модели\n",
+        "num_classes = 10\n",
+        "input_shape = (28, 28, 1)\n",
+        "\n",
+        "# Нормализация значений пикселей: приведение к диапазону [0, 1]\n",
+        "X_train = X_train / 255\n",
+        "X_test = X_test / 255\n",
+        "\n",
+        "# Расширяем размерность входных данных, чтобы каждое изображение имело\n",
+        "# размерность (высота, ширина, количество каналов)\n",
+        "\n",
+        "X_train = np.expand_dims(X_train, -1)\n",
+        "X_test = np.expand_dims(X_test, -1)\n",
+        "print('Shape of transformed X train:', X_train.shape)\n",
+        "print('Shape of transformed X test:', X_test.shape)\n",
+        "\n",
+        "# Преобразование меток в формат one-hot encoding\n",
+        "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
+        "y_test = keras.utils.to_categorical(y_test, num_classes)\n",
+        "print('Shape of transformed y train:', y_train.shape)\n",
+        "print('Shape of transformed y test:', y_test.shape)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "7x99O8ig1CLh"
+      },
+      "source": [
+        "### 5) Построение и обучение сверточной нейронной сети\n",
+        "\n",
+        "Создаем архитектуру сверточной нейронной сети с использованием сверточных слоев, пулинга и регуляризации. Обучаем модель на подготовленных данных с выделением части данных для валидации."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 5,
+      "metadata": {
+        "id": "Un561zSH1Cmv"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/keras/src/layers/convolutional/base_conv.py:113: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
+            "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
+          ]
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1mModel: \"sequential\"\u001b[0m\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
+              "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
+              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
+              "│ conv2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">26</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">26</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">13</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">11</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">5</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1600</span>)           │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)             │        <span style=\"color: #00af00; text-decoration-color: #00af00\">16,010</span> │\n",
+              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
+              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
+              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
+              "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m26\u001b[0m, \u001b[38;5;34m26\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │           \u001b[38;5;34m320\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m)    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m13\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m11\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │        \u001b[38;5;34m18,496\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m64\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout (\u001b[38;5;33mDropout\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m64\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1600\u001b[0m)           │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │        \u001b[38;5;34m16,010\u001b[0m │\n",
+              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">34,826</span> (136.04 KB)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m34,826\u001b[0m (136.04 KB)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">34,826</span> (136.04 KB)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m34,826\u001b[0m (136.04 KB)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "# Создание модели сверточной нейронной сети\n",
+        "model = Sequential()\n",
+        "model.add(layers.Conv2D(32, kernel_size=(3, 3), activation=\"relu\", input_shape=input_shape))\n",
+        "model.add(layers.MaxPooling2D(pool_size=(2, 2)))\n",
+        "model.add(layers.Conv2D(64, kernel_size=(3, 3), activation=\"relu\"))\n",
+        "model.add(layers.MaxPooling2D(pool_size=(2, 2)))\n",
+        "model.add(layers.Dropout(0.5))\n",
+        "model.add(layers.Flatten())\n",
+        "model.add(layers.Dense(num_classes, activation=\"softmax\"))\n",
+        "\n",
+        "model.summary()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 6,
+      "metadata": {
+        "id": "q_h8PxkN9m0v"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Epoch 1/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 184ms/step - accuracy: 0.7694 - loss: 0.7610 - val_accuracy: 0.9437 - val_loss: 0.2013\n",
+            "Epoch 2/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 161ms/step - accuracy: 0.9426 - loss: 0.1908 - val_accuracy: 0.9685 - val_loss: 0.1134\n",
+            "Epoch 3/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18s\u001b[0m 173ms/step - accuracy: 0.9609 - loss: 0.1283 - val_accuracy: 0.9747 - val_loss: 0.0851\n",
+            "Epoch 4/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 210ms/step - accuracy: 0.9688 - loss: 0.1022 - val_accuracy: 0.9785 - val_loss: 0.0708\n",
+            "Epoch 5/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 201ms/step - accuracy: 0.9730 - loss: 0.0871 - val_accuracy: 0.9808 - val_loss: 0.0602\n",
+            "Epoch 6/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 200ms/step - accuracy: 0.9758 - loss: 0.0779 - val_accuracy: 0.9823 - val_loss: 0.0547\n",
+            "Epoch 7/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 206ms/step - accuracy: 0.9781 - loss: 0.0707 - val_accuracy: 0.9820 - val_loss: 0.0515\n",
+            "Epoch 8/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 194ms/step - accuracy: 0.9805 - loss: 0.0637 - val_accuracy: 0.9858 - val_loss: 0.0468\n",
+            "Epoch 9/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 191ms/step - accuracy: 0.9813 - loss: 0.0611 - val_accuracy: 0.9865 - val_loss: 0.0419\n",
+            "Epoch 10/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 190ms/step - accuracy: 0.9816 - loss: 0.0574 - val_accuracy: 0.9865 - val_loss: 0.0402\n",
+            "Epoch 11/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 190ms/step - accuracy: 0.9831 - loss: 0.0531 - val_accuracy: 0.9873 - val_loss: 0.0401\n",
+            "Epoch 12/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 194ms/step - accuracy: 0.9840 - loss: 0.0503 - val_accuracy: 0.9880 - val_loss: 0.0367\n",
+            "Epoch 13/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m20s\u001b[0m 190ms/step - accuracy: 0.9846 - loss: 0.0476 - val_accuracy: 0.9882 - val_loss: 0.0372\n",
+            "Epoch 14/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 195ms/step - accuracy: 0.9845 - loss: 0.0479 - val_accuracy: 0.9880 - val_loss: 0.0360\n",
+            "Epoch 15/15\n",
+            "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 194ms/step - accuracy: 0.9852 - loss: 0.0453 - val_accuracy: 0.9888 - val_loss: 0.0330\n"
+          ]
+        },
+        {
+          "data": {
+            "text/plain": [
+              "<keras.src.callbacks.history.History at 0x1456e0b50>"
+            ]
+          },
+          "execution_count": 6,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "# Компиляция и обучение модели\n",
+        "batch_size = 512\n",
+        "epochs = 15\n",
+        "model.compile(loss=\"categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"])\n",
+        "model.fit(X_train, y_train, batch_size=batch_size, epochs=epochs, validation_split=0.1)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "HL2_LVga1C3l"
+      },
+      "source": [
+        "### 6) Оценка качества модели на тестовых данных\n",
+        "\n",
+        "Проводим финальную оценку обученной модели на независимой тестовой выборке, получая значения функции потерь и точности классификации."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 7,
+      "metadata": {
+        "id": "81Cgq8dn9uL6"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.9884 - loss: 0.0409\n",
+            "Loss on test data: 0.04092026501893997\n",
+            "Accuracy on test data: 0.9883999824523926\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Оценка качества работы обученной модели на тестовой выборке\n",
+        "scores = model.evaluate(X_test, y_test)\n",
+        "print('Loss on test data:', scores[0])\n",
+        "print('Accuracy on test data:', scores[1])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "KzrVY1SR1DZh"
+      },
+      "source": [
+        "### 7) Демонстрация работы модели на отдельных примерах\n",
+        "\n",
+        "Визуализируем результаты распознавания для двух тестовых изображений, сравнивая предсказания модели с истинными метками."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 8,
+      "metadata": {
+        "id": "dbfkWjDI1Dp7"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 93ms/step\n",
+            "NN output: [[1.8653100e-06 8.5978480e-10 4.9378517e-08 3.8702552e-11 2.3658897e-05\n",
+            "  1.0921732e-09 9.9997437e-01 5.6489594e-11 7.1016423e-08 2.4439044e-09]]\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Real mark:  6\n",
+            "NN answer:  6\n",
+            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 44ms/step\n",
+            "NN output: [[7.1973699e-12 5.2415072e-09 2.9768824e-08 9.9999547e-01 8.1457769e-14\n",
+            "  2.3912532e-08 1.0659815e-14 1.0085358e-09 6.9545116e-09 4.5778688e-06]]\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Real mark:  3\n",
+            "NN answer:  3\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Визуализация результатов распознавания для двух тестовых изображений\n",
+        "\n",
+        "for n in [3,26]:\n",
+        "  result = model.predict(X_test[n:n+1])\n",
+        "  print('NN output:', result)\n",
+        "\n",
+        "  plt.imshow(X_test[n].reshape(28,28), cmap=plt.get_cmap('gray'))\n",
+        "  plt.show()\n",
+        "  print('Real mark: ', np.argmax(y_test[n]))\n",
+        "  print('NN answer: ', np.argmax(result))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "YgiVGr5_1D3u"
+      },
+      "source": [
+        "### 8) Детальный анализ качества классификации\n",
