model_rnn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import data_extractor\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout\n",
    "from keras.layers import LSTM\n",
    "import matplotlib.pyplot as plt\n",
    "import importlib\n",
    "import os\n",
    "import saveload_workspace"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define parameters and settings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "root_wd = os.getcwd() # current wd should be model folder\n",
    "data_directory = os.path.join(root_wd, '..', 'Symmetric Experiment Data')\n",
    "file_names = [name for name in os.listdir(data_directory) if name.endswith('mat')]\n",
    "file_indexes = np.arange(12)\n",
    "Condition_Dictionary = {0: 'Short Symmetric', 1: 'Medium Symmetric', 2: 'Long Symmetric'}\n",
    "Condition_Colors_Dict = {0: '#0BB8FD', 1: '#FD8B0B', 2: '#E35D72'}\n",
    "Condition_Time_Dictionary = {0: 500, 1: 850, 2: 1500}\n",
    "Condition_Trials = [[50, 200], [200, 350], [350, 500]]\n",
    "Subject_Dict = {0: 'Predator', 1: 'Prey'}\n",
    "SAMPLE_TIME = 10  # samples coordinate data every 10ms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define functions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# function to obtain data samples \n",
    "def prepare_data(decisions_data, x_data, y_data, x_vel_data, y_vel_data, seq_size, condition_no, role_no, include_velocity):\n",
    "    trial_sample_length = int(Condition_Time_Dictionary[condition_no] / SAMPLE_TIME)\n",
    "    time_dataset, x_dataset, y_dataset, x_vel_dataset, y_vel_dataset, label_data = [], [], [], [], [], []\n",
    "    for file_no in range(len(file_indexes)):\n",
    "        x_sample_data = np.array(x_data[file_no, condition_no, :, role_no, :])\n",
    "        y_sample_data = np.array(y_data[file_no, condition_no, :, role_no, :])\n",
    "        x_vel_sample_data = np.array(x_vel_data[file_no, condition_no, :, role_no, :])\n",
    "        y_vel_sample_data = np.array(y_vel_data[file_no, condition_no, :, role_no, :])\n",
    "        for trial in range(decisions_data.shape[2]):\n",
    "            for start_idx in range(trial_sample_length - seq_size):\n",
    "                if not (np.isnan(np.sum(x_sample_data[trial, start_idx: start_idx + seq_size])) or \\\n",
    "                        np.isnan(np.sum(x_vel_sample_data[trial, start_idx: start_idx + seq_size]))):\n",
    "                    \n",
    "                    time_dataset.append(np.arange(start_idx, start_idx + seq_size) / (trial_sample_length - 1))\n",
    "                    x_dataset.append(x_sample_data[trial, start_idx: start_idx + seq_size])\n",
    "                    y_dataset.append(y_sample_data[trial, start_idx: start_idx + seq_size])\n",
    "                    x_vel_dataset.append(x_vel_sample_data[trial, start_idx: start_idx + seq_size])\n",
    "                    y_vel_dataset.append(y_vel_sample_data[trial, start_idx: start_idx + seq_size])\n",
    "                    \n",
    "                    one_hot_encode = np.zeros(3)\n",
    "                    one_hot_encode[int(decisions_data[file_no, condition_no, trial, role_no])] = 1\n",
    "                    label_data.append(one_hot_encode)\n",
    "\n",
    "    x_dataset_array = np.array(x_dataset)\n",
    "    y_dataset_array = np.array(y_dataset)\n",
    "    x_vel_dataset_array = np.array(x_vel_dataset)\n",
    "    y_vel_dataset_array = np.array(y_vel_dataset)\n",
    "    time_dataset_array = np.array(time_dataset)\n",
    "    \n",
    "    if include_velocity:\n",
    "        input_data_raw = np.array([time_dataset_array, x_dataset_array, y_dataset_array, x_vel_dataset_array, y_vel_dataset_array]).transpose(1, 2, 0)\n",
    "        x_dataset_norm = (x_dataset_array - np.min(x_dataset_array)) / (np.max(x_dataset_array) - np.min(x_dataset_array))\n",
    "        y_dataset_norm = (y_dataset_array - np.min(y_dataset_array)) / (np.max(y_dataset_array) - np.min(y_dataset_array))\n",
    "        x_vel_dataset_norm = (x_vel_dataset_array - np.min(x_vel_dataset_array)) / (np.max(x_vel_dataset_array) - np.min(x_vel_dataset_array))\n",
    "        y_vel_dataset_norm = (y_vel_dataset_array - np.min(y_vel_dataset_array)) / (np.max(y_vel_dataset_array) - np.min(y_vel_dataset_array))\n",
    "        input_data = np.array([time_dataset, x_dataset_norm, y_dataset_norm, x_vel_dataset_norm, y_vel_dataset_norm]).transpose(1, 2, 0)\n",
    "    else:\n",
    "        input_data_raw = np.array([time_dataset_array, x_dataset_array, y_dataset_array]).transpose(1, 2, 0)\n",
