Final Materials/Static_Signs.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from keras.utils.np_utils import to_categorical\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten, Conv2D, MaxPool2D\n",
    "from keras.optimizers import RMSprop\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.callbacks import ReduceLROnPlateau"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Visualizing Data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load files and assign variables to X and Y\n",
    "train = pd.read_csv(\"../Still_Images/sign_mnist_train.csv\")\n",
    "test = pd.read_csv(\"../Still_Images/sign_mnist_test.csv\")\n",
    "\n",
    "# Define train and test set\n",
    "X_train = train.drop(labels=['label'], axis=1)\n",
    "Y_train = train['label']\n",
    "X_test = test.drop(labels=['label'], axis=1)\n",
    "Y_test = test['label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show image from dataset\n",
    "def gen_image(image):\n",
    "    \"\"\"Return 28x28 image given grayscale values\"\"\"\n",
    "    pixels = image.reshape((28,28))\n",
    "    plt.imshow(pixels, cmap='gray')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get image from training set\n",
    "gen_image(np.array(X_train.iloc[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Pre-Processing**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Assign numbers to corresponding label\n",
    "label_dict = {0:'A',1:'B',2:'C',3:'D',4:'E',5:'F',6:'G',7:'H',8:'I',10:'K',11:'L',12:'M',13:'N',14:'O',15:'P',16:'Q',17:'R',\\\n",
    "             18:'S',19:'T',20:'U',21:'V',22:'W',23:'X',24:'Y'}\n",
    "label_dict_rev = {'A':0,'B':1,'C':2,'D':3,'E':4,'F':5,'G':6,'H':7,'I':8,'K':9,'L':10,'M':11,'N':12,'O':13,'P':14,'Q':15,'R':16,\\\n",
    "                 'S':17,'T':18,'U':19,'V':20,'W':21,'X':22,'Y':23}\n",
    "\n",
    "Y_test1 = []\n",
    "for i in Y_test:\n",
    "    Y_test1.append(label_dict_rev.get(label_dict.get(i)))\n",
    "\n",
    "Y_test2 = to_categorical(Y_test1, num_classes = 24)\n",
    "Y_train = train['label']\n",
    "\n",
    "Y_train1 = []\n",
    "for i in Y_train:\n",
    "    Y_train1.append(label_dict_rev.get(label_dict.get(i)))\n",
    "    \n",
    "# Classes labeled as 1-24; Need to change to 0-23 for to_categorical\n",
    "Y_train2 = to_categorical(Y_train1, num_classes = 24)\n",
    "X_train = train.drop(labels=['label'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Normalize pixels\n",
    "X_train1 = X_train/255\n",
    "X_test1 = X_test/255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert data to 2D form to represent height x width\n",
    "X_train2 = X_train1.values.reshape(-1,28,28,1)\n",
    "X_test2 = X_test1.values.reshape(-1,28,28,1)\n",
    "# Split training set into 10% validation and 90% training\n",
    "X_tr, X_val, Y_tr, Y_val = train_test_split(X_train2, Y_train2, test_size = 0.3, random_state=2, stratify=Y_train2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Modeling**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\ProgramData\\Anaconda2\\envs\\py36\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:4070: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# First 4 layers of model\n",
    "model = Sequential()\n",
    "# Layer 1\n",
    "model.add(Conv2D(filters=32, kernel_size=(5,5), padding='Same', activation='relu', input_shape=(28,28,1)))\n",
    "model.add(MaxPool2D(pool_size=(2,2))) # downsampling\n",
    "model.add(Dropout(0.25)) # Dropout reduces overfitting\n",
    "# Layer 2\n",
    "model.add(Conv2D(filters=64,kernel_size=(3,3),padding='Same',activation='relu'))\n",
    "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "# Fully connected layers\n",
    "model.add(Flatten())\n",
    "model.add(Dense(256,activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(24, activation='softmax'))\n",
    "\n",
    "optimizer = RMSprop(lr=0.001, rho=0.9, epsilon=1e-08, decay=0.0)\n",
    "\n",
    "# use categorical crossentropy as loss function\n",
    "model.compile(optimizer=optimizer, loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# Define model parameters\n",
    "epochs = 10\n",
    "batch_size = 64\n",
    "learning_rate_reduction = ReduceLROnPlateau(monitor='val_acc', patience=3, verbose=1, factor=0.5, min_lr=0.00001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 28, 28, 32)        832       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 14, 14, 32)        0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 14, 14, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 14, 14, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 7, 7, 64)          0         \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 7, 7, 64)          0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 3136)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 256)               803072    \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 256)               0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 24)                6168      \n",