+        "\n",
+        "Генерируем подробный отчет о качестве классификации и строим матрицу ошибок для визуального анализа работы модели по каждому классу."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 9,
+      "metadata": {
+        "id": "7MqcG_wl1EHI"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step\n",
+            "              precision    recall  f1-score   support\n",
+            "\n",
+            "           0       1.00      0.99      1.00      1001\n",
+            "           1       0.99      1.00      0.99      1143\n",
+            "           2       0.99      0.99      0.99       987\n",
+            "           3       0.99      0.99      0.99      1023\n",
+            "           4       0.99      0.99      0.99       974\n",
+            "           5       1.00      0.98      0.99       907\n",
+            "           6       0.99      0.99      0.99       974\n",
+            "           7       0.98      0.99      0.99      1032\n",
+            "           8       0.98      0.98      0.98      1006\n",
+            "           9       0.98      0.99      0.98       953\n",
+            "\n",
+            "    accuracy                           0.99     10000\n",
+            "   macro avg       0.99      0.99      0.99     10000\n",
+            "weighted avg       0.99      0.99      0.99     10000\n",
+            "\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 2 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "# Получение истинных и предсказанных меток для всех тестовых данных\n",
+        "true_labels = np.argmax(y_test, axis=1)\n",
+        "\n",
+        "predicted_labels = np.argmax(model.predict(X_test), axis=1)\n",
+        "\n",
+        "# Вывод подробного отчета о качестве классификации\n",
+        "print(classification_report(true_labels, predicted_labels))\n",
+        "# Построение и визуализация матрицы ошибок\n",
+        "conf_matrix = confusion_matrix(true_labels, predicted_labels)\n",
+        "\n",
+        "display = ConfusionMatrixDisplay(confusion_matrix=conf_matrix)\n",
+        "display.plot()\n",
+        "plt.show()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "amaspXGW1EVy"
+      },
+      "source": [
+        "### 9) Тестирование на собственных изображениях\n",
+        "\n",
+        "Загружаем и обрабатываем собственные изображения цифр, созданные ранее, и проверяем способность модели их корректно распознавать."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 10,
+      "metadata": {
+        "id": "ktWEeqWd1EyF"
+      },
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 45ms/step\n",
+            "I think it's 2\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step\n",
+            "I think it's 7\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Загрузка и обработка собственных изображений\n",
+        "from PIL import Image\n",
+        "\n",
+        "for name_image in ['2.png', '7.png']:\n",
+        "  file_data = Image.open(name_image)\n",
+        "  file_data = file_data.convert('L') # перевод в градации серого\n",
+        "  test_img = np.array(file_data)\n",
+        "\n",
+        "  # вывод собственного изображения\n",
+        "  plt.imshow(test_img, cmap=plt.get_cmap('gray'))\n",
+        "  plt.show()\n",
+        "\n",
+        "  # предобработка\n",
+        "  test_img = test_img / 255\n",
+        "  test_img = np.reshape(test_img, (1,28,28,1))\n",
+        "\n",
+        "  # распознавание\n",
+        "  result = model.predict(test_img)\n",
+        "  print('I think it\\'s', np.argmax(result))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "mgrihPd61E8w"
+      },
+      "source": [
+        "### 10) Сравнение с моделью из предыдущей лабораторной работы\n",
+        "\n",
+        "Загружаем сохраненную полносвязную нейронную сеть из лабораторной работы №1 и оцениваем ее производительность на тех же тестовых данных для последующего сравнения."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 11,
+      "metadata": {
+        "id": "DblXqn3l1FL2"
+      },
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1mModel: \"sequential\"\u001b[0m\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
+              "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
+              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
+              "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">7,850</span> │\n",
+              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
+              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
+              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
+              "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │         \u001b[38;5;34m7,850\u001b[0m │\n",
+              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">7,852</span> (30.68 KB)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m7,852\u001b[0m (30.68 KB)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">7,850</span> (30.66 KB)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m7,850\u001b[0m (30.66 KB)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Optimizer params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">2</span> (12.00 B)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Optimizer params: \u001b[0m\u001b[38;5;34m2\u001b[0m (12.00 B)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "model_lr1 = keras.models.load_model(\"best_mnist_model.keras\")\n",
+        "\n",
+        "model_lr1.summary()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 12,
+      "metadata": {
+        "id": "0ki8fhJrEyEt"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Shape of transformed X train: (60000, 784)\n",
+            "Shape of transformed X train: (10000, 784)\n",
+            "Shape of transformed y train: (60000, 10)\n",
+            "Shape of transformed y test: (10000, 10)\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Подготовка данных для полносвязной сети (преобразование изображений в векторы)\n",
+        "X_train, X_test, y_train, y_test = train_test_split(X, y,\n",
+        "                                                    test_size = 10000,\n",
+        "                                                    train_size = 60000,\n",
+        "                                                    random_state = 3)\n",
+        "num_pixels = X_train.shape[1] * X_train.shape[2]\n",
+        "X_train = X_train.reshape(X_train.shape[0], num_pixels) / 255\n",
+        "X_test = X_test.reshape(X_test.shape[0], num_pixels) / 255\n",
+        "print('Shape of transformed X train:', X_train.shape)\n",
+        "print('Shape of transformed X train:', X_test.shape)\n",
+        "\n",
+        "# Преобразование меток в формат one-hot encoding\n",
+        "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
+        "y_test = keras.utils.to_categorical(y_test, num_classes)\n",
+        "print('Shape of transformed y train:', y_train.shape)\n",
+        "print('Shape of transformed y test:', y_test.shape)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 13,
+      "metadata": {
+        "id": "0Yj0fzLNE12k"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m 34/313\u001b[0m \u001b[32m━━\u001b[0m\u001b[37m━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.9142 - loss: 0.2983    "
+          ]
+        },
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
+            "I0000 00:00:1765125204.514834  271959 service.cc:145] XLA service 0x7f89bb2d4be0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n",
+            "I0000 00:00:1765125204.514897  271959 service.cc:153]   StreamExecutor device (0): Host, Default Version\n",
+            "2025-12-07 19:33:24.515300: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
+            "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
+            "2025-12-07 19:33:24.542060: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:268] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
+            "I0000 00:00:1765125204.640642  271959 device_compiler.h:188] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n"
+          ]
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.9233 - loss: 0.2863\n",
+            "Loss on test data: 0.28625616431236267\n",
+            "Accuracy on test data: 0.92330002784729\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Оценка качества работы обученной модели на тестовой выборке\n",
+        "scores = model_lr1.evaluate(X_test, y_test)\n",
+        "print('Loss on test data:', scores[0])\n",
+        "print('Accuracy on test data:', scores[1])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "MsM3ew3d1FYq"
+      },
+      "source": [
+        "### 11) Сравнительный анализ моделей\n",
+        "\n",
+        "Сравниваем сверточную нейронную сеть с полносвязной сетью по ключевым показателям: количеству параметров, времени обучения и качеству классификации."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "xxFO4CXbIG88"
+      },
+      "source": [
+        "Таблица1:"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "xvoivjuNFlEf"
+      },
+      "source": [
+        "| Модель   | Количество настраиваемых параметров | Количество эпох обучения | Качество классификации тестовой выборки |\n",
+        "|----------|-------------------------------------|---------------------------|-----------------------------------------|\n",
+        "| Сверточная | 34 826                              | 15                        | accuracy: 0.988; loss: 0.041                                   |\n",
+        "| Полносвязная | 7 852                              | 50                        | accuracy: 0.923; loss: 0.286                                  |\n"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "YctF8h_sIB-P"
+      },
+      "source": [
+        "**Выводы:**\n",
+        "\n",
+        "На основе проведенного анализа можно заключить, что сверточная нейронная сеть демонстрирует существенные преимущества перед полносвязной сетью при решении задач распознавания изображений:\n",
+        "\n",
+        "1. **Эффективность параметров**: Сверточная сеть имеет больше параметров (34 826 против 7 852), но при этом показывает значительно лучшие результаты, что говорит о более эффективном использовании параметров для извлечения пространственных признаков.\n",
+        "\n",
+        "2. **Скорость обучения**: Для достижения высокого качества сверточной сети требуется в 3.3 раза меньше эпох обучения (15 против 50), что существенно сокращает время обучения.\n",