    "        x_dataset_norm = (x_dataset_array - np.min(x_dataset_array)) / (np.max(x_dataset_array) - np.min(x_dataset_array))\n",
    "        y_dataset_norm = (y_dataset_array - np.min(y_dataset_array)) / (np.max(y_dataset_array) - np.min(y_dataset_array))\n",
    "        input_data = np.array([time_dataset, x_dataset_norm, y_dataset_norm]).transpose(1, 2, 0)\n",
    "\n",
    "    return input_data_raw, input_data, np.array(label_data) \n",
    "\n",
    "# function to plot trajectory samples\n",
    "def plot_trajectory_sample(input_data_raw_, sample_list):\n",
    "    fig, ax = plt.subplots(figsize=(6, 6))\n",
    "    ax.axis('off')\n",
    "    ax.axis('equal')\n",
    "    ax.add_patch(plt.Circle((0, 0), 0.008, color='k', linewidth=2, fill=False))\n",
    "    ax.add_patch(plt.Circle((0.1, 0.15), 0.022, color='k', linewidth=4, fill=False))\n",
    "    ax.add_patch(plt.Circle((-0.1, 0.15), 0.022, color='k', linewidth=4, fill=False))\n",
    "    \n",
    "    for sample_no in sample_list:\n",
    "        ax.plot(input_data_raw_[sample_no, :, 1], input_data_raw_[sample_no, :, 2], linewidth=3.0, color='tomato')\n",
    "\n",
    "    ax.set_ylim((-0.01, 0.18))\n",
    "    ax.set_xlim((-0.15, 0.15))\n",
    "\n",
    "# function to plot prediction accuracies versus normalized time\n",
    "def plot_predictionprobability(prob_data, input_data_, label_data_):\n",
    "    time_bins = np.arange(0, 1, input_data[0, 1, 0] / 2)\n",
    "    prediction_accuracy = np.sum(label_data_ * prob_data, axis=1)\n",
    "    prediction_trial_time = np.mean(input_data_[:, :, 0], axis=1)\n",
    "    prediction_time_bins = np.digitize(prediction_trial_time, bins=time_bins)\n",
    "    \n",
    "    fig, ax = plt.subplots()\n",
    "    ax.set_xlabel('Normalized Time (%)')\n",
    "    ax.set_ylabel('Decision prediction probability')\n",
    "    ax.set_ylim((0, 1))\n",
    "    ax.set_xlim((0, 100)) \n",
    "    ax.scatter(prediction_trial_time * 100, prediction_accuracy, color=Condition_Colors_Dict[condition_no], s=0.1)\n",
    "    \n",
    "    time_bin_centers = (time_bins[1:] + time_bins[:-1]) / 2 * 100\n",
    "    df = pd.DataFrame({'pred_time': prediction_trial_time, 'pred_bin': prediction_time_bins, 'pred_accuracy': prediction_accuracy})\n",
    "    \n",
    "    bin_means = df.groupby('pred_bin').agg({'pred_accuracy': ['mean', 'min', 'max']})\n",
    "    \n",
    "    fig, ax = plt.subplots()\n",
    "    ax.set_xlabel('Normalized Time (%)')\n",
    "    ax.set_ylabel('Mean decision prediction probability')\n",
    "    ax.set_ylim((0, 1))\n",
    "    ax.set_xlim((0, 100)) \n",
    "    ax.plot(time_bin_centers[bin_means.index.values], bin_means.iloc[:, 0], color=Condition_Colors_Dict[condition_no], linewidth=2)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extract coordinate and decisions data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting trajectory and decision data\n",
      "Running file no 0\n",
      "Running condition no: 0 and type: 0\n",
      "Running condition no: 1 and type: 1\n",
      "Running condition no: 2 and type: 2\n",
      "Running file no 1\n",
      "Running condition no: 0 and type: 2\n",
      "Running condition no: 1 and type: 0\n",
      "Running condition no: 2 and type: 1\n",
      "Running file no 2\n",
      "Running condition no: 0 and type: 0\n",
      "Running condition no: 1 and type: 2\n",
      "Running condition no: 2 and type: 1\n",
      "Running file no 3\n",
      "Running condition no: 0 and type: 1\n",
      "Running condition no: 1 and type: 2\n",
      "Running condition no: 2 and type: 0\n",
      "Running file no 4\n",
      "Running condition no: 0 and type: 1\n",
      "Running condition no: 1 and type: 0\n",
      "Running condition no: 2 and type: 2\n",
      "Running file no 5\n",
      "Running condition no: 0 and type: 2\n",
      "Running condition no: 1 and type: 1\n",
      "Running condition no: 2 and type: 0\n",
      "Running file no 6\n",
      "Running condition no: 0 and type: 0\n",
      "Running condition no: 1 and type: 1\n",
      "Running condition no: 2 and type: 2\n",
      "Running file no 7\n",
      "Running condition no: 0 and type: 2\n",
      "File: 7, Condition: 0, Role: 0, Trial No: 114 is corrupted\n",
      "Running condition no: 1 and type: 0\n",
      "Running condition no: 2 and type: 1\n",
      "Running file no 8\n",
      "Running condition no: 0 and type: 1\n",
      "Running condition no: 1 and type: 2\n",
      "Running condition no: 2 and type: 0\n",
      "Running file no 9\n",
      "Running condition no: 0 and type: 1\n",
      "Running condition no: 1 and type: 0\n",
