main.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "AJ5SW6-Dh62y",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Digit recognizer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FZ20j3uwiSZN",
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Used libraries\n",
    "\n",
    "• tensorflow — TensorFlow is an open source software library for high performance numerical computation.\n",
    "• keras — TensorFlow Keras is an implementation of the Keras API that uses TensorFlow as a backend."
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Used classes\n",
    "• Sequential - model that groups linear set of layers to `tf.keras.Model`.\n",
    "• Dense - densely connected NN layer.\n",
    "• Dropout - randomly sets input values to 0 with the speed that helps to prevent overtraining. Input values not set to 0 are scaled by 1/(1 - speed) so that the sum of all input values remains the same."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "zEWO9gwpiZVK",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from keras.layers import Dense, Dropout\n",
    "from keras.models import Sequential\n",
    "from keras.optimizers import RMSprop\n",
    "from tensorflow import keras"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Model creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "aQbRaMq0jGgf",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def mnist_make_model(image_w: int, image_h: int) -> Sequential:\n",
    "    model = Sequential()\n",
    "    model.add(Dense(1024, activation=\"relu\", input_shape=(image_w * image_h,)))\n",
    "    model.add(Dropout(0.2))     # rate 0.2 - set 20% of inputs to zero\n",
    "    model.add(Dense(1024, activation=\"relu\"))\n",
    "    model.add(Dropout(0.2))\n",
    "    model.add(Dense(10, activation=\"softmax\"))\n",
    "    model.compile(loss=\"categorical_crossentropy\", optimizer=RMSprop(), metrics=[\"accuracy\"])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Neural network training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "dUPiRJUuk6Rp",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def mnist_mlp_train(model: Sequential) -> tuple:\n",
    "    (x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()\n",
    "\n",
    "    image_size = x_train.shape[1]\n",
    "    train_data = x_train.reshape(x_train.shape[0], image_size * image_size)\n",
    "    test_data = x_test.reshape(x_test.shape[0], image_size * image_size)\n",
    "\n",
    "    train_data = train_data.astype(\"float32\")\n",
    "    test_data = test_data.astype(\"float32\")\n",
    "    train_data /= 255.0\n",
    "    test_data /= 255.0\n",
    "\n",
    "    # encode the labels - we have 10 output classes\n",
    "    # 3 -> [0 0 0 1 0 0 0 0 0 0], 5 -> [0 0 0 0 0 1 0 0 0 0]\n",
    "    num_classes = 10\n",
    "    train_labels_cat = keras.utils.to_categorical(y_train, num_classes)\n",
    "    test_labels_cat = keras.utils.to_categorical(y_test, num_classes)\n",
    "\n",
    "    print(\"Training the network...\")\n",
    "\n",
    "    # Start training the network\n",
    "    history = model.fit(\n",
    "        x=train_data,\n",
    "        y=train_labels_cat,\n",
    "        epochs=21,\n",
    "        batch_size=100,\n",
    "        verbose=1,\n",
    "        validation_data=(test_data, test_labels_cat),\n",
    "    )\n",
    "\n",
    "    return model, history"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Model creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ZcHgh8dvk_YV",
    "outputId": "3ce0e994-d1ac-405d-eb0b-31d5ff075ce6",
