{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oZs0KGcz01BY"
   },
   "source": [
    "## Задание 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gz18QPRz03Ec"
   },
   "source": [
    "### 1) Подготовили рабочую среду в Google Colab, создав новый блокнот. Выполнили импорт требуемых библиотек и модулей для дальнейшей работы."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "mr9IszuQ1ANG"
   },
   "outputs": [],
   "source": [
    "# импорт модулей\n",
    "import os\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) Произвели загрузку датасета MNIST, который включает размеченные изображения рукописных цифр.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "Ixw5Sp0_1A-w"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
      "\u001b[1m11490434/11490434\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 0us/step\n"
     ]
    }
   ],
   "source": [
    "# загрузка датасета\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) Выполнили разделение датасета на обучающую и тестовую выборки в пропорции 60 000:10 000. Для воспроизводимости результатов установили параметр random_state равным (4k – 1)=31, где k=8 соответствует номеру нашей бригады. Отобразили размерности полученных массивов данных."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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 своего разбиения датасета\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# объединяем в один набор\n",
    "X = np.concatenate((X_train, X_test))\n",
    "y = np.concatenate((y_train, y_test))\n",
    "\n",
    "# разбиваем по вариантам\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 = 31)\n",
    "# вывод размерностей\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) Осуществили предобработку данных для подготовки к обучению сверточной нейронной сети. Нормализовали пиксели изображений в диапазон [0, 1], а метки классов преобразовали в формат one-hot encoding. Продемонстрировали размерности обработанных массивов."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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": [
    "# Зададим параметры данных и модели\n",
    "num_classes = 10\n",
    "input_shape = (28, 28, 1)\n",
    "\n",
    "# Приведение входных данных 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",
    "# переведем метки в one-hot\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) Разработали архитектуру сверточной нейронной сети и провели ее обучение на обучающей выборке, выделив часть данных для валидации. Представили структуру созданной модели."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "Un561zSH1Cmv"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\Admin\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\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": [
    "# создаем модель\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": 8,
   "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[1m11s\u001b[0m 85ms/step - accuracy: 0.7738 - loss: 0.7503 - val_accuracy: 0.9435 - val_loss: 0.1959\n",
      "Epoch 2/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 73ms/step - accuracy: 0.9429 - loss: 0.1898 - val_accuracy: 0.9670 - val_loss: 0.1182\n",
      "Epoch 3/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 72ms/step - accuracy: 0.9603 - loss: 0.1322 - val_accuracy: 0.9743 - val_loss: 0.0887\n",
      "Epoch 4/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 74ms/step - accuracy: 0.9678 - loss: 0.1057 - val_accuracy: 0.9760 - val_loss: 0.0762\n",
      "Epoch 5/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 70ms/step - accuracy: 0.9719 - loss: 0.0902 - val_accuracy: 0.9787 - val_loss: 0.0687\n",
      "Epoch 6/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 71ms/step - accuracy: 0.9749 - loss: 0.0810 - val_accuracy: 0.9800 - val_loss: 0.0622\n",
      "Epoch 7/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 70ms/step - accuracy: 0.9775 - loss: 0.0719 - val_accuracy: 0.9820 - val_loss: 0.0575\n",
      "Epoch 8/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 70ms/step - accuracy: 0.9790 - loss: 0.0659 - val_accuracy: 0.9823 - val_loss: 0.0524\n",
      "Epoch 9/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 73ms/step - accuracy: 0.9812 - loss: 0.0627 - val_accuracy: 0.9827 - val_loss: 0.0525\n",
      "Epoch 10/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 73ms/step - accuracy: 0.9815 - loss: 0.0574 - val_accuracy: 0.9837 - val_loss: 0.0480\n",