      "Running condition no: 2 and type: 2\n",
      "Running file no 10\n",
      "Running condition no: 0 and type: 0\n",
      "Running condition no: 1 and type: 2\n",
      "Running condition no: 2 and type: 1\n",
      "Running file no 11\n",
      "Running condition no: 0 and type: 2\n",
      "Running condition no: 1 and type: 1\n",
      "Running condition no: 2 and type: 0\n"
     ]
    }
   ],
   "source": [
    "decisions_data, x_data, y_data, x_vel_data, y_vel_data = data_extractor.extract_data(directory=data_directory, file_indexes=file_indexes,\n",
    "                                                             sample_time=SAMPLE_TIME, Condition_Dictionary=Condition_Dictionary,\n",
    "                                                             Condition_Time_Dictionary=Condition_Time_Dictionary,\n",
    "                                                             Condition_Trials=Condition_Trials)\n",
    "os.chdir(root_wd)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Saving data variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "decisions_data\n",
      "x_data\n",
      "y_data\n",
      "x_vel_data\n",
      "y_vel_data\n"
     ]
    }
   ],
   "source": [
    "save_variables = ['decisions_data', 'x_data', 'y_data', 'x_vel_data', 'y_vel_data']\n",
    "save_directory = root_wd + '\\Saved Variables and Data\\Coord and Decision Data'\n",
    "save_name = 'data_workspace'\n",
    "saveload_workspace.save_workspace(workspace=save_variables, globals=globals(), save_directory=save_directory,\n",
    "                                   save_session_name=save_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading data variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "decisions_data\n",
      "x_data\n",
      "y_data\n",
      "x_vel_data\n",
      "y_vel_data\n"
     ]
    }
   ],
   "source": [
    "save_directory = root_wd + '\\Saved Variables and Data\\Coord and Decision Data'\n",
    "save_name = 'data_workspace'\n",
    "variables = saveload_workspace.return_workspace(saved_directory=save_directory, saved_session_name=save_name)\n",
    "for variable in variables:\n",
    "    globals()[variable] = variables[variable]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create input and label data\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "sequence_size = 10  # samples in a window of 100 ms\n",
    "condition_no = 1 # 0 - short condition, 1 - medium condition, 2 - long condition\n",
    "role_no = 0 # 0 - predator role, 1 = prey role\n",
    "include_velocity = False # whether to include velocity data as input for RNN\n",
    "input_data_raw, input_data, label_data = prepare_data(decisions_data=decisions_data, x_data=x_data, y_data=y_data, x_vel_data=x_vel_data, y_vel_data=y_vel_data,\\\n",
    "                                                      seq_size=sequence_size, condition_no=condition_no, role_no=role_no, include_velocity=include_velocity)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Trajectory sample visualizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 480x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_trajectory_sample(input_data_raw_=input_data_raw, sample_list=[399, 400, 401, 403, 404])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Splitting into train and test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "TRAIN_PERCENTAGE = 0.8 # proportion of samples to train on\n",
    "TRAIN_SAMPLES = int(TRAIN_PERCENTAGE * label_data.shape[0])\n",
    "train_indexes = np.random.choice(np.arange(label_data.shape[0]), size=TRAIN_SAMPLES, replace=False)\n",
    "test_indexes = np.setdiff1d(np.arange(label_data.shape[0]), train_indexes)\n",
    "input_data_train, label_data_train = input_data[train_indexes, :], label_data[train_indexes, :] # training data\n",
    "input_data_test, label_data_test = input_data[test_indexes, :], label_data[test_indexes, :] # test data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "844/844 [==============================] - 7s 7ms/step - loss: 0.8486\n",
      "Epoch 2/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.7831\n",
      "Epoch 3/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.7595\n",
      "Epoch 4/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.7474\n",
      "Epoch 5/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.7368\n",
      "Epoch 6/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.7261\n",
      "Epoch 7/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.7177\n",
      "Epoch 8/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.7125\n",