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training the network...\n",
      "Epoch 1/21\n",
      "600/600 [==============================] - 25s 41ms/step - loss: 0.2211 - accuracy: 0.9322 - val_loss: 0.0956 - val_accuracy: 0.9700\n",
      "Epoch 2/21\n",
      "600/600 [==============================] - 26s 44ms/step - loss: 0.0983 - accuracy: 0.9716 - val_loss: 0.0900 - val_accuracy: 0.9757\n",
      "Epoch 3/21\n",
      "600/600 [==============================] - 25s 41ms/step - loss: 0.0751 - accuracy: 0.9782 - val_loss: 0.0750 - val_accuracy: 0.9796\n",
      "Epoch 4/21\n",
      "600/600 [==============================] - 24s 40ms/step - loss: 0.0661 - accuracy: 0.9826 - val_loss: 0.0948 - val_accuracy: 0.9790\n",
      "Epoch 5/21\n",
      "600/600 [==============================] - 25s 41ms/step - loss: 0.0581 - accuracy: 0.9845 - val_loss: 0.0866 - val_accuracy: 0.9800\n",
      "Epoch 6/21\n",
      "600/600 [==============================] - 30s 50ms/step - loss: 0.0520 - accuracy: 0.9860 - val_loss: 0.1078 - val_accuracy: 0.9785\n",
      "Epoch 7/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0502 - accuracy: 0.9880 - val_loss: 0.1028 - val_accuracy: 0.9805\n",
      "Epoch 8/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0476 - accuracy: 0.9888 - val_loss: 0.0992 - val_accuracy: 0.9816\n",
      "Epoch 9/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0409 - accuracy: 0.9898 - val_loss: 0.1217 - val_accuracy: 0.9811\n",
      "Epoch 10/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0434 - accuracy: 0.9898 - val_loss: 0.1328 - val_accuracy: 0.9808\n",
      "Epoch 11/21\n",
      "600/600 [==============================] - 25s 42ms/step - loss: 0.0389 - accuracy: 0.9911 - val_loss: 0.1110 - val_accuracy: 0.9828\n",
      "Epoch 12/21\n",
      "600/600 [==============================] - 25s 42ms/step - loss: 0.0330 - accuracy: 0.9925 - val_loss: 0.1372 - val_accuracy: 0.9825\n",
      "Epoch 13/21\n",
      "600/600 [==============================] - 24s 40ms/step - loss: 0.0370 - accuracy: 0.9918 - val_loss: 0.1743 - val_accuracy: 0.9798\n",
      "Epoch 14/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0338 - accuracy: 0.9931 - val_loss: 0.1510 - val_accuracy: 0.9818\n",
      "Epoch 15/21\n",
      "600/600 [==============================] - 25s 41ms/step - loss: 0.0333 - accuracy: 0.9931 - val_loss: 0.1485 - val_accuracy: 0.9837\n",
      "Epoch 16/21\n",
      "600/600 [==============================] - 25s 42ms/step - loss: 0.0345 - accuracy: 0.9930 - val_loss: 0.1403 - val_accuracy: 0.9837\n",
      "Epoch 17/21\n",
      "600/600 [==============================] - 27s 45ms/step - loss: 0.0281 - accuracy: 0.9938 - val_loss: 0.2032 - val_accuracy: 0.9820\n",
      "Epoch 18/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0282 - accuracy: 0.9946 - val_loss: 0.1936 - val_accuracy: 0.9824\n",
      "Epoch 19/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0273 - accuracy: 0.9948 - val_loss: 0.2231 - val_accuracy: 0.9822\n",
      "Epoch 20/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0259 - accuracy: 0.9947 - val_loss: 0.1863 - val_accuracy: 0.9828\n",
      "Epoch 21/21\n",
      "600/600 [==============================] - 26s 43ms/step - loss: 0.0281 - accuracy: 0.9945 - val_loss: 0.2352 - val_accuracy: 0.9807\n"
     ]
    }
   ],
   "source": [
    "model = mnist_make_model(image_w=28, image_h=28)\n",
    "model, history = mnist_mlp_train(model)\n",
