break-linear-example.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.037407,
     "end_time": "2020-11-13T07:01:39.628293",
     "exception": false,
     "start_time": "2020-11-13T07:01:39.590886",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Break linearity example\n",
    "\n",
    "> Why does neural network rules the fitting power since 2006\n",
    "\n",
    "![Simple nn view from wikipedia](https://upload.wikimedia.org/wikipedia/commons/e/e4/Artificial_neural_network.svg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.035762,
     "end_time": "2020-11-13T07:01:39.700782",
     "exception": false,
     "start_time": "2020-11-13T07:01:39.665020",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Dataset\n",
    "### Simplified Human Activity Recognition w/Smartphone\n",
    "\n",
    "See the [dataset detail](https://www.kaggle.com/mboaglio/simplifiedhuarus)\n",
    "\n",
    "> Abstract: Human Activity Recognition database built from the recordings of 30 subjects performing activities of daily living (ADL) while carrying a waist-mounted smartphone with embedded inertial sensors.\n",
    "\n",
    "> Data Set Characteristics: Multivariate, Time-Series\n",
    "\n",
    "> Data Set Information:\n",
    "\n",
    "> The experiments have been carried out with a group of 30 volunteers within an age bracket of 19-48 years. Each person performed six activities (WALKING, WALKINGUPSTAIRS, WALKINGDOWNSTAIRS, SITTING, STANDING, LAYING) wearing a smartphone (Samsung Galaxy S II) on the waist. Using its embedded accelerometer and gyroscope, we captured 3-axial linear acceleration and 3-axial angular velocity at a constant rate of 50Hz. The experiments have been video-recorded to label the data manually. The obtained dataset has been randomly partitioned into two sets, where 70% of the volunteers was selected for generating the training data and 30% the test data.\n",
    "\n",
    "> The sensor signals (accelerometer and gyroscope) were pre-processed by applying noise filters and then sampled in fixed-width sliding windows of 2.56 sec and 50% overlap (128 readings/window). The sensor acceleration signal, which has gravitational and body motion components, was separated using a Butterworth low-pass filter into body acceleration and gravity. The gravitational force is assumed to have only low frequency components, therefore a filter with 0.3 Hz cutoff frequency was used. From each window, a vector of features was obtained by calculating variables from the time and frequency domain.\n",
    "\n",
    "> Check the README.txt file for further details about this dataset.\n",
    "\n",
    "> An updated version of this dataset can be found at (https://archive.ics.uci.edu/ml/datasets/human+activity+recognition+using+smartphones) It includes labels of postural transitions between activities and also the full raw inertial signals instead of the ones pre-processed into windows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:39.781401Z",
     "iopub.status.busy": "2020-11-13T07:01:39.780469Z",
     "iopub.status.idle": "2020-11-13T07:01:40.944454Z",
     "shell.execute_reply": "2020-11-13T07:01:40.943603Z"
    },
    "papermill": {
     "duration": 1.207596,
     "end_time": "2020-11-13T07:01:40.944587",
     "exception": false,
     "start_time": "2020-11-13T07:01:39.736991",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/kaggle/input/simplifiedhuarus/test.csv\n",
      "/kaggle/input/simplifiedhuarus/train.csv\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "for dirname, _, filenames in os.walk('/kaggle/input'):\n",
    "    for filename in filenames:\n",
    "        print(os.path.join(dirname, filename))\n",
    "\n",
    "from sklearn.decomposition import PCA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.036379,
     "end_time": "2020-11-13T07:01:41.018503",
     "exception": false,
     "start_time": "2020-11-13T07:01:40.982124",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Install extra [packages](https://github.com/raynardj/forgebox)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:41.098482Z",
     "iopub.status.busy": "2020-11-13T07:01:41.097609Z",
     "iopub.status.idle": "2020-11-13T07:01:53.405849Z",
     "shell.execute_reply": "2020-11-13T07:01:53.406397Z"
    },
    "papermill": {
     "duration": 12.350739,
     "end_time": "2020-11-13T07:01:53.406565",
     "exception": false,
     "start_time": "2020-11-13T07:01:41.055826",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "!pip install -q forgebox"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:53.486905Z",
     "iopub.status.busy": "2020-11-13T07:01:53.486130Z",
     "iopub.status.idle": "2020-11-13T07:01:54.705202Z",