      "Epoch 11/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 70ms/step - accuracy: 0.9833 - loss: 0.0530 - val_accuracy: 0.9845 - val_loss: 0.0454\n",
      "Epoch 12/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 72ms/step - accuracy: 0.9838 - loss: 0.0521 - val_accuracy: 0.9853 - val_loss: 0.0438\n",
      "Epoch 13/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 71ms/step - accuracy: 0.9841 - loss: 0.0498 - val_accuracy: 0.9857 - val_loss: 0.0436\n",
      "Epoch 14/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 74ms/step - accuracy: 0.9857 - loss: 0.0472 - val_accuracy: 0.9865 - val_loss: 0.0413\n",
      "Epoch 15/15\n",
      "\u001b[1m106/106\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 73ms/step - accuracy: 0.9859 - loss: 0.0445 - val_accuracy: 0.9873 - val_loss: 0.0396\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.src.callbacks.history.History at 0x22c3ba83ed0>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# компилируем и обучаем модель\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) Протестировали обученную модель на тестовой выборке. Определили значения функции потерь и точности классификации."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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 3ms/step - accuracy: 0.9873 - loss: 0.0396\n",
      "Loss on test data: 0.03962046653032303\n",
      "Accuracy on test data: 0.9872999787330627\n"
     ]
    }
   ],
   "source": [
    "# Оценка качества работы модели на тестовых данных\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) Протестировали модель на двух произвольных изображениях из тестовой выборки. Визуализировали изображения и сравнили истинные метки с предсказаниями модели."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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 76ms/step\n",
      "NN output: [[1.02410545e-10 1.21296682e-06 6.38641040e-06 8.52757785e-06\n",
      "  3.49328509e-12 4.14857287e-10 9.68709627e-17 9.99978900e-01\n",
      "  6.33653556e-08 4.88464457e-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:  7\n",
      "NN answer:  7\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n",
      "NN output: [[2.6005425e-03 3.0321027e-07 2.2106780e-05 1.3148747e-05 8.1389046e-01\n",
      "  1.5582073e-04 3.5853500e-05 2.7356921e-03 2.7238589e-02 1.5330753e-01]]\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:  4\n"
     ]
    }
   ],
   "source": [
    "# вывод двух тестовых изображений и результатов распознавания\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) Сформировали детальный отчет о качестве классификации на тестовой выборке, включая матрицу ошибок (confusion matrix)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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 2ms/step\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.99      0.99       967\n",
      "           1       1.00      0.99      0.99      1107\n",
      "           2       0.98      0.99      0.99       970\n",
      "           3       0.99      0.98      0.99      1023\n",
      "           4       1.00      0.99      0.99      1008\n",
      "           5       0.98      0.99      0.98       866\n",
      "           6       0.99      0.99      0.99       965\n",
      "           7       0.98      0.98      0.98      1070\n",
      "           8       0.98      0.99      0.99       943\n",
      "           9       0.98      0.98      0.98      1081\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": [
    "# истинные метки классов\n",
    "true_labels = np.argmax(y_test, axis=1)\n",
    "# предсказанные метки классов\n",
    "predicted_labels = np.argmax(model.predict(X_test), axis=1)\n",
    "\n",
    "# отчет о качестве классификации\n",
    "print(classification_report(true_labels, predicted_labels))\n",
    "# вычисление матрицы ошибок\n",
    "conf_matrix = confusion_matrix(true_labels, predicted_labels)\n",
    "# отрисовка матрицы ошибок в виде \"тепловой карты\"\n",
    "display = ConfusionMatrixDisplay(confusion_matrix=conf_matrix)\n",
    "display.plot()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "amaspXGW1EVy"
   },
   "source": [
    "### 9) Загрузили собственные изображения, подготовленные в рамках лабораторной работы №1. После предобработки передали их на вход обученной модели и получили результаты распознавания."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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 49ms/step\n",