    "model.save(os.path.join(\"models\", \"mlp_digits_28x28.h5\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Building training graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "history_dict = history.history\n",
    "acc_values = history_dict[\"accuracy\"]\n",
    "val_acc_values = history_dict[\"val_accuracy\"]\n",
    "epochs = range(1, len(acc_values) + 1)\n",
    "plt.plot(epochs, acc_values, \"bo\", label=\"Training acc\")\n",
    "plt.plot(epochs, val_acc_values, \"b\", label=\"Validation acc\")\n",
    "plt.xlabel(\"Epochs\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Evaluate the quality of model training on test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on test data: 98.40%\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy on test data: %.2f%%\" % (val_acc_values[-11] * 100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Digit prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def mlp_digits_predict(model: Sequential, image_file: str) -> np.int64:\n",
    "    image_size = 28\n",
    "    img = keras.preprocessing.image.load_img(image_file, target_size=(image_size, image_size), color_mode=\"grayscale\")\n",
    "    plt.imshow(img)\n",
    "    plt.show()\n",
    "\n",
    "    img_arr = np.expand_dims(img, axis=0)\n",
    "    img_arr = 1 - img_arr / 255.0\n",
    "    img_arr = img_arr.reshape((1, image_size * image_size))\n",
    "    predict_classes = model.predict([img_arr])\n",
    "    result = np.argmax(predict_classes, axis=1)\n",
    "    return result[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Loading the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "model = tf.keras.models.load_model(os.path.join(\"models\", \"mlp_digits_28x28.h5\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Digit recognition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 112ms/step\n",
      "Result for 0: 0\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 28ms/step\n",
      "Result for 1: 1\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 28ms/step\n",
      "Result for 2: 2\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 28ms/step\n",
      "Result for 2_: 2\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 24ms/step\n",
      "Result for 3: 3\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 23ms/step\n",
      "Result for 4: 4\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 25ms/step\n",
      "Result for 5: 5\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 25ms/step\n",
      "Result for 6: 5\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAAALJ0lEQVR4nO3dXYgdhRnG8edxG1v8uMhXlxhDY20ohIJr2YaCtlikGnPR6I2YC0lBWAsKSr2o2Au9DKUqvSjCWoNpsYqgYi7SahqEIKU2q6T5MG2TSsRs12xMLow31axvL3Yix7jnwzMzZ07y/n+wnLMzZ3deDv6dszPnZBwRAnDhu6jpAQAMBrEDSRA7kASxA0kQO5DEVwa5sWVLRmL1qkWD3CSQytH3PtEHp+a80LpSsdteL+k3kkYk/S4itnR6/OpVi/T3V1aV2SSADtbd/F7bdX2/jLc9Ium3km6RtFbSJttr+/19AOpV5m/2dZKORMQ7EfGxpOckbaxmLABVKxP7SkmtrxmOFcs+x/aE7SnbUydOzpXYHIAyaj8aHxGTETEeEePLl47UvTkAbZSJfVpS69G2K4tlAIZQmdj3SFpj+yrbF0u6Q9L2asYCULW+T71FxBnb90p6RfOn3rZGxMHKJgNQqVLn2SNih6QdFc0CoEa8XRZIgtiBJIgdSILYgSSIHUiC2IEkiB1IgtiBJIgdSILYgSSIHUiC2IEkiB1IgtiBJIgdSILYgSSIHUiC2IEkiB1IgtiBJIgdSILYgSSIHUiC2IEkiB1IgtiBJIgdSILYgSSIHUiC2IEkSl2y2fZRSaclzUk6ExHjVQwFoHqlYi/8KCI+qOD3AKgRL+OBJMrGHpJetf2m7YmFHmB7wvaU7akTJ+dKbg5Av8q+jL8+IqZtf13STtv/jIjdrQ+IiElJk5I0fs3XouT2APSp1J49IqaL21lJL0laV8VQAKrXd+y2L7V9+dn7km6SdKCqwQBUq8zL+FFJL9k++3v+GBF/rmQqDI2brxhrbNuv/HdvY9u+EPUde0S8I+maCmcBUCNOvQFJEDuQBLEDSRA7kASxA0lU8UGY9Jo8PQX0ij07kASxA0kQO5AEsQNJEDuQBLEDSRA7kASxA0kQO5AEsQNJEDuQBLEDSRA7kASxA0kQO5AEn2evQOZ/8viWb/+g4/pPT58e0CTohj07kASxA0kQO5AEsQNJEDuQBLEDSRA7kATn2dHR/+KTjuvLnEfP/P6EJnTds9veanvW9oGWZUts77R9uLhdXO+YAMrq5WX805LWn7PsQUm7ImKNpF3F9wCGWNfYI2K3pFPnLN4oaVtxf5ukW6sdC0DV+j1ANxoRM8X99yWNtnug7QnbU7anTpyc63NzAMoqfTQ+IkJSdFg/GRHjETG+fOlI2c0B6FO/sR+3vUKSitvZ6kYCUId+Y98uaXNxf7Okl6sZB0Bdup5nt/2spBskLbN9TNLDkrZIet72XZLelXR7nUOiOT9Z+b1SP3/RJZdUNAnK6hp7RGxqs+rGimcBUCPeLgskQexAEsQOJEHsQBLEDiTBR1yTu/q5n3Vc/y39rdTv/9ORv5b6eVSHPTuQBLEDSRA7kASxA0kQO5AEsQNJEDuQBOfZk/vWz8udR8f5gz07kASxA0kQO5AEsQNJEDuQBLEDSRA7kATn2S9wN18xVuvv57LL5w/27EASxA4kQexAEsQOJEHsQBLEDiRB7EASnGdHR5xHv3B03bPb3mp71vaBlmWP2J62vbf42lDvmADK6uVl/NOS1i+w/PGIGCu+dlQ7FoCqdY09InZLOjWAWQDUqMwBuntt7yte5i9u9yDbE7anbE+dODlXYnMAyug39ickXS1pTNKMpEfbPTAiJiNiPCLGly8d6XNzAMrqK/aIOB4RcxHxqaQnJa2rdiwAVesrdtsrWr69TdKBdo8FMBy6nme3/aykGyQts31M0sOSbrA9JikkHZV0d30jopu6P7OOC0PX2CNi0wKLn6phFgA14u2yQBLEDiRB7EASxA4kQexAEnzENTk+wpoHe3YgCWIHkiB2IAliB5IgdiAJYgeSIHYgCc6znwf4CCuqwJ4dSILYgSSIHUiC2IEkiB1IgtiBJIgdSILz7Bc4Pq+Os9izA0kQO5AEsQNJEDuQBLEDSRA7kASxA0lwnn0I8Hl1DELXPbvtVbZfs/227YO27yuWL7G90/bh4nZx/eMC6FcvL+PPSHogItZK+r6ke2yvlfSgpF0RsUbSruJ7AEOqa+wRMRMRbxX3T0s6JGmlpI2SthUP2ybp1ppmBFCBL3WAzvZqSddKekPSaETMFKvelzTa5mcmbE/Znjpxcq7MrABK6Dl225dJekHS/RHxYeu6iAhJsdDPRcRkRIxHxPjypSOlhgXQv55it71I86E/ExEvFouP215RrF8habaeEQFUoeupN9uW9JSkQxHxWMuq7ZI2S9pS3L5cy4QXgLpPrW2f3tNh7aJat43zRy/n2a+TdKek/bb3Fsse0nzkz9u+S9K7km6vZUIAlegae0S8LsltVt9Y7TgA6sLbZYEkiB1IgtiBJIgdSILYgST4iOsAdPvnnMueh/+qOZeO7tizA0kQO5AEsQNJEDuQBLEDSRA7kASxA0lwnn0IcFllDAJ7diAJYgeSIHYgCWIHkiB2IAliB5IgdiAJYgeSIHYgCWIHkiB2IAliB5IgdiAJYgeSIHYgia6x215l+zXbb9s+aPu+Yvkjtqdt7y2+NtQ/LoB+9fKPV5yR9EBEvGX7cklv2t5ZrHs8In5d33gAqtLL9dlnJM0U90/bPiRpZd2DAajWl/qb3fZqSddKeqNYdK/tfba32l7c5mcmbE/Znjpxcq7ctAD61nPsti+T9IKk+yPiQ0lPSLpa0pjm9/yPLvRzETEZEeMRMb586Uj5iQH0pafYbS/SfOjPRMSLkhQRxyNiLiI+lfSkpHX1jQmgrF6OxlvSU5IORcRjLctXtDzsNkkHqh8PQFV6ORp/naQ7Je23vbdY9pCkTbbHJIWko5LurmE+ABXp5Wj865K8wKod1Y8DoC68gw5IgtiBJIgdSILYgSSIHUiC2IEkiB1IgtiBJIgdSILYgSSIHUiC2IEkiB1IgtiBJBwRg9uYfULSuy2Llkn6YGADfDnDOtuwziUxW7+qnO0bEbF8oRUDjf0LG7enImK8sQE6GNbZhnUuidn6NajZeBkPJEHsQBJNxz7Z8PY7GdbZhnUuidn6NZDZGv2bHcDgNL1nBzAgxA4k0Ujsttfb/pftI7YfbGKGdmwftb2/uAz1VMOzbLU9a/tAy7IltnfaPlzcLniNvYZmG4rLeHe4zHijz13Tlz8f+N/stkck/VvSjyUdk7RH0qaIeHugg7Rh+6ik8Yho/A0Ytn8o6SNJv4+I7xTLfiXpVERsKf5HuTgifjEksz0i6aOmL+NdXK1oRetlxiXdKumnavC56zDX7RrA89bEnn2dpCMR8U5EfCzpOUkbG5hj6EXEbkmnzlm8UdK24v42zf/HMnBtZhsKETETEW8V909LOnuZ8Uafuw5zDUQTsa+U9F7L98c0XNd7D0mv2n7T9kTTwyxgNCJmivvvSxptcpgFdL2M9yCdc5nxoXnu+rn8eVkcoPui6yPiu5JukXRP8XJ1KMX832DDdO60p8t4D8oClxn/TJPPXb+XPy+ridinJa1q+f7KYtlQiIjp4nZW0ksavktRHz97Bd3idrbheT4zTJfxXugy4xqC567Jy583EfseSWtsX2X7Ykl3SNrewBxfYPvS4sCJbF8q6SYN36Wot0vaXNzfLOnlBmf5nGG5jHe7y4yr4eeu8cufR8TAvyRt0PwR+f9I+mUTM7SZ65uS/lF8HWx6NknPav5l3SeaP7Zxl6SlknZJOizpL5KWDNFsf5C0X9I+zYe1oqHZrtf8S/R9kvYWXxuafu46zDWQ5423ywJJcIAOSILYgSSIHUiC2IEkiB1IgtiBJIgdSOL/uCt3CWyMwaQAAAAASUVORK5CYII=\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 28ms/step\n",
      "Result for 7: 7\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 25ms/step\n",
      "Result for 8: 8\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 27ms/step\n",
      "Result for 9: 9\n"
     ]
    }
   ],
   "source": [
    "images_list = sorted(os.listdir(\"source\"))\n",
    "for image_name in images_list:\n",
    "    name, extension = image_name.split('.')\n",
    "    print(f\"Result for {name}: {mlp_digits_predict(model, os.path.join('source', image_name))}\")"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "AI_TextRecognizer.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "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": "3.10.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}