      "=================================================================\n",
      "Total params: 828,568\n",
      "Trainable params: 828,568\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary() # get model architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\ProgramData\\Anaconda2\\envs\\py36\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:422: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n",
      "\n",
      "Train on 27455 samples, validate on 8237 samples\n",
      "Epoch 1/10\n",
      "27455/27455 [==============================] - 41s 1ms/step - loss: 1.5915 - accuracy: 0.5056 - val_loss: 0.3228 - val_accuracy: 0.9354\n",
      "Epoch 2/10\n",
      "  128/27455 [..............................] - ETA: 38s - loss: 0.6128 - accuracy: 0.7578"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda2\\envs\\py36\\lib\\site-packages\\keras\\callbacks\\callbacks.py:1042: RuntimeWarning: Reduce LR on plateau conditioned on metric `val_acc` which is not available. Available metrics are: val_loss,val_accuracy,loss,accuracy,lr\n",
      "  (self.monitor, ','.join(list(logs.keys()))), RuntimeWarning\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "27455/27455 [==============================] - 40s 1ms/step - loss: 0.3186 - accuracy: 0.8944 - val_loss: 0.0377 - val_accuracy: 0.9959\n",
      "Epoch 3/10\n",
      "27455/27455 [==============================] - 40s 1ms/step - loss: 0.1124 - accuracy: 0.9629 - val_loss: 0.0066 - val_accuracy: 1.0000\n",
      "Epoch 4/10\n",
      "27455/27455 [==============================] - 43s 2ms/step - loss: 0.0602 - accuracy: 0.9792 - val_loss: 8.6654e-04 - val_accuracy: 1.0000\n",
      "Epoch 5/10\n",
      "27455/27455 [==============================] - 42s 2ms/step - loss: 0.0398 - accuracy: 0.9870 - val_loss: 4.4569e-04 - val_accuracy: 1.0000\n",
      "Epoch 6/10\n",
      "27455/27455 [==============================] - 41s 1ms/step - loss: 0.0287 - accuracy: 0.9907 - val_loss: 1.5071e-04 - val_accuracy: 1.0000\n",
      "Epoch 7/10\n",
      "27455/27455 [==============================] - 41s 1ms/step - loss: 0.0233 - accuracy: 0.9918 - val_loss: 2.9333e-04 - val_accuracy: 1.0000\n",
      "Epoch 8/10\n",
      "27455/27455 [==============================] - 41s 1ms/step - loss: 0.0205 - accuracy: 0.9940 - val_loss: 2.5093e-05 - val_accuracy: 1.0000\n",
      "Epoch 9/10\n",
      "27455/27455 [==============================] - 41s 2ms/step - loss: 0.0192 - accuracy: 0.9940 - val_loss: 2.1653e-05 - val_accuracy: 1.0000\n",
      "Epoch 10/10\n",
      "27455/27455 [==============================] - 41s 1ms/step - loss: 0.0186 - accuracy: 0.9947 - val_loss: 1.1502e-05 - val_accuracy: 1.0000\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.callbacks.History at 0x20c217a3a90>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Fit model - Use first line when predicting and second \"final\" line when using validation set to find # of epochs\n",
    "\n",
    "model.fit(X_train2, Y_train2, batch_size=batch_size, epochs=epochs, validation_data=(X_val,Y_val), callbacks=[learning_rate_reduction])\n",
    "#final = model.fit(X_train2, Y_train2, batch_size=batch_size, epochs=epochs, validation_data=(X_val,Y_val), callbacks=[learning_rate_reduction])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Results**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Plot training and validation accuracy\n",
    "# plt.plot(final.history['accuracy'][1:30])\n",
    "# plt.plot(final.history['val_accuracy'][1:30])\n",
    "# plt.title('model accuracy')\n",
    "# plt.ylabel('accuracy')\n",
    "# plt.xlabel('epoch')\n",
    "# plt.legend(['train', 'validation'])\n",
    "# plt.ylim((0.9,1)) # limit y-axis to 90-100% accuracy\n",
    "# plt.show()\n",
    "\n",
    "# # Plot training and validation loss\n",
    "# plt.plot(final.history['loss'][1:30])\n",
    "# plt.plot(final.history['val_loss'][1:30])\n",
    "# plt.title('model loss')\n",
    "# plt.ylabel('loss')\n",
    "# plt.xlabel('epoch')\n",
    "# plt.legend(['train', 'validation'])\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7172/7172 [==============================] - 3s 354us/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.2862028573632938, 0.9428332448005676]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get CNN loss and test error\n",
    "model.evaluate(X_test2, Y_test2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "results = model.predict(X_test2) # predict test labels\n",
    "Y_pred_classes = np.argmax(results, axis = 1) # Convert predictions classes to one hot vectors \n",
    "Y_true = np.argmax(Y_test2,axis = 1) # Convert validation observations to one hot vectors\n",
    "\n",
    "# Create confusion matrix\n",
    "confusion_mtx = confusion_matrix(Y_true, Y_pred_classes) \n",