      "I think it's 0\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 56ms/step\n",
      "I think it's 1\n"
     ]
    }
   ],
   "source": [
    "# загрузка собственного изображения\n",
    "from PIL import Image\n",
    "\n",
    "for name_image in ['0.png', '1.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) Загрузили ранее сохраненную модель из лабораторной работы №1. Изучили ее архитектуру и провели оценку качества на тестовых данных аналогично пункту 6."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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_7\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential_7\"\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_15 (<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\">100</span>)            │        <span style=\"color: #00af00; text-decoration-color: #00af00\">78,500</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_16 (<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\">100</span>)            │        <span style=\"color: #00af00; text-decoration-color: #00af00\">10,100</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_17 (<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\">1,010</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",
       "│ dense_15 (\u001b[38;5;33mDense\u001b[0m)                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m100\u001b[0m)            │        \u001b[38;5;34m78,500\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_16 (\u001b[38;5;33mDense\u001b[0m)                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m100\u001b[0m)            │        \u001b[38;5;34m10,100\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_17 (\u001b[38;5;33mDense\u001b[0m)                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)             │         \u001b[38;5;34m1,010\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\">89,612</span> (350.05 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m89,612\u001b[0m (350.05 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\">89,610</span> (350.04 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m89,610\u001b[0m (350.04 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"
    },
    {
     "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_model.keras\")\n",
    "\n",
    "model_lr1.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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": [
    "# развернем каждое изображение 28*28 в вектор 784\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 = 31)\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",
    "# переведем метки в one-hot\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": 21,
   "metadata": {
    "id": "0Yj0fzLNE12k"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 3ms/step - accuracy: 0.9440 - loss: 0.1897\n",
      "Loss on test data: 0.18974457681179047\n",
      "Accuracy on test data: 0.9440000057220459\n"
     ]
    }
   ],
   "source": [
    "# Оценка качества работы модели на тестовых данных\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) Выполнили сравнительный анализ сверточной нейронной сети и лучшей полносвязной модели из лабораторной работы №1. Сравнение проводилось по трем критериям:\n",
    "### - число обучаемых параметров модели\n",
    "### - количество эпох, необходимое для обучения\n",
    "### - итоговое качество классификации на тестовой выборке\n",
    "### На основе полученных результатов сформулировали выводы об эффективности применения сверточных нейронных сетей для задач распознавания изображений.  "
   ]
  },
  {
   "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:0.987     ; loss:0.040                                   |\n",
    "| Полносвязная | 84 062                              | 50                        | accuracy:0.944     ; loss:0.190                                  |\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YctF8h_sIB-P"
   },
   "source": [
    "##### Проведенный сравнительный анализ, результаты которого представлены в таблице 1, наглядно демонстрирует превосходство сверточной нейронной сети над полносвязной архитектурой в задачах классификации изображений. \n",
    "\n",
    "**Эффективность по параметрам:** Сверточная сеть содержит в 2.4 раза меньше обучаемых параметров (34 826 против 84 062), что свидетельствует о более эффективном использовании вычислительных ресурсов благодаря механизму разделения весов в сверточных слоях.\n",
    "\n",