+        "\n",
+        "3. **Точность классификации**: Сверточная сеть показывает более высокую точность (98.8% против 92.3%) и значительно меньшую функцию потерь (0.041 против 0.286). Разница в точности составляет 6.5%, что является существенным улучшением.\n",
+        "\n",
+        "4. **Обобщающая способность**: Сверточная сеть демонстрирует лучшую способность к обобщению, что видно из более низкой функции потерь на тестовых данных.\n",
+        "\n",
+        "Эти результаты подтверждают, что архитектура сверточных сетей, учитывающая пространственную структуру изображений через операции свертки и пулинга, является более подходящим выбором для задач компьютерного зрения, несмотря на большее количество параметров."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "wCLHZPGB1F1y"
+      },
+      "source": [
+        "## Задание 2"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "DUOYls124TT8"
+      },
+      "source": [
+        "### В новом блокноте выполнили п. 2–8 задания 1, изменив набор данных MNIST на CIFAR-10, содержащий размеченные цветные изображения объектов, разделенные на 10 классов.  \n",
+        "### При этом:\n",
+        "### - в п. 3 разбиение данных на обучающие и тестовые произвели в соотношении 50 000:10 000\n",
+        "### - после разбиения данных (между п. 3 и 4) вывели 25 изображений из обучающей выборки с подписями классов\n",
+        "### - в п. 7 одно из тестовых изображений должно распознаваться корректно, а другое – ошибочно.   "
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "XDStuSpEJa8o"
+      },
+      "source": [
+        "### 1) Загрузка датасета CIFAR-10\n",
+        "\n",
+        "Загружаем набор данных CIFAR-10, который содержит цветные изображения размером 32x32 пикселя, разделенные на 10 классов: самолет, автомобиль, птица, кошка, олень, собака, лягушка, лошадь, корабль, грузовик."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 14,
+      "metadata": {
+        "id": "y0qK7eKL4Tjy"
+      },
+      "outputs": [],
+      "source": [
+        "# Импорт и загрузка датасета MNIST\n",
+        "from keras.datasets import cifar10\n",
+        "\n",
+        "(X_train, y_train), (X_test, y_test) = cifar10.load_data()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "wTHiBy-ZJ5oh"
+      },
+      "source": [
+        "### 2) Разделение данных на обучающую и тестовую выборки\n",
+        "\n",
+        "Создаем собственное разбиение датасета CIFAR-10 в соотношении 50 000:10 000. Используем random_state = 3 для воспроизводимости результатов (k = 1 - номер нашей бригады)."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 15,
+      "metadata": {
+        "id": "DlnFbQogKD2v"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Shape of X train: (50000, 32, 32, 3)\n",
+            "Shape of y train: (50000, 1)\n",
+            "Shape of X test: (10000, 32, 32, 3)\n",
+            "Shape of y test: (10000, 1)\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Создание собственного разбиения датасета\n",
+        "\n",
+        "# Объединение исходных обучающей и тестовой выборок в единый набор\n",
+        "X = np.concatenate((X_train, X_test))\n",
+        "y = np.concatenate((y_train, y_test))\n",
+        "\n",
+        "# Разделение на обучающую и тестовую выборки согласно заданию\n",
+        "X_train, X_test, y_train, y_test = train_test_split(X, y,\n",
+        "                                                    test_size = 10000,\n",
+        "                                                    train_size = 50000,\n",
+        "                                                    random_state = 3)\n",
+        "# Вывод размерностей полученных массивов\n",
+        "print('Shape of X train:', X_train.shape)\n",
+        "print('Shape of y train:', y_train.shape)\n",
+        "print('Shape of X test:', X_test.shape)\n",
+        "print('Shape of y test:', y_test.shape)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "pj3bMaz1KZ3a"
+      },
+      "source": [
+        "### Визуализация примеров из обучающей выборки\n",
+        "\n",
+        "Отображаем сетку из 25 изображений из обучающей выборки с подписями соответствующих классов для визуального ознакомления с данными."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 16,
+      "metadata": {
+        "id": "TW8D67KEKhVE"
+      },
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 1000x1000 with 25 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',\n",
+        "               'dog', 'frog', 'horse', 'ship', 'truck']\n",
+        "\n",
+        "plt.figure(figsize=(10,10))\n",
+        "for i in range(25):\n",
+        "    plt.subplot(5,5,i+1)\n",
+        "    plt.xticks([])\n",
+        "    plt.yticks([])\n",
+        "    plt.grid(False)\n",
+        "    plt.imshow(X_train[i])\n",
+        "    plt.xlabel(class_names[y_train[i][0]])\n",
+        "plt.show()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "d3TPr2w1KQTK"
+      },
+      "source": [
+        "### 3) Предобработка данных CIFAR-10\n",
+        "\n",
+        "Нормализуем значения пикселей и преобразуем метки в формат one-hot encoding для работы с категориальной функцией потерь."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 17,
+      "metadata": {
+        "id": "iFDpxEauLZ8j"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Shape of transformed X train: (50000, 32, 32, 3)\n",
+            "Shape of transformed X test: (10000, 32, 32, 3)\n",
+            "Shape of transformed y train: (50000, 10)\n",
+            "Shape of transformed y test: (10000, 10)\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Определение параметров данных и модели\n",
+        "num_classes = 10\n",
+        "input_shape = (32, 32, 3)\n",
+        "\n",
+        "# Нормализация значений пикселей: приведение к диапазону [0, 1]\n",
+        "X_train = X_train / 255\n",
+        "X_test = X_test / 255\n",
+        "\n",
+        "print('Shape of transformed X train:', X_train.shape)\n",
+        "print('Shape of transformed X test:', X_test.shape)\n",
+        "\n",
+        "# Преобразование меток в формат one-hot encoding\n",
+        "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
+        "y_test = keras.utils.to_categorical(y_test, num_classes)\n",
+        "print('Shape of transformed y train:', y_train.shape)\n",
+        "print('Shape of transformed y test:', y_test.shape)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ydNITXptLeGT"
+      },
+      "source": [
+        "### 4) Построение и обучение сверточной сети для CIFAR-10\n",
+        "\n",
+        "Создаем более сложную архитектуру сверточной сети с использованием батч-нормализации и нескольких блоков свертки для работы с цветными изображениями."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 18,
+      "metadata": {
+        "id": "YhAD5CllLlv7"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "/Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages/keras/src/layers/convolutional/base_conv.py:113: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
+            "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
+          ]
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_1\"</span>\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1mModel: \"sequential_1\"\u001b[0m\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
+              "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
+              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
+              "│ conv2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">896</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
+              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │         <span style=\"color: #00af00; text-decoration-color: #00af00\">9,248</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_1           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
+              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)     │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_2           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │\n",
+              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │        <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_3           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)     │           <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │\n",
+              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)       │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)      │        <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_4           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)      │           <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
+              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)      │       <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_5           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)      │           <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
+              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)      │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ flatten_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">2048</span>)           │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)            │       <span style=\"color: #00af00; text-decoration-color: #00af00\">262,272</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)            │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,290</span> │\n",
+              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
+              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
+              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
+              "│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │           \u001b[38;5;34m896\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │           \u001b[38;5;34m128\u001b[0m │\n",
+              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_3 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │         \u001b[38;5;34m9,248\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_1           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │           \u001b[38;5;34m128\u001b[0m │\n",
+              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d_2 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m32\u001b[0m)     │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │        \u001b[38;5;34m18,496\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_2           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │           \u001b[38;5;34m256\u001b[0m │\n",
