4장. 3절. CNN 알고리즘 - 3.1. LeNet.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ad1dbfe4",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras import Sequential\n",
    "from tensorflow.keras.layers import Conv2D, AveragePooling2D\n",
    "from tensorflow.keras.layers import Flatten, Dense, Input\n",
    " \n",
    "class LeNet:\n",
    "\n",
    "    @staticmethod \n",
    "    def build(input_shape=(32,32,1), activation='sigmoid'):\n",
    "        model = Sequential()\n",
    "        model.add(Input(shape=input_shape))\n",
    "\n",
    "        # 첫 번째 CONV => ACTIVATION => POOL 계층 \n",
    "        model.add(Conv2D(6, 5, activation=activation,\n",
    "                         kernel_initializer='random_uniform'))\n",
    "        model.add(AveragePooling2D(pool_size=(2,2)))\n",
    "\n",
    "        # 두 번째 CONV => ACTIVATION => POOL 계층 \n",
    "        model.add(Conv2D(16, 5, \n",
    "                         activation=activation,\n",
    "                         kernel_initializer='random_uniform'))\n",
    "        model.add(AveragePooling2D(pool_size=(2,2)))\n",
    "\n",
    "\n",
    "        # 첫 번째 FC 계층 \n",
    "        model.add(Flatten())\n",
    "        model.add(Dense(120, activation=activation))\n",
    "        # 두 번째 FC 계층 \n",
    "        model.add(Dense(84, activation=activation))\n",
    "        # 출력층 soft-max 활성화 함수 사용\n",
    "        model.add(Dense(10, activation='softmax'))\n",
    "\n",
    "        return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c368e05b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
       "│ 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\">24</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">24</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</span>)           │             <span style=\"color: #00af00; text-decoration-color: #00af00\">156</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ average_pooling2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">AveragePooling2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">12</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6</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\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">8</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)            │           <span style=\"color: #00af00; text-decoration-color: #00af00\">2,416</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ average_pooling2d_1                  │ (<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\">16</span>)            │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">AveragePooling2D</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\">256</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\">120</span>)                 │          <span style=\"color: #00af00; text-decoration-color: #00af00\">30,840</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\">84</span>)                  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">10,164</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\">850</span> │\n",
       "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
       "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m24\u001b[0m, \u001b[38;5;34m24\u001b[0m, \u001b[38;5;34m6\u001b[0m)           │             \u001b[38;5;34m156\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ average_pooling2d (\u001b[38;5;33mAveragePooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12\u001b[0m, \u001b[38;5;34m12\u001b[0m, \u001b[38;5;34m6\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;34m8\u001b[0m, \u001b[38;5;34m8\u001b[0m, \u001b[38;5;34m16\u001b[0m)            │           \u001b[38;5;34m2,416\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ average_pooling2d_1                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m4\u001b[0m, \u001b[38;5;34m16\u001b[0m)            │               \u001b[38;5;34m0\u001b[0m │\n",
       "│ (\u001b[38;5;33mAveragePooling2D\u001b[0m)                   │                             │                 │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\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;34m120\u001b[0m)                 │          \u001b[38;5;34m30,840\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m84\u001b[0m)                  │          \u001b[38;5;34m10,164\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;34m850\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\">44,426</span> (173.54 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m44,426\u001b[0m (173.54 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\">44,426</span> (173.54 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m44,426\u001b[0m (173.54 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": [
    "model = LeNet.build(input_shape=(28, 28, 1), activation=\"relu\")\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6d64ca2b",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss=\"sparse_categorical_crossentropy\",\n",
    "              optimizer=\"sgd\", \n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1f004f7d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.datasets import mnist\n",
    "\n",
    "(X_train, y_train), (X_test, y_test) = mnist.load_data()\n",
    "X_train = X_train.reshape(-1, 28, 28, 1) / 255.0\n",
    "X_test = X_test.reshape(-1, 28, 28, 1) / 255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c030a976",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 6ms/step - accuracy: 0.1499 - loss: 2.2981\n",
      "Epoch 2/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.2942 - loss: 2.2239\n",
      "Epoch 3/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.7901 - loss: 0.7808\n",
      "Epoch 4/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.8756 - loss: 0.4145\n",
      "Epoch 5/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9032 - loss: 0.3265\n",
      "Epoch 6/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9154 - loss: 0.2854\n",
      "Epoch 7/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9237 - loss: 0.2503\n",
      "Epoch 8/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - accuracy: 0.9289 - loss: 0.2312\n",
      "Epoch 9/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9351 - loss: 0.2129\n",
      "Epoch 10/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9372 - loss: 0.2051\n",
      "Epoch 11/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9429 - loss: 0.1872\n",
      "Epoch 12/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9477 - loss: 0.1757\n",
      "Epoch 13/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9495 - loss: 0.1651\n",
      "Epoch 14/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - accuracy: 0.9527 - loss: 0.1612\n",
      "Epoch 15/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step - accuracy: 0.9550 - loss: 0.1502\n",
      "Epoch 16/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - accuracy: 0.9581 - loss: 0.1416\n",
      "Epoch 17/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - accuracy: 0.9582 - loss: 0.1385\n",
      "Epoch 18/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 9ms/step - accuracy: 0.9588 - loss: 0.1334\n",
      "Epoch 19/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 10ms/step - accuracy: 0.9616 - loss: 0.1216\n",
      "Epoch 20/20\n",
      "\u001b[1m300/300\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 7ms/step - accuracy: 0.9622 - loss: 0.1219\n"
     ]
    }
   ],
   "source": [
    "hist = model.fit(X_train, y_train,\n",
    "                 batch_size=200, epochs=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b693d842",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 96.50%\n"
     ]
    }
   ],
   "source": [
    "loss, accuracy = model.evaluate(X_test, y_test, verbose=0)\n",
    "print(f\"Accuracy: {accuracy * 100:.2f}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "15730d05",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted: 1, Actual: 1\n",
      "Predicted: 7, Actual: 7\n",
      "Predicted: 3, Actual: 3\n",
      "Predicted: 6, Actual: 6\n",
      "Predicted: 9, Actual: 4\n",
      "Predicted: 8, Actual: 8\n",
      "Predicted: 4, Actual: 4\n",
      "Predicted: 3, Actual: 3\n",
      "Predicted: 1, Actual: 1\n",
      "Predicted: 0, Actual: 0\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import cv2\n",
    "\n",
    "samples = np.random.choice(np.arange(0,len(y_test)),size=(10,))\n",
    "\n",
    "for i in samples:\n",
    "    probs = model.predict(X_test[np.newaxis, i], verbose=0)\n",
    "    prediction = probs.argmax(axis=1)\n",
    "    image = (X_test[i] * 255).astype(\"uint8\")\n",
    "    image = cv2.cvtColor(image, cv2.COLOR_GRAY2BGR)\n",
    "    image = cv2.resize(image, (96, 96), \n",
    "                       interpolation=cv2.INTER_LINEAR)\n",
    " \n",
    "    cv2.putText(image, str(prediction[0]), (5, 20),\n",
    "                cv2.FONT_HERSHEY_SIMPLEX, 0.75, (0,255,0), 2)\n",
    "    print(\"Predicted: {}, Actual: {}\".format(prediction[0], y_test[i]))\n",
    "    cv2.imshow(\"Digit\", image)\n",
    "    cv2.waitKey(0)\n",
    "\n",
    "cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f80bb652",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('mnist-lenet.keras')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "52d0db12-d76c-4d9b-9957-b49e83cfc62b",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.11.7"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}