     "shell.execute_reply": "2020-11-13T07:01:54.704375Z"
    },
    "papermill": {
     "duration": 1.261875,
     "end_time": "2020-11-13T07:01:54.705340",
     "exception": false,
     "start_time": "2020-11-13T07:01:53.443465",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from forgebox.imports import *\n",
    "from forgebox.ftorch.prepro import split_df\n",
    "from forgebox.html import DOM\n",
    "from forgebox.images.widgets import view_images"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.036743,
     "end_time": "2020-11-13T07:01:54.779171",
     "exception": false,
     "start_time": "2020-11-13T07:01:54.742428",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Train data and test(validation) data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a",
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:54.864686Z",
     "iopub.status.busy": "2020-11-13T07:01:54.863823Z",
     "iopub.status.idle": "2020-11-13T07:01:55.674409Z",
     "shell.execute_reply": "2020-11-13T07:01:55.673541Z"
    },
    "papermill": {
     "duration": 0.858198,
     "end_time": "2020-11-13T07:01:55.674543",
     "exception": false,
     "start_time": "2020-11-13T07:01:54.816345",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# constant\n",
    "DATA = Path(\"/kaggle/input/simplifiedhuarus\")\n",
    "VALID_RATIO = .2\n",
    "\n",
    "def read_data(filename: str) -> pd.DataFrame:return pd.read_csv(DATA/f'{filename}.csv')\n",
    "\n",
    "total_df = read_data(\"train\")\n",
    "# train/valid split\n",
    "train_df, valid_df = split_df(total_df, valid = VALID_RATIO)\n",
    "\n",
    "test_df = read_data(\"test\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:55.777399Z",
     "iopub.status.busy": "2020-11-13T07:01:55.776535Z",
     "iopub.status.idle": "2020-11-13T07:01:55.798302Z",
     "shell.execute_reply": "2020-11-13T07:01:55.798900Z"
    },
    "papermill": {
     "duration": 0.085955,
     "end_time": "2020-11-13T07:01:55.799065",
     "exception": false,
     "start_time": "2020-11-13T07:01:55.713110",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rn</th>\n",
       "      <th>activity</th>\n",
       "      <th>tBodyAcc.mean.X</th>\n",
       "      <th>tBodyAcc.mean.Y</th>\n",
       "      <th>tBodyAcc.mean.Z</th>\n",
       "      <th>tBodyAcc.std.X</th>\n",
       "      <th>tBodyAcc.std.Y</th>\n",
       "      <th>tBodyAcc.std.Z</th>\n",
       "      <th>tBodyAcc.mad.X</th>\n",
       "      <th>tBodyAcc.mad.Y</th>\n",
       "      <th>...</th>\n",
       "      <th>fBodyBodyGyroJerkMag.meanFreq</th>\n",
       "      <th>fBodyBodyGyroJerkMag.skewness</th>\n",
       "      <th>fBodyBodyGyroJerkMag.kurtosis</th>\n",
       "      <th>angle.tBodyAccMean.gravity</th>\n",
       "      <th>angle.tBodyAccJerkMean.gravityMean</th>\n",
       "      <th>angle.tBodyGyroMean.gravityMean</th>\n",
       "      <th>angle.tBodyGyroJerkMean.gravityMean</th>\n",
       "      <th>angle.X.gravityMean</th>\n",
       "      <th>angle.Y.gravityMean</th>\n",
       "      <th>angle.Z.gravityMean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2360</th>\n",
       "      <td>6735</td>\n",
       "      <td>WALKING</td>\n",
       "      <td>0.219</td>\n",
       "      <td>-0.01700</td>\n",
       "      <td>-0.1110</td>\n",
       "      <td>-0.194</td>\n",
       "      <td>-0.1470</td>\n",
       "      <td>-0.283</td>\n",
       "      <td>-0.203</td>\n",
       "      <td>-0.1880</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0255</td>\n",
       "      <td>-0.299</td>\n",
       "      <td>-0.720</td>\n",
       "      <td>0.9400</td>\n",
       "      <td>0.6620</td>\n",
       "      <td>0.3060</td>\n",
       "      <td>-0.1670</td>\n",
       "      <td>-0.843</td>\n",
       "      <td>0.1750</td>\n",
       "      <td>-0.0628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2890</th>\n",
       "      <td>8274</td>\n",
       "      <td>STANDING</td>\n",
       "      <td>0.275</td>\n",
       "      <td>-0.02100</td>\n",
       "      <td>-0.1110</td>\n",
       "      <td>-0.990</td>\n",
       "      <td>-0.9650</td>\n",
       "      <td>-0.993</td>\n",
       "      <td>-0.992</td>\n",
       "      <td>-0.9640</td>\n",
       "      <td>...</td>\n",
       "      <td>0.1190</td>\n",
       "      <td>-0.110</td>\n",
       "      <td>-0.430</td>\n",
       "      <td>0.0180</td>\n",
       "      <td>0.3210</td>\n",
       "      <td>-0.3390</td>\n",
       "      <td>0.6040</td>\n",
       "      <td>-0.801</td>\n",
       "      <td>0.2320</td>\n",
       "      <td>0.0521</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3115</th>\n",
       "      <td>8908</td>\n",