    "**Скорость обучения:** Сверточная модель достигает оптимального качества за 15 эпох, в то время как полносвязная требует 50 эпох. Это указывает на более быструю сходимость алгоритма обучения благодаря индуктивным смещениям, заложенным в архитектуру сверточных сетей.\n",
    "\n",
    "**Качество классификации:** Сверточная сеть демонстрирует значительно более высокую точность (98.7% против 94.4%) и существенно меньшие потери (0.040 против 0.190). Разница в точности составляет более 4 процентных пунктов, что является существенным улучшением для задачи распознавания рукописных цифр.\n",
    "\n",
    "**Выводы:** Полученные результаты подтверждают, что использование сверточных слоев позволяет эффективно извлекать иерархические пространственные признаки из изображений, что критически важно для задач компьютерного зрения. Инвариантность к сдвигам и способность выявлять локальные паттерны делают сверточные нейронные сети предпочтительным выбором для работы с изображениями по сравнению с полносвязными архитектурами."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wCLHZPGB1F1y"
   },
   "source": [
    "## Задание 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DUOYls124TT8"
   },
   "source": [
    "### В отдельном блокноте повторили этапы 2–8 из задания 1, заменив датасет MNIST на CIFAR-10, который содержит цветные изображения объектов, распределенные по 10 категориям.  \n",
    "### Особенности выполнения:\n",
    "### - разделение на обучающую и тестовую выборки выполнено в пропорции 50 000:10 000\n",
    "### - после разделения данных (между этапами 3 и 4) визуализировали 25 примеров из обучающей выборки с указанием соответствующих классов\n",
    "### - при тестировании на двух изображениях (этап 7) одно должно быть распознано верно, а второе – с ошибкой   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XDStuSpEJa8o"
   },
   "source": [
    "### 1) Произвели загрузку датасета CIFAR-10, включающего цветные изображения, распределенные по 10 категориям: самолет, автомобиль, птица, кошка, олень, собака, лягушка, лошадь, корабль, грузовик."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "id": "y0qK7eKL4Tjy"
   },
   "outputs": [],
   "source": [
    "# загрузка датасета\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) Осуществили разделение датасета на обучающую и тестовую части в соотношении 50 000:10 000. Для обеспечения воспроизводимости установили random_state = 31, что соответствует формуле (4k – 1) при k=8 (номер нашей бригады). Отобразили размерности сформированных массивов."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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 своего разбиения датасета\n",
    "\n",
    "# объединяем в один набор\n",
    "X = np.concatenate((X_train, X_test))\n",
    "y = np.concatenate((y_train, y_test))\n",
    "\n",
    "# разбиваем по вариантам\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 = 31)\n",
    "# вывод размерностей\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": [
    "### Визуализировали 25 примеров из обучающей выборки с указанием их классов."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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) Выполнили предобработку данных для обучения сверточной нейронной сети. Нормализовали значения пикселей в диапазон [0, 1] и преобразовали метки классов в формат one-hot encoding. Показали размерности обработанных массивов."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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": [
    "# Зададим параметры данных и модели\n",
    "num_classes = 10\n",
    "input_shape = (32, 32, 3)\n",
    "\n",
    "# Приведение входных данных 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",
    "# переведем метки в one-hot\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) Построили архитектуру сверточной нейронной сети и провели обучение на обучающей выборке с использованием части данных для валидации. Представили детальную структуру модели."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "YhAD5CllLlv7"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\Admin\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\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": [
    "# создаем модель\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": 27,
   "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[1m87s\u001b[0m 115ms/step - accuracy: 0.3052 - loss: 1.8713 - val_accuracy: 0.4752 - val_loss: 1.3957\n",
      "Epoch 2/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.4705 - loss: 1.4488 - val_accuracy: 0.5730 - val_loss: 1.1992\n",