    "\n",
    "# Print confusion matrix\n",
    "fig = plt.figure(figsize=(10,8))\n",
    "ax1 = fig.add_subplot()\n",
    "sns.heatmap(confusion_mtx, annot=True, fmt=\"d\");\n",
    "labels=['A','B','C','D','E','F','G','H','I','K','L','M','N','O','P','Q','R',\\\n",
    "             'S','T','U','V','W','X','Y']\n",
    "ax1.xaxis.set_ticklabels(labels); ax1.yaxis.set_ticklabels(labels);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Visualize CNN layer outputs**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import models\n",
    "layer_outputs = [layer.output for layer in model.layers[:10]] # Get outputs of top 10 layers\n",
    "activation_model = models.Model(inputs=model.input, outputs=layer_outputs) # Model returns outputs given inputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.preprocessing import image\n",
    "pix = np.array(X_train.iloc[0]).reshape((28,28)) # reshape input to 28x28\n",
    "img_tensor = image.img_to_array(pix) # convert image to numpy array\n",
    "img_tensor = np.expand_dims(img_tensor, axis=0)\n",
    "img_tensor /= 255. # normalize pixels\n",
    "activations = activation_model.predict(img_tensor) # Returns a list Numpy arrays w/one array per layer activation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda2\\envs\\py36\\lib\\site-packages\\ipykernel_launcher.py:16: RuntimeWarning: invalid value encountered in true_divide\n",
      "  app.launch_new_instance()\n"
     ]
    },
    {
     "ename": "MemoryError",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mMemoryError\u001b[0m                               Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-16-5648c74a1615>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      9\u001b[0m     \u001b[0msize\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mlayer_activation\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;31m#The feature map has shape (1, size, size, n_features).\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     10\u001b[0m     \u001b[0mn_cols\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mn_features\u001b[0m \u001b[1;33m//\u001b[0m \u001b[0mimages_per_row\u001b[0m \u001b[1;31m# Tiles the activation channels in this matrix\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 11\u001b[1;33m     \u001b[0mdisplay_grid\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mzeros\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msize\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mn_cols\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mimages_per_row\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0msize\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     12\u001b[0m     \u001b[1;32mfor\u001b[0m \u001b[0mcol\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mn_cols\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m \u001b[1;31m# Create grid showing the filters\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     13\u001b[0m         \u001b[1;32mfor\u001b[0m \u001b[0mrow\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mimages_per_row\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mMemoryError\u001b[0m: "
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "layer_names = []\n",
    "for layer in model.layers[:10]:\n",
    "    layer_names.append(layer.name) # Names of the layers\n",
    "    \n",
    "images_per_row = 16\n",
    "\n",
    "for layer_name, layer_activation in zip(layer_names, activations): # Displays the feature maps\n",
    "    n_features = layer_activation.shape[-1] # Number of features in the feature map\n",
    "    size = layer_activation.shape[1] #The feature map has shape (1, size, size, n_features).\n",
    "    n_cols = n_features // images_per_row # Tiles the activation channels in this matrix\n",
    "    display_grid = np.zeros((size * n_cols, images_per_row * size))\n",
    "    for col in range(n_cols): # Create grid showing the filters\n",
    "        for row in range(images_per_row):\n",
    "            channel_image = layer_activation[0,:, :, col * images_per_row + row]\n",
    "            channel_image -= channel_image.mean() # Post-processes the feature\n",
    "            channel_image /= channel_image.std()\n",
    "            channel_image *= 64\n",
    "            channel_image += 128\n",
    "            channel_image = np.clip(channel_image, 0, 255).astype('uint8')\n",
    "            display_grid[col * size : (col + 1) * size, # Displays the grid\n",
    "                         row * size : (row + 1) * size] = channel_image\n",
    "    scale = 1. / size\n",
    "    plt.figure(figsize=(scale * display_grid.shape[1],\n",
    "                        scale * display_grid.shape[0]))\n",
    "    plt.title(layer_name)\n",
    "    plt.grid(False)\n",
    "    plt.imshow(display_grid, aspect='auto', cmap='gray')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}