       "      <td>SITTING</td>\n",
       "      <td>0.278</td>\n",
       "      <td>-0.01700</td>\n",
       "      <td>-0.1090</td>\n",
       "      <td>-0.999</td>\n",
       "      <td>-0.9940</td>\n",
       "      <td>-0.992</td>\n",
       "      <td>-0.999</td>\n",
       "      <td>-0.9940</td>\n",
       "      <td>...</td>\n",
       "      <td>0.1290</td>\n",
       "      <td>-0.263</td>\n",
       "      <td>-0.634</td>\n",
       "      <td>-0.1980</td>\n",
       "      <td>-0.1880</td>\n",
       "      <td>0.1700</td>\n",
       "      <td>-0.6350</td>\n",
       "      <td>-0.684</td>\n",
       "      <td>-0.0389</td>\n",
       "      <td>-0.1600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2125</th>\n",
       "      <td>6063</td>\n",
       "      <td>STANDING</td>\n",
       "      <td>0.280</td>\n",
       "      <td>-0.01710</td>\n",
       "      <td>-0.1160</td>\n",
       "      <td>-0.996</td>\n",
       "      <td>-0.9790</td>\n",
       "      <td>-0.964</td>\n",
       "      <td>-0.996</td>\n",
       "      <td>-0.9770</td>\n",
       "      <td>...</td>\n",
       "      <td>0.3300</td>\n",
       "      <td>-0.658</td>\n",
       "      <td>-0.911</td>\n",
       "      <td>-0.1400</td>\n",
       "      <td>-0.2010</td>\n",
       "      <td>-0.3180</td>\n",
       "      <td>0.8180</td>\n",
       "      <td>-0.819</td>\n",
       "      <td>0.2230</td>\n",
       "      <td>0.0336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>486</th>\n",
       "      <td>1355</td>\n",
       "      <td>SITTING</td>\n",
       "      <td>0.277</td>\n",
       "      <td>-0.01670</td>\n",
       "      <td>-0.1110</td>\n",
       "      <td>-0.997</td>\n",
       "      <td>-0.9930</td>\n",
       "      <td>-0.989</td>\n",
       "      <td>-0.997</td>\n",
       "      <td>-0.9930</td>\n",
       "      <td>...</td>\n",
       "      <td>0.2600</td>\n",
       "      <td>-0.444</td>\n",
       "      <td>-0.784</td>\n",
       "      <td>0.0423</td>\n",
       "      <td>-0.0805</td>\n",
       "      <td>-0.0761</td>\n",
       "      <td>-0.3640</td>\n",
       "      <td>-0.977</td>\n",
       "      <td>0.0828</td>\n",
       "      <td>0.0297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1935</th>\n",
       "      <td>5507</td>\n",
       "      <td>STANDING</td>\n",
       "      <td>0.279</td>\n",
       "      <td>-0.01420</td>\n",
       "      <td>-0.1070</td>\n",
       "      <td>-0.998</td>\n",
       "      <td>-0.9860</td>\n",
       "      <td>-0.987</td>\n",
       "      <td>-0.998</td>\n",
       "      <td>-0.9860</td>\n",
       "      <td>...</td>\n",
       "      <td>0.5260</td>\n",
       "      <td>-0.777</td>\n",
       "      <td>-0.920</td>\n",
       "      <td>-0.0222</td>\n",
       "      <td>-0.0249</td>\n",
       "      <td>0.7790</td>\n",
       "      <td>0.0705</td>\n",
       "      <td>-0.862</td>\n",
       "      <td>0.1820</td>\n",
       "      <td>-0.0308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1442</th>\n",
       "      <td>4145</td>\n",
       "      <td>WALKING_UPSTAIRS</td>\n",
       "      <td>0.334</td>\n",
       "      <td>-0.01860</td>\n",
       "      <td>-0.1460</td>\n",
       "      <td>-0.283</td>\n",
       "      <td>-0.0465</td>\n",
       "      <td>-0.270</td>\n",
       "      <td>-0.301</td>\n",
       "      <td>-0.0739</td>\n",
       "      <td>...</td>\n",
       "      <td>0.3190</td>\n",
       "      <td>-0.492</td>\n",
       "      <td>-0.765</td>\n",
       "      <td>-0.6170</td>\n",
       "      <td>-0.1160</td>\n",
       "      <td>-0.9780</td>\n",
       "      <td>0.3540</td>\n",
       "      <td>-0.468</td>\n",
       "      <td>0.3870</td>\n",
       "      <td>0.2400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3605</th>\n",
       "      <td>10278</td>\n",
       "      <td>WALKING_UPSTAIRS</td>\n",
       "      <td>0.344</td>\n",
       "      <td>0.00479</td>\n",
       "      <td>-0.1220</td>\n",
       "      <td>-0.320</td>\n",
       "      <td>-0.0667</td>\n",
       "      <td>-0.182</td>\n",
       "      <td>-0.380</td>\n",
       "      <td>-0.0710</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.0292</td>\n",
       "      <td>0.181</td>\n",
       "      <td>-0.250</td>\n",
       "      <td>-0.4030</td>\n",
       "      <td>-0.7060</td>\n",
       "      <td>0.7390</td>\n",
       "      <td>0.8710</td>\n",
       "      <td>-0.653</td>\n",
       "      <td>0.2780</td>\n",
       "      <td>0.1800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3418</th>\n",
       "      <td>9746</td>\n",
       "      <td>STANDING</td>\n",