+              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │        \u001b[38;5;34m36,928\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_3           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m64\u001b[0m)     │           \u001b[38;5;34m256\u001b[0m │\n",
+              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d_3 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m64\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout_2 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m64\u001b[0m)       │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_6 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m)      │        \u001b[38;5;34m73,856\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_4           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m)      │           \u001b[38;5;34m512\u001b[0m │\n",
+              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ conv2d_7 (\u001b[38;5;33mConv2D\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m)      │       \u001b[38;5;34m147,584\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ batch_normalization_5           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m128\u001b[0m)      │           \u001b[38;5;34m512\u001b[0m │\n",
+              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ max_pooling2d_4 (\u001b[38;5;33mMaxPooling2D\u001b[0m)  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m128\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout_3 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m128\u001b[0m)      │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ flatten_1 (\u001b[38;5;33mFlatten\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2048\u001b[0m)           │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)            │       \u001b[38;5;34m262,272\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dropout_4 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)            │             \u001b[38;5;34m0\u001b[0m │\n",
+              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+              "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │         \u001b[38;5;34m1,290\u001b[0m │\n",
+              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">552,362</span> (2.11 MB)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m552,362\u001b[0m (2.11 MB)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">551,466</span> (2.10 MB)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m551,466\u001b[0m (2.10 MB)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "text/html": [
+              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">896</span> (3.50 KB)\n",
+              "</pre>\n"
+            ],
+            "text/plain": [
+              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m896\u001b[0m (3.50 KB)\n"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "# Создание модели сверточной нейронной сети\n",
+        "model = Sequential()\n",
+        "\n",
+        "# Блок 1\n",
+        "model.add(layers.Conv2D(32, (3, 3), padding=\"same\",\n",
+        "                        activation=\"relu\", input_shape=input_shape))\n",
+        "model.add(layers.BatchNormalization())\n",
+        "model.add(layers.Conv2D(32, (3, 3), padding=\"same\", activation=\"relu\"))\n",
+        "model.add(layers.BatchNormalization())\n",
+        "model.add(layers.MaxPooling2D((2, 2)))\n",
+        "model.add(layers.Dropout(0.25))\n",
+        "\n",
+        "# Блок 2\n",
+        "model.add(layers.Conv2D(64, (3, 3), padding=\"same\", activation=\"relu\"))\n",
+        "model.add(layers.BatchNormalization())\n",
+        "model.add(layers.Conv2D(64, (3, 3), padding=\"same\", activation=\"relu\"))\n",
+        "model.add(layers.BatchNormalization())\n",
+        "model.add(layers.MaxPooling2D((2, 2)))\n",
+        "model.add(layers.Dropout(0.25))\n",
+        "\n",
+        "# Блок 3\n",
+        "model.add(layers.Conv2D(128, (3, 3), padding=\"same\", activation=\"relu\"))\n",
+        "model.add(layers.BatchNormalization())\n",
+        "model.add(layers.Conv2D(128, (3, 3), padding=\"same\", activation=\"relu\"))\n",
+        "model.add(layers.BatchNormalization())\n",
+        "model.add(layers.MaxPooling2D((2, 2)))\n",
+        "model.add(layers.Dropout(0.4))\n",
+        "\n",
+        "model.add(layers.Flatten())\n",
+        "model.add(layers.Dense(128, activation='relu'))\n",
+        "model.add(layers.Dropout(0.5))\n",
+        "model.add(layers.Dense(num_classes, activation=\"softmax\"))\n",
+        "\n",
+        "\n",
+        "model.summary()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 19,
+      "metadata": {
+        "id": "3otvqMjjOdq5"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Epoch 1/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m214s\u001b[0m 295ms/step - accuracy: 0.3409 - loss: 1.8087 - val_accuracy: 0.4302 - val_loss: 1.6950\n",
+            "Epoch 2/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m210s\u001b[0m 299ms/step - accuracy: 0.5008 - loss: 1.3835 - val_accuracy: 0.6096 - val_loss: 1.1257\n",
+            "Epoch 3/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m209s\u001b[0m 297ms/step - accuracy: 0.5871 - loss: 1.1704 - val_accuracy: 0.6310 - val_loss: 1.1089\n",
+            "Epoch 4/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m205s\u001b[0m 291ms/step - accuracy: 0.6421 - loss: 1.0381 - val_accuracy: 0.6666 - val_loss: 0.9580\n",
+            "Epoch 5/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m195s\u001b[0m 277ms/step - accuracy: 0.6788 - loss: 0.9402 - val_accuracy: 0.7004 - val_loss: 0.8947\n",
+            "Epoch 6/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m160s\u001b[0m 227ms/step - accuracy: 0.7065 - loss: 0.8630 - val_accuracy: 0.6856 - val_loss: 0.9637\n",
+            "Epoch 7/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m159s\u001b[0m 226ms/step - accuracy: 0.7256 - loss: 0.8078 - val_accuracy: 0.7604 - val_loss: 0.6995\n",
+            "Epoch 8/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m159s\u001b[0m 225ms/step - accuracy: 0.7458 - loss: 0.7463 - val_accuracy: 0.7388 - val_loss: 0.7766\n",
+            "Epoch 9/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m160s\u001b[0m 227ms/step - accuracy: 0.7601 - loss: 0.7104 - val_accuracy: 0.7420 - val_loss: 0.7523\n",
+            "Epoch 10/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m159s\u001b[0m 226ms/step - accuracy: 0.7770 - loss: 0.6658 - val_accuracy: 0.7782 - val_loss: 0.6714\n",
+            "Epoch 11/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m196s\u001b[0m 278ms/step - accuracy: 0.7876 - loss: 0.6321 - val_accuracy: 0.7852 - val_loss: 0.6610\n",
+            "Epoch 12/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.7978 - loss: 0.6006 - val_accuracy: 0.8026 - val_loss: 0.5872\n",
+            "Epoch 13/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.8073 - loss: 0.5754 - val_accuracy: 0.7994 - val_loss: 0.5945\n",
+            "Epoch 14/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.8146 - loss: 0.5536 - val_accuracy: 0.7776 - val_loss: 0.6921\n",
+            "Epoch 15/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.8226 - loss: 0.5308 - val_accuracy: 0.8016 - val_loss: 0.6051\n",
+            "Epoch 16/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m173s\u001b[0m 246ms/step - accuracy: 0.8295 - loss: 0.5097 - val_accuracy: 0.8082 - val_loss: 0.6001\n",
+            "Epoch 17/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m162s\u001b[0m 231ms/step - accuracy: 0.8333 - loss: 0.4900 - val_accuracy: 0.8204 - val_loss: 0.5621\n",
+            "Epoch 18/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m175s\u001b[0m 248ms/step - accuracy: 0.8399 - loss: 0.4763 - val_accuracy: 0.8202 - val_loss: 0.5716\n",
+            "Epoch 19/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m199s\u001b[0m 283ms/step - accuracy: 0.8458 - loss: 0.4535 - val_accuracy: 0.8132 - val_loss: 0.5784\n",
+            "Epoch 20/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.8494 - loss: 0.4406 - val_accuracy: 0.8276 - val_loss: 0.5378\n",
+            "Epoch 21/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m177s\u001b[0m 251ms/step - accuracy: 0.8536 - loss: 0.4293 - val_accuracy: 0.8132 - val_loss: 0.5989\n",
+            "Epoch 22/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m174s\u001b[0m 248ms/step - accuracy: 0.8591 - loss: 0.4120 - val_accuracy: 0.8398 - val_loss: 0.5143\n",
+            "Epoch 23/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m183s\u001b[0m 260ms/step - accuracy: 0.8610 - loss: 0.4031 - val_accuracy: 0.8216 - val_loss: 0.5681\n",
+            "Epoch 24/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.8668 - loss: 0.3945 - val_accuracy: 0.8358 - val_loss: 0.5374\n",
+            "Epoch 25/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m180s\u001b[0m 256ms/step - accuracy: 0.8708 - loss: 0.3810 - val_accuracy: 0.8166 - val_loss: 0.6225\n",
+            "Epoch 26/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m181s\u001b[0m 257ms/step - accuracy: 0.8706 - loss: 0.3787 - val_accuracy: 0.8380 - val_loss: 0.5285\n",
+            "Epoch 27/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m182s\u001b[0m 259ms/step - accuracy: 0.8771 - loss: 0.3634 - val_accuracy: 0.8410 - val_loss: 0.5138\n",
+            "Epoch 28/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m177s\u001b[0m 251ms/step - accuracy: 0.8776 - loss: 0.3538 - val_accuracy: 0.8280 - val_loss: 0.5548\n",
+            "Epoch 29/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m168s\u001b[0m 238ms/step - accuracy: 0.8824 - loss: 0.3486 - val_accuracy: 0.8390 - val_loss: 0.5372\n",
+            "Epoch 30/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m173s\u001b[0m 246ms/step - accuracy: 0.8838 - loss: 0.3398 - val_accuracy: 0.8434 - val_loss: 0.4986\n",
+            "Epoch 31/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m183s\u001b[0m 260ms/step - accuracy: 0.8876 - loss: 0.3322 - val_accuracy: 0.8380 - val_loss: 0.5392\n",
+            "Epoch 32/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m185s\u001b[0m 262ms/step - accuracy: 0.8899 - loss: 0.3235 - val_accuracy: 0.8086 - val_loss: 0.6294\n",