      "Epoch 9/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.7080\n",
      "Epoch 10/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.7058\n",
      "Epoch 11/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.7014\n",
      "Epoch 12/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6973\n",
      "Epoch 13/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6929\n",
      "Epoch 14/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6910\n",
      "Epoch 15/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6875\n",
      "Epoch 16/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6851\n",
      "Epoch 17/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6836\n",
      "Epoch 18/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6812\n",
      "Epoch 19/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6790\n",
      "Epoch 20/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6781\n",
      "Epoch 21/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6751\n",
      "Epoch 22/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6732\n",
      "Epoch 23/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6719\n",
      "Epoch 24/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6702\n",
      "Epoch 25/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6673\n",
      "Epoch 26/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6639\n",
      "Epoch 27/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6638\n",
      "Epoch 28/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6617\n",
      "Epoch 29/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6607\n",
      "Epoch 30/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6603\n",
      "Epoch 31/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6572\n",
      "Epoch 32/50\n",
      "844/844 [==============================] - 6s 8ms/step - loss: 0.6555\n",
      "Epoch 33/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6547\n",
      "Epoch 34/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6532\n",
      "Epoch 35/50\n",
      "844/844 [==============================] - 9s 10ms/step - loss: 0.6520\n",
      "Epoch 36/50\n",
      "844/844 [==============================] - 8s 9ms/step - loss: 0.6505\n",
      "Epoch 37/50\n",
      "844/844 [==============================] - 7s 9ms/step - loss: 0.6502\n",
      "Epoch 38/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6490\n",
      "Epoch 39/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6468\n",
      "Epoch 40/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6469\n",
      "Epoch 41/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6462\n",
      "Epoch 42/50\n",
      "844/844 [==============================] - 7s 8ms/step - loss: 0.6449\n",
      "Epoch 43/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6435\n",
      "Epoch 44/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6422\n",
      "Epoch 45/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6410\n",
      "Epoch 46/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6399\n",
      "Epoch 47/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6394\n",
      "Epoch 48/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6385\n",
      "Epoch 49/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6371\n",
      "Epoch 50/50\n",
      "844/844 [==============================] - 6s 7ms/step - loss: 0.6367\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1d88345b550>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(LSTM(64, input_shape=(input_data_train.shape[1], input_data_train.shape[2])))\n",
    "model.add(Dropout(0.6))\n",
    "model.add(Dense(label_data_train.shape[1], activation='softmax'))\n",
    "model.compile(loss='categorical_crossentropy', optimizer='adam')\n",
    "model.fit(x=input_data_train, y=label_data_train, epochs=50, batch_size=128)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_data_predictions = model.predict(x=input_data_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decision prediction accuracy as a function of time\n",
    "\n",
    "### The prediction accuracy increases as time passes in the trial"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_predictionprobability(prob_data=test_data_predictions, input_data_=input_data_test, label_data_=label_data_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "93c7ba2e20fe519c2ce70fe21cbce94564c192acca215071f3b6f4ab17c24aa4"
  },
  "kernelspec": {
   "display_name": "Python 3.9.6 64-bit",
   "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": "3.9.6"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}