      "Epoch 3/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m78s\u001b[0m 111ms/step - accuracy: 0.5626 - loss: 1.2235 - val_accuracy: 0.6470 - val_loss: 1.0268\n",
      "Epoch 4/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m75s\u001b[0m 107ms/step - accuracy: 0.6261 - loss: 1.0727 - val_accuracy: 0.6940 - val_loss: 0.8987\n",
      "Epoch 5/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m75s\u001b[0m 106ms/step - accuracy: 0.6678 - loss: 0.9739 - val_accuracy: 0.7042 - val_loss: 0.8850\n",
      "Epoch 6/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m73s\u001b[0m 104ms/step - accuracy: 0.6986 - loss: 0.8855 - val_accuracy: 0.7360 - val_loss: 0.7630\n",
      "Epoch 7/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m74s\u001b[0m 105ms/step - accuracy: 0.7183 - loss: 0.8263 - val_accuracy: 0.7624 - val_loss: 0.7084\n",
      "Epoch 8/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m77s\u001b[0m 110ms/step - accuracy: 0.7344 - loss: 0.7800 - val_accuracy: 0.7724 - val_loss: 0.6707\n",
      "Epoch 9/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m83s\u001b[0m 118ms/step - accuracy: 0.7575 - loss: 0.7222 - val_accuracy: 0.7818 - val_loss: 0.6691\n",
      "Epoch 10/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m79s\u001b[0m 112ms/step - accuracy: 0.7705 - loss: 0.6802 - val_accuracy: 0.7970 - val_loss: 0.6004\n",
      "Epoch 11/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m83s\u001b[0m 118ms/step - accuracy: 0.7839 - loss: 0.6496 - val_accuracy: 0.7932 - val_loss: 0.6760\n",
      "Epoch 12/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m83s\u001b[0m 118ms/step - accuracy: 0.7897 - loss: 0.6216 - val_accuracy: 0.8122 - val_loss: 0.5603\n",
      "Epoch 13/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m84s\u001b[0m 119ms/step - accuracy: 0.8016 - loss: 0.5895 - val_accuracy: 0.7936 - val_loss: 0.6226\n",
      "Epoch 14/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m84s\u001b[0m 120ms/step - accuracy: 0.8129 - loss: 0.5600 - val_accuracy: 0.8160 - val_loss: 0.5553\n",
      "Epoch 15/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8173 - loss: 0.5403 - val_accuracy: 0.8282 - val_loss: 0.5158\n",
      "Epoch 16/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m82s\u001b[0m 116ms/step - accuracy: 0.8224 - loss: 0.5228 - val_accuracy: 0.8338 - val_loss: 0.5143\n",
      "Epoch 17/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m82s\u001b[0m 117ms/step - accuracy: 0.8313 - loss: 0.4944 - val_accuracy: 0.8190 - val_loss: 0.5393\n",
      "Epoch 18/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8374 - loss: 0.4780 - val_accuracy: 0.7674 - val_loss: 0.7332\n",
      "Epoch 19/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m82s\u001b[0m 116ms/step - accuracy: 0.8427 - loss: 0.4673 - val_accuracy: 0.8398 - val_loss: 0.4830\n",
      "Epoch 20/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m85s\u001b[0m 121ms/step - accuracy: 0.8463 - loss: 0.4508 - val_accuracy: 0.8292 - val_loss: 0.5125\n",
      "Epoch 21/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 114ms/step - accuracy: 0.8526 - loss: 0.4341 - val_accuracy: 0.8374 - val_loss: 0.5082\n",
      "Epoch 22/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8560 - loss: 0.4201 - val_accuracy: 0.8382 - val_loss: 0.5002\n",
      "Epoch 23/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 115ms/step - accuracy: 0.8617 - loss: 0.4127 - val_accuracy: 0.8262 - val_loss: 0.5137\n",
      "Epoch 24/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8676 - loss: 0.3964 - val_accuracy: 0.8400 - val_loss: 0.4983\n",
      "Epoch 25/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8669 - loss: 0.3931 - val_accuracy: 0.8416 - val_loss: 0.4823\n",
      "Epoch 26/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m82s\u001b[0m 116ms/step - accuracy: 0.8692 - loss: 0.3839 - val_accuracy: 0.8462 - val_loss: 0.4897\n",
      "Epoch 27/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 115ms/step - accuracy: 0.8740 - loss: 0.3722 - val_accuracy: 0.8338 - val_loss: 0.5208\n",
      "Epoch 28/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8764 - loss: 0.3643 - val_accuracy: 0.8480 - val_loss: 0.4734\n",
      "Epoch 29/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 115ms/step - accuracy: 0.8812 - loss: 0.3498 - val_accuracy: 0.8514 - val_loss: 0.4512\n",