+            "Epoch 33/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m191s\u001b[0m 271ms/step - accuracy: 0.8931 - loss: 0.3156 - val_accuracy: 0.8430 - val_loss: 0.5467\n",
+            "Epoch 34/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.8920 - loss: 0.3133 - val_accuracy: 0.8454 - val_loss: 0.5099\n",
+            "Epoch 35/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m181s\u001b[0m 256ms/step - accuracy: 0.8965 - loss: 0.3024 - val_accuracy: 0.8468 - val_loss: 0.5167\n",
+            "Epoch 36/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m180s\u001b[0m 256ms/step - accuracy: 0.8956 - loss: 0.3030 - val_accuracy: 0.8296 - val_loss: 0.5907\n",
+            "Epoch 37/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m181s\u001b[0m 257ms/step - accuracy: 0.8980 - loss: 0.2956 - val_accuracy: 0.8426 - val_loss: 0.5412\n",
+            "Epoch 38/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m181s\u001b[0m 257ms/step - accuracy: 0.9006 - loss: 0.2887 - val_accuracy: 0.8438 - val_loss: 0.5187\n",
+            "Epoch 39/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.9013 - loss: 0.2874 - val_accuracy: 0.8478 - val_loss: 0.5139\n",
+            "Epoch 40/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.9029 - loss: 0.2824 - val_accuracy: 0.8068 - val_loss: 0.6571\n",
+            "Epoch 41/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.9062 - loss: 0.2758 - val_accuracy: 0.8542 - val_loss: 0.5129\n",
+            "Epoch 42/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m180s\u001b[0m 255ms/step - accuracy: 0.9060 - loss: 0.2727 - val_accuracy: 0.8538 - val_loss: 0.4998\n",
+            "Epoch 43/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m187s\u001b[0m 265ms/step - accuracy: 0.9096 - loss: 0.2650 - val_accuracy: 0.8504 - val_loss: 0.4944\n",
+            "Epoch 44/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m179s\u001b[0m 254ms/step - accuracy: 0.9100 - loss: 0.2646 - val_accuracy: 0.8480 - val_loss: 0.5352\n",
+            "Epoch 45/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m175s\u001b[0m 249ms/step - accuracy: 0.9109 - loss: 0.2560 - val_accuracy: 0.8510 - val_loss: 0.5218\n",
+            "Epoch 46/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m177s\u001b[0m 252ms/step - accuracy: 0.9130 - loss: 0.2518 - val_accuracy: 0.8552 - val_loss: 0.4983\n",
+            "Epoch 47/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m181s\u001b[0m 257ms/step - accuracy: 0.9124 - loss: 0.2525 - val_accuracy: 0.8578 - val_loss: 0.5095\n",
+            "Epoch 48/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m177s\u001b[0m 251ms/step - accuracy: 0.9161 - loss: 0.2456 - val_accuracy: 0.8488 - val_loss: 0.5426\n",
+            "Epoch 49/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m174s\u001b[0m 247ms/step - accuracy: 0.9171 - loss: 0.2449 - val_accuracy: 0.8530 - val_loss: 0.5429\n",
+            "Epoch 50/50\n",
+            "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m197s\u001b[0m 280ms/step - accuracy: 0.9195 - loss: 0.2354 - val_accuracy: 0.8454 - val_loss: 0.5250\n"
+          ]
+        },
+        {
+          "data": {
+            "text/plain": [
+              "<keras.src.callbacks.history.History at 0x146a42610>"
+            ]
+          },
+          "execution_count": 19,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "# Компиляция и обучение модели\n",
+        "batch_size = 64\n",
+        "epochs = 50\n",
+        "model.compile(loss=\"categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"])\n",
+        "model.fit(X_train, y_train, batch_size=batch_size, epochs=epochs, validation_split=0.1)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Vv1kUHWTLl9B"
+      },
+      "source": [
+        "### 5) Оценка качества модели на тестовых данных\n",
+        "\n",
+        "Оцениваем финальную производительность обученной модели на тестовой выборке CIFAR-10."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 20,
+      "metadata": {
+        "id": "SaDxydiyLmRX"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 38ms/step - accuracy: 0.8539 - loss: 0.5047\n",
+            "Loss on test data: 0.5046958327293396\n",
+            "Accuracy on test data: 0.8539000153541565\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Оценка качества работы обученной модели на тестовой выборке\n",
+        "scores = model.evaluate(X_test, y_test)\n",
+        "print('Loss on test data:', scores[0])\n",
+        "print('Accuracy on test data:', scores[1])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "OdgEiyUGLmhP"
+      },
+      "source": [
+        "### 6) Демонстрация работы модели на отдельных примерах\n",
+        "\n",
+        "Визуализируем результаты распознавания для двух тестовых изображений: одно должно быть распознано корректно, другое - ошибочно."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 32,
+      "metadata": {
+        "id": "t3yGj1MlLm9H"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 114ms/step\n",
+            "NN output: [[1.1500329e-13 1.3227661e-13 1.2523241e-10 1.1334410e-10 7.6330755e-13\n",
+            "  2.6473241e-14 1.0000000e+00 7.4253257e-18 1.0440019e-13 2.6218085e-13]]\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Real mark:  6\n",
+            "NN answer:  6\n",
+            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 98ms/step\n",
+            "NN output: [[2.0177353e-09 1.4144381e-12 9.2603455e-05 1.9454856e-04 2.1098103e-02\n",
+            "  9.7849804e-01 2.5194169e-05 9.1513859e-05 3.2438701e-11 4.6777016e-10]]\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Real mark:  4\n",
+            "NN answer:  5\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Визуализация результатов распознавания для двух тестовых изображений\n",
+        "\n",
+        "for n in [3,14]:\n",
+        "  result = model.predict(X_test[n:n+1])\n",
+        "  print('NN output:', result)\n",
+        "\n",
+        "  plt.imshow(X_test[n].reshape(32,32,3), cmap=plt.get_cmap('gray'))\n",
+        "  plt.show()\n",
+        "  print('Real mark: ', np.argmax(y_test[n]))\n",
+        "  print('NN answer: ', np.argmax(result))"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "3h6VGDRrLnNC"
+      },
+      "source": [
+        "### 7) Детальный анализ качества классификации CIFAR-10\n",
+        "\n",
+        "Генерируем подробный отчет о качестве классификации и строим матрицу ошибок для анализа работы модели по каждому классу."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 22,
+      "metadata": {
+        "id": "od56oyyzM0nw"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 35ms/step\n",
+            "              precision    recall  f1-score   support\n",
+            "\n",
+            "    airplane       0.84      0.91      0.87      1007\n",
+            "  automobile       0.95      0.91      0.93      1037\n",
+            "        bird       0.83      0.79      0.81      1030\n",
+            "         cat       0.77      0.65      0.70       990\n",
+            "        deer       0.83      0.82      0.82       966\n",
+            "         dog       0.72      0.83      0.77      1009\n",
+            "        frog       0.90      0.89      0.89       972\n",
+            "       horse       0.87      0.89      0.88       991\n",
+            "        ship       0.95      0.92      0.93       990\n",
+            "       truck       0.89      0.93      0.91      1008\n",
+            "\n",
+            "    accuracy                           0.85     10000\n",
+            "   macro avg       0.86      0.85      0.85     10000\n",
+            "weighted avg       0.86      0.85      0.85     10000\n",
+            "\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 600x600 with 2 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "# Получение истинных и предсказанных меток для всех тестовых данных\n",
+        "true_labels = np.argmax(y_test, axis=1)\n",
+        "\n",
+        "predicted_labels = np.argmax(model.predict(X_test), axis=1)\n",
+        "\n",
+        "# Вывод подробного отчета о качестве классификации\n",
+        "print(classification_report(true_labels, predicted_labels, target_names=class_names))\n",
+        "# Построение и визуализация матрицы ошибок\n",
+        "conf_matrix = confusion_matrix(true_labels, predicted_labels)\n",
+        "\n",
+        "fig, ax = plt.subplots(figsize=(6, 6))\n",
+        "disp = ConfusionMatrixDisplay(confusion_matrix=conf_matrix,display_labels=class_names)\n",
+        "disp.plot(ax=ax, xticks_rotation=45)  # поворот подписей по X и приятная палитра\n",
+        "plt.tight_layout()  # чтобы всё влезло\n",
+        "plt.show()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "RF4xK1cxamBc"
+      },
+      "source": [
+        "**Выводы по результатам классификации CIFAR-10:**\n",
+        "\n",
+        "Разработанная сверточная нейронная сеть показала хорошие результаты при классификации цветных изображений из датасета CIFAR-10. Модель достигла точности классификации около 86%, что является достойным результатом для данной задачи, учитывая сложность различения объектов в низком разрешении (32x32 пикселя) и наличие 10 различных классов.\n",
+        "\n",
+        "Использование батч-нормализации и dropout-регуляризации позволило улучшить обобщающую способность модели и предотвратить переобучение. Архитектура с тремя блоками сверточных слоев эффективно извлекает иерархические признаки из изображений, что подтверждается полученными метриками качества."
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": []
+    }
+  ],
+  "metadata": {
+    "accelerator": "GPU",
+    "colab": {
+      "gpuType": "T4",
+      "provenance": []
+    },
+    "kernelspec": {
+      "display_name": "Python 3",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.11.9"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
diff --git a/labworks/LW3/report_new.md b/labworks/LW3/report_new.md
new file mode 100644
index 0000000..f73f6f4
--- /dev/null
+++ b/labworks/LW3/report_new.md
@@ -0,0 +1,612 @@
+# Отчёт по лабораторной работе №3
+
+**Троянов Д.С., Чернов Д.Е. — А-01-22**
+
+---
+
+## Задание 1
+
+### 1) Подготовка рабочей среды и импорт библиотек
+
+Инициализируем рабочую среду и подключаем необходимые библиотеки для работы с нейронными сетями и обработки данных. Также настраиваем SSL для корректной загрузки датасетов.