       "      <td>0.291</td>\n",
       "      <td>-0.03390</td>\n",
       "      <td>-0.1800</td>\n",
       "      <td>-0.895</td>\n",
       "      <td>-0.7250</td>\n",
       "      <td>-0.796</td>\n",
       "      <td>-0.911</td>\n",
       "      <td>-0.6970</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.1350</td>\n",
       "      <td>-0.440</td>\n",
       "      <td>-0.811</td>\n",
       "      <td>-0.0976</td>\n",
       "      <td>0.4290</td>\n",
       "      <td>-0.3760</td>\n",
       "      <td>-0.0973</td>\n",
       "      <td>-0.634</td>\n",
       "      <td>0.3470</td>\n",
       "      <td>-0.0286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3060</th>\n",
       "      <td>8758</td>\n",
       "      <td>SITTING</td>\n",
       "      <td>0.281</td>\n",
       "      <td>-0.01470</td>\n",
       "      <td>-0.0994</td>\n",
       "      <td>-0.984</td>\n",
       "      <td>-0.9290</td>\n",
       "      <td>-0.960</td>\n",
       "      <td>-0.987</td>\n",
       "      <td>-0.9360</td>\n",
       "      <td>...</td>\n",
       "      <td>0.3690</td>\n",
       "      <td>-0.533</td>\n",
       "      <td>-0.819</td>\n",
       "      <td>-0.0917</td>\n",
       "      <td>0.3260</td>\n",
       "      <td>0.4170</td>\n",
       "      <td>-0.5400</td>\n",
       "      <td>-0.927</td>\n",
       "      <td>0.1000</td>\n",
       "      <td>0.0722</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 563 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         rn          activity  tBodyAcc.mean.X  tBodyAcc.mean.Y  \\\n",
       "2360   6735           WALKING            0.219         -0.01700   \n",
       "2890   8274          STANDING            0.275         -0.02100   \n",
       "3115   8908           SITTING            0.278         -0.01700   \n",
       "2125   6063          STANDING            0.280         -0.01710   \n",
       "486    1355           SITTING            0.277         -0.01670   \n",
       "1935   5507          STANDING            0.279         -0.01420   \n",
       "1442   4145  WALKING_UPSTAIRS            0.334         -0.01860   \n",
       "3605  10278  WALKING_UPSTAIRS            0.344          0.00479   \n",
       "3418   9746          STANDING            0.291         -0.03390   \n",
       "3060   8758           SITTING            0.281         -0.01470   \n",
       "\n",
       "      tBodyAcc.mean.Z  tBodyAcc.std.X  tBodyAcc.std.Y  tBodyAcc.std.Z  \\\n",
       "2360          -0.1110          -0.194         -0.1470          -0.283   \n",
       "2890          -0.1110          -0.990         -0.9650          -0.993   \n",
       "3115          -0.1090          -0.999         -0.9940          -0.992   \n",
       "2125          -0.1160          -0.996         -0.9790          -0.964   \n",
       "486           -0.1110          -0.997         -0.9930          -0.989   \n",
       "1935          -0.1070          -0.998         -0.9860          -0.987   \n",
       "1442          -0.1460          -0.283         -0.0465          -0.270   \n",
       "3605          -0.1220          -0.320         -0.0667          -0.182   \n",
       "3418          -0.1800          -0.895         -0.7250          -0.796   \n",
       "3060          -0.0994          -0.984         -0.9290          -0.960   \n",
       "\n",
       "      tBodyAcc.mad.X  tBodyAcc.mad.Y  ...  fBodyBodyGyroJerkMag.meanFreq  \\\n",
       "2360          -0.203         -0.1880  ...                         0.0255   \n",
       "2890          -0.992         -0.9640  ...                         0.1190   \n",
       "3115          -0.999         -0.9940  ...                         0.1290   \n",
       "2125          -0.996         -0.9770  ...                         0.3300   \n",
       "486           -0.997         -0.9930  ...                         0.2600   \n",
       "1935          -0.998         -0.9860  ...                         0.5260   \n",
       "1442          -0.301         -0.0739  ...                         0.3190   \n",
       "3605          -0.380         -0.0710  ...                        -0.0292   \n",
       "3418          -0.911         -0.6970  ...                        -0.1350   \n",
       "3060          -0.987         -0.9360  ...                         0.3690   \n",
       "\n",
       "      fBodyBodyGyroJerkMag.skewness  fBodyBodyGyroJerkMag.kurtosis  \\\n",
       "2360                         -0.299                         -0.720   \n",
       "2890                         -0.110                         -0.430   \n",
       "3115                         -0.263                         -0.634   \n",
       "2125                         -0.658                         -0.911   \n",