      "Epoch 30/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8820 - loss: 0.3463 - val_accuracy: 0.8432 - val_loss: 0.5021\n",
      "Epoch 31/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8874 - loss: 0.3368 - val_accuracy: 0.8486 - val_loss: 0.4834\n",
      "Epoch 32/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 115ms/step - accuracy: 0.8866 - loss: 0.3299 - val_accuracy: 0.8424 - val_loss: 0.5011\n",
      "Epoch 33/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8880 - loss: 0.3268 - val_accuracy: 0.8398 - val_loss: 0.5170\n",
      "Epoch 34/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 114ms/step - accuracy: 0.8909 - loss: 0.3200 - val_accuracy: 0.8482 - val_loss: 0.4952\n",
      "Epoch 35/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8911 - loss: 0.3198 - val_accuracy: 0.8516 - val_loss: 0.4742\n",
      "Epoch 36/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 114ms/step - accuracy: 0.8956 - loss: 0.3110 - val_accuracy: 0.8588 - val_loss: 0.4497\n",
      "Epoch 37/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.8966 - loss: 0.2992 - val_accuracy: 0.8512 - val_loss: 0.4598\n",
      "Epoch 38/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 115ms/step - accuracy: 0.8999 - loss: 0.2949 - val_accuracy: 0.8478 - val_loss: 0.5029\n",
      "Epoch 39/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 114ms/step - accuracy: 0.9016 - loss: 0.2857 - val_accuracy: 0.8632 - val_loss: 0.4740\n",
      "Epoch 40/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 115ms/step - accuracy: 0.9013 - loss: 0.2915 - val_accuracy: 0.8578 - val_loss: 0.4687\n",
      "Epoch 41/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 114ms/step - accuracy: 0.9038 - loss: 0.2822 - val_accuracy: 0.8588 - val_loss: 0.4607\n",
      "Epoch 42/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.9054 - loss: 0.2767 - val_accuracy: 0.8594 - val_loss: 0.4645\n",
      "Epoch 43/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 114ms/step - accuracy: 0.9067 - loss: 0.2722 - val_accuracy: 0.8628 - val_loss: 0.4632\n",
      "Epoch 44/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 113ms/step - accuracy: 0.9073 - loss: 0.2697 - val_accuracy: 0.8656 - val_loss: 0.4409\n",
      "Epoch 45/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 114ms/step - accuracy: 0.9101 - loss: 0.2590 - val_accuracy: 0.8668 - val_loss: 0.4596\n",
      "Epoch 46/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 114ms/step - accuracy: 0.9082 - loss: 0.2638 - val_accuracy: 0.8522 - val_loss: 0.4907\n",
      "Epoch 47/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m80s\u001b[0m 113ms/step - accuracy: 0.9149 - loss: 0.2519 - val_accuracy: 0.8600 - val_loss: 0.4572\n",
      "Epoch 48/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m79s\u001b[0m 113ms/step - accuracy: 0.9154 - loss: 0.2475 - val_accuracy: 0.8542 - val_loss: 0.4735\n",
      "Epoch 49/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 116ms/step - accuracy: 0.9126 - loss: 0.2498 - val_accuracy: 0.8628 - val_loss: 0.4717\n",
      "Epoch 50/50\n",
      "\u001b[1m704/704\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m81s\u001b[0m 115ms/step - accuracy: 0.9165 - loss: 0.2455 - val_accuracy: 0.8586 - val_loss: 0.4725\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.src.callbacks.history.History at 0x22c3db6a650>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# компилируем и обучаем модель\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) Проанализировали качество обученной модели на тестовой выборке. Определили значения функции потерь и метрики точности классификации."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "id": "SaDxydiyLmRX"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 15ms/step - accuracy: 0.8549 - loss: 0.5139\n",
      "Loss on test data: 0.5139228701591492\n",
      "Accuracy on test data: 0.8549000024795532\n"
     ]
    }
   ],
   "source": [
    "# Оценка качества работы модели на тестовых данных\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) Протестировали модель на двух изображениях из тестовой выборки. Визуализировали изображения и сопоставили истинные метки с предсказаниями нейронной сети."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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 64ms/step\n",
      "NN output: [[1.2917297e-03 1.5173923e-03 7.3140259e-03 8.7915343e-01 5.2461558e-04\n",
      "  1.0724516e-01 9.8486373e-04 1.8565248e-03 6.8461086e-05 4.3758133e-05]]\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",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 59ms/step\n",