+
+```python
+# Подключение необходимых библиотек и модулей
+import os
+import ssl
+
+# Обход проблемы с SSL сертификатами на macOS
+ssl._create_default_https_context = ssl._create_unverified_context
+
+# Для работы в Google Colab раскомментируйте следующую строку:
+# os.chdir('/content/drive/MyDrive/Colab Notebooks/is_lab3')
+
+from tensorflow import keras
+from tensorflow.keras import layers
+from tensorflow.keras.models import Sequential
+import matplotlib.pyplot as plt
+import numpy as np
+from sklearn.metrics import classification_report, confusion_matrix
+from sklearn.metrics import ConfusionMatrixDisplay
+```
+
+### 2) Загрузка датасета MNIST
+
+Загружаем стандартный набор данных MNIST, который содержит изображения рукописных цифр от 0 до 9 с соответствующими метками.
+
+```python
+# Импорт и загрузка датасета MNIST
+from keras.datasets import mnist
+(X_train, y_train), (X_test, y_test) = mnist.load_data()
+```
+
+### 3) Разделение данных на обучающую и тестовую выборки
+
+Производим собственное разбиение датасета в соотношении 60 000:10 000. Для воспроизводимости результатов используем параметр random_state = 3 (вычисляется как 4k - 1, где k = 1 - номер нашей бригады). Выводим размерности полученных массивов данных.
+
+```python
+# Создание собственного разбиения датасета
+from sklearn.model_selection import train_test_split
+
+# Объединение исходных обучающей и тестовой выборок в единый набор
+X = np.concatenate((X_train, X_test))
+y = np.concatenate((y_train, y_test))
+
+# Разделение на обучающую и тестовую выборки согласно заданию
+X_train, X_test, y_train, y_test = train_test_split(X, y,
+                                                    test_size = 10000,
+                                                    train_size = 60000,
+                                                    random_state = 3)
+# Вывод размерностей полученных массивов
+print('Shape of X train:', X_train.shape)
+print('Shape of y train:', y_train.shape)
+print('Shape of X test:', X_test.shape)
+print('Shape of y test:', y_test.shape)
+```
+```
+Shape of X train: (60000, 28, 28)
+Shape of y train: (60000,)
+Shape of X test: (10000, 28, 28)
+Shape of y test: (10000,)
+```
+
+### 4) Предобработка данных
+
+Выполняем нормализацию пикселей изображений (приведение к диапазону [0, 1]) и преобразование меток в формат one-hot encoding для корректной работы с категориальной функцией потерь. Выводим размерности предобработанных массивов данных.
+
+```python
+# Определение параметров данных и модели
+num_classes = 10
+input_shape = (28, 28, 1)
+
+# Нормализация значений пикселей: приведение к диапазону [0, 1]
+X_train = X_train / 255
+X_test = X_test / 255
+
+# Добавление размерности канала для корректной работы с Conv2D слоями
+# Преобразование из (высота, ширина) в (высота, ширина, каналы)
+X_train = np.expand_dims(X_train, -1)
+X_test = np.expand_dims(X_test, -1)
+print('Shape of transformed X train:', X_train.shape)
+print('Shape of transformed X test:', X_test.shape)
+
+# Преобразование меток в формат one-hot encoding
+y_train = keras.utils.to_categorical(y_train, num_classes)
+y_test = keras.utils.to_categorical(y_test, num_classes)
+print('Shape of transformed y train:', y_train.shape)
+print('Shape of transformed y test:', y_test.shape)
+```
+```
+Shape of transformed X train: (60000, 28, 28, 1)
+Shape of transformed X test: (10000, 28, 28, 1)
+Shape of transformed y train: (60000, 10)
+Shape of transformed y test: (10000, 10)
+```
+
+### 5) Построение и обучение сверточной нейронной сети
+
+Создаем архитектуру сверточной нейронной сети с использованием сверточных слоев, пулинга и регуляризации. Обучаем модель на подготовленных данных с выделением части данных для валидации. Выводим информацию об архитектуре нейронной сети.
+
+```python
+# Создание модели сверточной нейронной сети
+model = Sequential()
+model.add(layers.Conv2D(32, kernel_size=(3, 3), activation="relu", input_shape=input_shape))
+model.add(layers.MaxPooling2D(pool_size=(2, 2)))
+model.add(layers.Conv2D(64, kernel_size=(3, 3), activation="relu"))
+model.add(layers.MaxPooling2D(pool_size=(2, 2)))
+model.add(layers.Dropout(0.5))
+model.add(layers.Flatten())
+model.add(layers.Dense(num_classes, activation="softmax"))
+
+model.summary()
+```
+**Model: "sequential"**
+| Layer (type)                   | Output Shape        | Param # |
+|--------------------------------|---------------------|--------:|
+| conv2d (Conv2D)                | (None, 26, 26, 32)  |     320 |
+| max_pooling2d (MaxPooling2D)   | (None, 13, 13, 32)  |       0 |
+| conv2d_1 (Conv2D)              | (None, 11, 11, 64)  |  18,496 |
+| max_pooling2d_1 (MaxPooling2D) | (None, 5, 5, 64)    |       0 |
+| dropout (Dropout)              | (None, 5, 5, 64)    |       0 |
+| flatten (Flatten)              | (None, 1600)        |       0 |
+| dense (Dense)                  | (None, 10)          |  16,010 |
+**Total params:** 34,826 (136.04 KB)  
+**Trainable params:** 34,826 (136.04 KB)  
+**Non-trainable params:** 0 (0.00 B)
+
+```python
+# Компиляция и обучение модели
+batch_size = 512
+epochs = 15
+model.compile(loss="categorical_crossentropy", optimizer="adam", metrics=["accuracy"])
+model.fit(X_train, y_train, batch_size=batch_size, epochs=epochs, validation_split=0.1)
+```
+
+### 6) Оценка качества модели на тестовых данных
+
+Проводим финальную оценку обученной модели на независимой тестовой выборке, получая значения функции потерь и точности классификации.
+
+```python
+# Оценка качества работы обученной модели на тестовой выборке
+scores = model.evaluate(X_test, y_test)
+print('Loss on test data:', scores[0])
+print('Accuracy on test data:', scores[1])
+```
+```
+313/313 ━━━━━━━━━━━━━━━━━━━━ 1s 4ms/step - accuracy: 0.9884 - loss: 0.0409
+Loss on test data: 0.04092026501893997
+Accuracy on test data: 0.9883999824523926
+```
+
+### 7) Демонстрация работы модели на отдельных примерах
+
+Визуализируем результаты распознавания для двух тестовых изображений, сравнивая предсказания модели с истинными метками.
+
+```python
+# Визуализация результатов распознавания для двух тестовых изображений
+
+for n in [3,26]:
+  result = model.predict(X_test[n:n+1])
+  print('NN output:', result)
+
+  plt.imshow(X_test[n].reshape(28,28), cmap=plt.get_cmap('gray'))
+  plt.show()
+  print('Real mark: ', np.argmax(y_test[n]))
+  print('NN answer: ', np.argmax(result))
+```
+![MNIST тестовое изображение - цифра 6](images/1.png)
+```
+Real mark:  6
+NN answer:  6
+```
+![MNIST тестовое изображение - цифра 3](images/2.png)
+```
+Real mark:  3
+NN answer:  3
+```
+
+### 8) Детальный анализ качества классификации
+
+Генерируем подробный отчет о качестве классификации и строим матрицу ошибок для визуального анализа работы модели по каждому классу.
+
+```python
+# Получение истинных и предсказанных меток для всех тестовых данных
+true_labels = np.argmax(y_test, axis=1)
+# Предсказанные метки классов
+predicted_labels = np.argmax(model.predict(X_test), axis=1)
+
+# Вывод подробного отчета о качестве классификации
+print(classification_report(true_labels, predicted_labels))
+# Построение и визуализация матрицы ошибок
+conf_matrix = confusion_matrix(true_labels, predicted_labels)
+# Отрисовка матрицы ошибок в виде "тепловой карты"
+display = ConfusionMatrixDisplay(confusion_matrix=conf_matrix)
+display.plot()
+plt.show()
+```
+```
+313/313 ━━━━━━━━━━━━━━━━━━━━ 2s 5ms/step
+              precision    recall  f1-score   support
+
+           0       1.00      0.99      1.00      1001
+           1       0.99      1.00      0.99      1143
+           2       0.99      0.99      0.99       987
+           3       0.99      0.99      0.99      1023
+           4       0.99      0.99      0.99       974
+           5       1.00      0.98      0.99       907
+           6       0.99      0.99      0.99       974
+           7       0.98      0.99      0.99      1032
+           8       0.98      0.98      0.98      1006
+           9       0.98      0.99      0.98       953
+
+    accuracy                           0.99     10000
+   macro avg       0.99      0.99      0.99     10000
+weighted avg       0.99      0.99      0.99     10000
+```
+![Матрица ошибок для MNIST](images/3.png)
+
+### 9) Тестирование на собственных изображениях
+
+Загружаем и обрабатываем собственные изображения цифр, созданные ранее, и проверяем способность модели их корректно распознавать.
+
+```python
+# Загрузка и обработка собственных изображений
+from PIL import Image
+
+for name_image in ['2.png', '7.png']:
+  file_data = Image.open(name_image)
+  file_data = file_data.convert('L') # Перевод в градации серого
+  test_img = np.array(file_data)
+
+  # Вывод собственного изображения
+  plt.imshow(test_img, cmap=plt.get_cmap('gray'))
+  plt.show()
+
+  # Предобработка
+  test_img = test_img / 255
+  test_img = np.reshape(test_img, (1,28,28,1))
+
+  # Распознавание
+  result = model.predict(test_img)
+  print('I think it\'s', np.argmax(result))
+```
+![Собственное изображение - цифра 2](images/4.png)
+```
+I think it's 2
+```
+![Собственное изображение - цифра 7](images/5.png)
+```
+I think it's 7
+```
+
+### 10) Сравнение с моделью из предыдущей лабораторной работы
+
+Загружаем сохраненную полносвязную нейронную сеть из лабораторной работы №1 и оцениваем ее производительность на тех же тестовых данных для последующего сравнения.
+
+```python
+model_lr1 = keras.models.load_model("best_mnist_model.keras")
+
+model_lr1.summary()
+```
+**Model: "sequential"**
+| Layer (type)     | Output Shape | Param # |
+|------------------|-------------:|--------:|
+| dense (Dense) | (None, 10)  |  7,850 |
+**Total params:** 7,852 (30.68 KB)  
+**Trainable params:** 7,850 (30.66 KB)  
+**Non-trainable params:** 0 (0.00 B)  
+**Optimizer params:** 2 (12.00 B)
+
+
+```python
+# Подготовка данных для полносвязной сети (преобразование изображений в векторы)
+X_train, X_test, y_train, y_test = train_test_split(X, y,
+                                                    test_size = 10000,
+                                                    train_size = 60000,
+                                                    random_state = 3)
+num_pixels = X_train.shape[1] * X_train.shape[2]
+X_train = X_train.reshape(X_train.shape[0], num_pixels) / 255
+X_test = X_test.reshape(X_test.shape[0], num_pixels) / 255
+print('Shape of transformed X train:', X_train.shape)
+print('Shape of transformed X train:', X_test.shape)
+
+# Преобразование меток в формат one-hot encoding
+y_train = keras.utils.to_categorical(y_train, num_classes)
+y_test = keras.utils.to_categorical(y_test, num_classes)
+print('Shape of transformed y train:', y_train.shape)
+print('Shape of transformed y test:', y_test.shape)
+```
+```
+Shape of transformed X train: (60000, 784)
+Shape of transformed X train: (10000, 784)
+Shape of transformed y train: (60000, 10)
+Shape of transformed y test: (10000, 10)
+```
+
+```python
+# Оценка качества работы модели на тестовых данных
+scores = model_lr1.evaluate(X_test, y_test)
+print('Loss on test data:', scores[0])
+print('Accuracy on test data:', scores[1])
+```
+```
+313/313 ━━━━━━━━━━━━━━━━━━━━ 1s 3ms/step - accuracy: 0.9233 - loss: 0.2863
+Loss on test data: 0.28625616431236267
+Accuracy on test data: 0.92330002784729
+```
+
+### 11) Сравнительный анализ моделей
+
+Сравниваем сверточную нейронную сеть с полносвязной сетью по ключевым показателям: количеству параметров, времени обучения и качеству классификации. Делаем выводы по результатам применения сверточной нейронной сети для распознавания изображений.
+
+**Таблица сравнения моделей:**
+
+| Модель   | Количество настраиваемых параметров | Количество эпох обучения | Качество классификации тестовой выборки |
+|----------|-------------------------------------|---------------------------|-----------------------------------------|
+| Сверточная | 34 826                              | 15                        | accuracy: 0.988; loss: 0.041                                   |
+| Полносвязная | 7 852                              | 50                        | accuracy: 0.923; loss: 0.286                                  |
+
+
+**Выводы:**
+
+На основе проведенного анализа можно заключить, что сверточная нейронная сеть демонстрирует существенные преимущества перед полносвязной сетью при решении задач распознавания изображений:
+
+1. **Эффективность параметров**: Сверточная сеть имеет больше параметров (34 826 против 7 852), но при этом показывает значительно лучшие результаты, что говорит о более эффективном использовании параметров для извлечения пространственных признаков.
+
+2. **Скорость обучения**: Для достижения высокого качества сверточной сети требуется в 3.3 раза меньше эпох обучения (15 против 50), что существенно сокращает время обучения.
+
+3. **Точность классификации**: Сверточная сеть показывает более высокую точность (98.8% против 92.3%) и значительно меньшую функцию потерь (0.041 против 0.286). Разница в точности составляет 6.5 процентных пункта, что является существенным улучшением.
+
+4. **Обобщающая способность**: Сверточная сеть демонстрирует лучшую способность к обобщению, что видно из более низкой функции потерь на тестовых данных.
+
+Эти результаты подтверждают, что архитектура сверточных сетей, учитывающая пространственную структуру изображений через операции свертки и пулинга, является более подходящим выбором для задач компьютерного зрения, несмотря на большее количество параметров.
+
+## Задание 2
+
+### Работа с датасетом CIFAR-10
+
+Повторяем основные этапы задания 1, но используем датасет CIFAR-10, содержащий цветные изображения объектов 10 различных классов.
+
+Особенности выполнения:
+- Разделение данных производится в соотношении 50 000:10 000
+- После разделения визуализируем 25 изображений из обучающей выборки
+- При демонстрации работы модели выбираем примеры так, чтобы одно изображение распознавалось корректно, а другое - ошибочно
+
+### 1) Загрузка датасета CIFAR-10
+
+Загружаем набор данных CIFAR-10, который содержит цветные изображения размером 32x32 пикселя, разделенные на 10 классов: самолет, автомобиль, птица, кошка, олень, собака, лягушка, лошадь, корабль, грузовик.
+
+```python
+# Импорт и загрузка датасета CIFAR-10
+from keras.datasets import cifar10
+
+(X_train, y_train), (X_test, y_test) = cifar10.load_data()
+```
+
+### 2) Разделение данных на обучающую и тестовую выборки
+
+Создаем собственное разбиение датасета CIFAR-10 в соотношении 50 000:10 000. Используем random_state = 3 для воспроизводимости результатов (k = 1 - номер нашей бригады). Выводим размерности полученных массивов данных.
+
+```python
+# Создание собственного разбиения датасета
+
+# Объединение исходных выборок
+X = np.concatenate((X_train, X_test))
+y = np.concatenate((y_train, y_test))
+
+# Разделение на обучающую и тестовую выборки
+X_train, X_test, y_train, y_test = train_test_split(X, y,
+                                                    test_size = 10000,
+                                                    train_size = 50000,
+                                                    random_state = 3)
+# Вывод размерностей
+print('Shape of X train:', X_train.shape)
+print('Shape of y train:', y_train.shape)
+print('Shape of X test:', X_test.shape)
+print('Shape of y test:', y_test.shape)
+```
+```
+Shape of X train: (50000, 32, 32, 3)
+Shape of y train: (50000, 1)
+Shape of X test: (10000, 32, 32, 3)
+Shape of y test: (10000, 1)
+```
+
+### Визуализация примеров из обучающей выборки
+
+Отображаем сетку из 25 изображений из обучающей выборки с подписями соответствующих классов для визуального ознакомления с данными.
+
+```python
+class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',
+               'dog', 'frog', 'horse', 'ship', 'truck']
+
+plt.figure(figsize=(10,10))
+for i in range(25):
+    plt.subplot(5,5,i+1)
+    plt.xticks([])
+    plt.yticks([])
+    plt.grid(False)
+    plt.imshow(X_train[i])
+    plt.xlabel(class_names[y_train[i][0]])
+plt.show()
+```
+![Сетка из 25 изображений CIFAR-10](images/6.png)
+
+### 3) Предобработка данных CIFAR-10
+
+Нормализуем значения пикселей и преобразуем метки в формат one-hot encoding для работы с категориальной функцией потерь. Выводим размерности предобработанных массивов данных.
+
+```python
+# Определение параметров данных и модели
+num_classes = 10
+input_shape = (32, 32, 3)
+
+# Нормализация значений пикселей: приведение к диапазону [0, 1]
+X_train = X_train / 255
+X_test = X_test / 255
+
+print('Shape of transformed X train:', X_train.shape)
+print('Shape of transformed X test:', X_test.shape)
+
+# Преобразование меток в формат one-hot encoding
+y_train = keras.utils.to_categorical(y_train, num_classes)
+y_test = keras.utils.to_categorical(y_test, num_classes)
+print('Shape of transformed y train:', y_train.shape)
+print('Shape of transformed y test:', y_test.shape)
+```
+```
+Shape of transformed X train: (50000, 32, 32, 3)
+Shape of transformed X test: (10000, 32, 32, 3)
+Shape of transformed y train: (50000, 10)
+Shape of transformed y test: (10000, 10)
+```
+
+### 4) Построение и обучение сверточной сети для CIFAR-10
+
+Создаем более сложную архитектуру сверточной сети с использованием батч-нормализации и нескольких блоков свертки для работы с цветными изображениями. Обучаем модель на подготовленных данных с выделением части данных для валидации.