      "NN output: [[1.0608504e-06 1.7653504e-08 9.3135744e-01 1.1895873e-03 9.9542603e-06\n",
      "  6.6781670e-02 6.5458257e-04 5.6261097e-06 5.1841993e-09 1.2788083e-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:  5\n",
      "NN answer:  2\n"
     ]
    }
   ],
   "source": [
    "# вывод двух тестовых изображений и результатов распознавания\n",
    "\n",
    "for n in [3,10]:\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) Сформировали подробный отчет о результатах классификации тестовой выборки и построили матрицу ошибок (confusion matrix)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "id": "od56oyyzM0nw"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 13ms/step\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "    airplane       0.81      0.91      0.86      1004\n",
      "  automobile       0.85      0.97      0.91       985\n",
      "        bird       0.79      0.80      0.80       998\n",
      "         cat       0.76      0.70      0.73       985\n",
      "        deer       0.85      0.84      0.85       992\n",
      "         dog       0.82      0.77      0.79       968\n",
      "        frog       0.86      0.93      0.89      1010\n",
      "       horse       0.91      0.86      0.89      1020\n",
      "        ship       0.97      0.86      0.91      1002\n",
      "       truck       0.93      0.90      0.91      1036\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": [
    "# истинные метки классов\n",
    "true_labels = np.argmax(y_test, axis=1)\n",
    "# предсказанные метки классов\n",
    "predicted_labels = np.argmax(model.predict(X_test), axis=1)\n",
    "\n",
    "# отчет о качестве классификации\n",
    "print(classification_report(true_labels, predicted_labels, target_names=class_names))\n",
    "# вычисление матрицы ошибок\n",
    "conf_matrix = confusion_matrix(true_labels, predicted_labels)\n",
    "# отрисовка матрицы ошибок в виде \"тепловой карты\"\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": [
    "#### Анализ результатов классификации датасета CIFAR-10 показал, что разработанная сверточная нейронная сеть с архитектурой, включающей три блока сверточных слоев с batch normalization и dropout, успешно справилась с задачей классификации цветных изображений.\n",
    "\n",
    "**Общая производительность:** Достигнутая точность классификации составляет 85.49%, что является хорошим результатом для данного датасета, учитывая его сложность (малый размер изображений 32×32, высокая вариативность объектов, наличие фоновых элементов).\n",
    "\n",
    "**Анализ по классам:** Модель демонстрирует различную эффективность для разных категорий объектов:\n",
    "- **Высокая точность (≥90%):** ship (precision 0.97, recall 0.86), truck (precision 0.93, recall 0.90), horse (precision 0.91, recall 0.86) - объекты с четкими геометрическими формами и характерными признаками\n",
    "- **Средняя точность (80-90%):** automobile (precision 0.85, recall 0.97), airplane (precision 0.81, recall 0.91), deer (precision 0.85, recall 0.84), frog (precision 0.86, recall 0.93), dog (precision 0.82, recall 0.77) - объекты с более сложной структурой\n",
    "- **Пониженная точность (<80%):** bird (precision 0.79, recall 0.80), cat (precision 0.76, recall 0.70) - объекты с высокой внутриклассовой вариативностью и схожестью между классами\n",
    "\n",
    "**Особенности классификации:** Наибольшие трудности модель испытывает при классификации кошек (precision 0.76, recall 0.70), что связано с высокой вариативностью этого класса и схожестью с собаками. При этом модель демонстрирует сбалансированные метрики precision и recall для большинства классов, что указывает на отсутствие систематических смещений в предсказаниях. Интересно отметить, что для некоторых классов (automobile, airplane, frog) recall выше precision, что говорит о склонности модели чаще предсказывать эти классы.\n",
    "\n",
    "**Выводы:** Полученные результаты подтверждают эффективность применения сверточных нейронных сетей с batch normalization и dropout для классификации цветных изображений. Архитектура успешно извлекает пространственные признаки различного уровня абстракции, что позволяет достигать высокого качества классификации даже на сложных наборах данных с ограниченным разрешением изображений."
   ]
  }
 ],
 "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": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}