+
+```python
+# Создание модели сверточной нейронной сети
+model = Sequential()
+
+# Блок 1
+model.add(layers.Conv2D(32, (3, 3), padding="same",
+                        activation="relu", input_shape=input_shape))
+model.add(layers.BatchNormalization())
+model.add(layers.Conv2D(32, (3, 3), padding="same", activation="relu"))
+model.add(layers.BatchNormalization())
+model.add(layers.MaxPooling2D((2, 2)))
+model.add(layers.Dropout(0.25))
+
+# Блок 2
+model.add(layers.Conv2D(64, (3, 3), padding="same", activation="relu"))
+model.add(layers.BatchNormalization())
+model.add(layers.Conv2D(64, (3, 3), padding="same", activation="relu"))
+model.add(layers.BatchNormalization())
+model.add(layers.MaxPooling2D((2, 2)))
+model.add(layers.Dropout(0.25))
+
+# Блок 3
+model.add(layers.Conv2D(128, (3, 3), padding="same", activation="relu"))
+model.add(layers.BatchNormalization())
+model.add(layers.Conv2D(128, (3, 3), padding="same", activation="relu"))
+model.add(layers.BatchNormalization())
+model.add(layers.MaxPooling2D((2, 2)))
+model.add(layers.Dropout(0.4))
+
+model.add(layers.Flatten())
+model.add(layers.Dense(128, activation='relu'))
+model.add(layers.Dropout(0.5))
+model.add(layers.Dense(num_classes, activation="softmax"))
+
+
+model.summary()
+```
+**Model: "sequential_9"**
+| Layer (type)                               | Output Shape      |  Param # |
+|--------------------------------------------|-------------------|---------:|
+| conv2d_41 (Conv2D)                         | (None, 32, 32, 32) |      896 |
+| batch_normalization_6 (BatchNormalization) | (None, 32, 32, 32) |      128 |
+| conv2d_42 (Conv2D)                         | (None, 32, 32, 32) |    9,248 |
+| batch_normalization_7 (BatchNormalization) | (None, 32, 32, 32) |      128 |
+| max_pooling2d_26 (MaxPooling2D)           | (None, 16, 16, 32) |        0 |
+| dropout_24 (Dropout)                       | (None, 16, 16, 32) |        0 |
+| conv2d_43 (Conv2D)                         | (None, 16, 16, 64) |   18,496 |
+| batch_normalization_8 (BatchNormalization) | (None, 16, 16, 64) |      256 |
+| conv2d_44 (Conv2D)                         | (None, 16, 16, 64) |   36,928 |
+| batch_normalization_9 (BatchNormalization) | (None, 16, 16, 64) |      256 |
+| max_pooling2d_27 (MaxPooling2D)           | (None, 8, 8, 64)   |        0 |
+| dropout_25 (Dropout)                       | (None, 8, 8, 64)   |        0 |
+| conv2d_45 (Conv2D)                         | (None, 8, 8, 128)  |   73,856 |
+| batch_normalization_10 (BatchNormalization)| (None, 8, 8, 128)  |      512 |
+| conv2d_46 (Conv2D)                         | (None, 8, 8, 128)  |  147,584 |
+| batch_normalization_11 (BatchNormalization)| (None, 8, 8, 128)  |      512 |
+| max_pooling2d_28 (MaxPooling2D)           | (None, 4, 4, 128)  |        0 |
+| dropout_26 (Dropout)                       | (None, 4, 4, 128)  |        0 |
+| flatten_9 (Flatten)                        | (None, 2048)       |        0 |
+| dense_17 (Dense)                           | (None, 128)        |  262,272 |
+| dropout_27 (Dropout)                       | (None, 128)        |        0 |
+| dense_18 (Dense)                           | (None, 10)         |    1,290 |
+**Total params:** 552,362 (2.11 MB)  
+**Trainable params:** 551,466 (2.10 MB)  
+**Non-trainable params:** 896 (3.50 KB)
+
+```python
+# Компиляция и обучение модели
+batch_size = 64
+epochs = 50
+model.compile(loss="categorical_crossentropy", optimizer="adam", metrics=["accuracy"])
+model.fit(X_train, y_train, batch_size=batch_size, epochs=epochs, validation_split=0.1)
+```
+
+### 5) Оценка качества модели на тестовых данных
+
+Оцениваем финальную производительность обученной модели на тестовой выборке CIFAR-10.
+
+```python
+# Оценка качества работы модели на тестовых данных
+scores = model.evaluate(X_test, y_test)
+print('Loss on test data:', scores[0])
+print('Accuracy on test data:', scores[1])
+```
+```
+313/313 ━━━━━━━━━━━━━━━━━━━━ 7s 22ms/step - accuracy: 0.8553 - loss: 0.5210
+Loss on test data: 0.5209607481956482
+Accuracy on test data: 0.8553000092506409
+```
+
+### 6) Демонстрация работы модели на отдельных примерах
+
+Визуализируем результаты распознавания для двух тестовых изображений: одно должно быть распознано корректно, другое - ошибочно.
+
+```python
+# Визуализация результатов распознавания для двух тестовых изображений
+
+for n in [3,14]:
+  result = model.predict(X_test[n:n+1])
+  print('NN output:', result)
+
+  plt.imshow(X_test[n].reshape(32,32,3), cmap=plt.get_cmap('gray'))
+  plt.show()
+  print('Real mark: ', np.argmax(y_test[n]))
+  print('NN answer: ', np.argmax(result))
+```
+![CIFAR-10 тестовое изображение](images/7.png)
+```
+Real mark:  6
+NN answer:  6
+```
+![CIFAR-10 тестовое изображение - олень (ошибочно распознано)](images/8.png)
+```
+Real mark:  4
+NN answer:  5
+```
+
+### 7) Детальный анализ качества классификации CIFAR-10
+
+Генерируем подробный отчет о качестве классификации и строим матрицу ошибок для анализа работы модели по каждому классу.
+
+```python
+# Получение истинных и предсказанных меток для всех тестовых данных
+true_labels = np.argmax(y_test, axis=1)
+# Предсказанные метки классов
+predicted_labels = np.argmax(model.predict(X_test), axis=1)
+
+# Вывод подробного отчета о качестве классификации
+print(classification_report(true_labels, predicted_labels, target_names=class_names))
+# Построение и визуализация матрицы ошибок
+conf_matrix = confusion_matrix(true_labels, predicted_labels)
+# Отрисовка матрицы ошибок в виде "тепловой карты"
+fig, ax = plt.subplots(figsize=(6, 6))
+disp = ConfusionMatrixDisplay(confusion_matrix=conf_matrix,display_labels=class_names)
+disp.plot(ax=ax, xticks_rotation=45)  # Поворот подписей по X и приятная палитра
+plt.tight_layout()  # Чтобы всё влезло
+plt.show()
+```
+```
+313/313 ━━━━━━━━━━━━━━━━━━━━ 1s 4ms/step
+              precision    recall  f1-score   support
+
+    airplane       0.84      0.91      0.87      1007
+  automobile       0.95      0.91      0.93      1037
+        bird       0.83      0.79      0.81      1030
+         cat       0.77      0.65      0.70       990
+        deer       0.83      0.82      0.82       966
+         dog       0.72      0.83      0.77      1009
+        frog       0.90      0.89      0.89       972
+       horse       0.87      0.89      0.88       991
+        ship       0.95      0.92      0.93       990
+       truck       0.89      0.93      0.91      1008
+
+    accuracy                           0.85     10000
+   macro avg       0.86      0.85      0.85     10000
+weighted avg       0.86      0.85      0.85     10000
+```
+![Матрица ошибок для CIFAR-10](images/9.png)
+
+**Выводы по результатам классификации CIFAR-10:**
+
+Разработанная сверточная нейронная сеть показала хорошие результаты при классификации цветных изображений из датасета CIFAR-10. Модель достигла точности классификации 85.5% (accuracy: 0.855, loss: 0.521) на тестовой выборке, а в детальном отчете о классификации показала accuracy 0.85, что является достойным результатом для данной задачи, учитывая сложность различения объектов в низком разрешении (32x32 пикселя) и наличие 10 различных классов.
+
+Использование батч-нормализации и dropout-регуляризации позволило улучшить обобщающую способность модели и предотвратить переобучение. Архитектура с тремя блоками сверточных слоев эффективно извлекает иерархические признаки из изображений, что подтверждается полученными метриками качества.
+