       "486                          -0.444                         -0.784   \n",
       "1935                         -0.777                         -0.920   \n",
       "1442                         -0.492                         -0.765   \n",
       "3605                          0.181                         -0.250   \n",
       "3418                         -0.440                         -0.811   \n",
       "3060                         -0.533                         -0.819   \n",
       "\n",
       "      angle.tBodyAccMean.gravity  angle.tBodyAccJerkMean.gravityMean  \\\n",
       "2360                      0.9400                              0.6620   \n",
       "2890                      0.0180                              0.3210   \n",
       "3115                     -0.1980                             -0.1880   \n",
       "2125                     -0.1400                             -0.2010   \n",
       "486                       0.0423                             -0.0805   \n",
       "1935                     -0.0222                             -0.0249   \n",
       "1442                     -0.6170                             -0.1160   \n",
       "3605                     -0.4030                             -0.7060   \n",
       "3418                     -0.0976                              0.4290   \n",
       "3060                     -0.0917                              0.3260   \n",
       "\n",
       "      angle.tBodyGyroMean.gravityMean  angle.tBodyGyroJerkMean.gravityMean  \\\n",
       "2360                           0.3060                              -0.1670   \n",
       "2890                          -0.3390                               0.6040   \n",
       "3115                           0.1700                              -0.6350   \n",
       "2125                          -0.3180                               0.8180   \n",
       "486                           -0.0761                              -0.3640   \n",
       "1935                           0.7790                               0.0705   \n",
       "1442                          -0.9780                               0.3540   \n",
       "3605                           0.7390                               0.8710   \n",
       "3418                          -0.3760                              -0.0973   \n",
       "3060                           0.4170                              -0.5400   \n",
       "\n",
       "      angle.X.gravityMean  angle.Y.gravityMean  angle.Z.gravityMean  \n",
       "2360               -0.843               0.1750              -0.0628  \n",
       "2890               -0.801               0.2320               0.0521  \n",
       "3115               -0.684              -0.0389              -0.1600  \n",
       "2125               -0.819               0.2230               0.0336  \n",
       "486                -0.977               0.0828               0.0297  \n",
       "1935               -0.862               0.1820              -0.0308  \n",
       "1442               -0.468               0.3870               0.2400  \n",
       "3605               -0.653               0.2780               0.1800  \n",
       "3418               -0.634               0.3470              -0.0286  \n",
       "3060               -0.927               0.1000               0.0722  \n",
       "\n",
       "[10 rows x 563 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_df.sample(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:55.891265Z",
     "iopub.status.busy": "2020-11-13T07:01:55.890108Z",
     "iopub.status.idle": "2020-11-13T07:01:55.894907Z",
     "shell.execute_reply": "2020-11-13T07:01:55.894155Z"
    },
    "papermill": {
     "duration": 0.056453,
     "end_time": "2020-11-13T07:01:55.895037",
     "exception": false,
     "start_time": "2020-11-13T07:01:55.838584",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>activity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LAYING</th>\n",
       "      <td>558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STANDING</th>\n",
       "      <td>553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SITTING</th>\n",
       "      <td>497</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WALKING</th>\n",
       "      <td>487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WALKING_UPSTAIRS</th>\n",
       "      <td>461</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WALKING_DOWNSTAIRS</th>\n",
       "      <td>398</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    activity\n",
       "LAYING                   558\n",
       "STANDING                 553\n",
       "SITTING                  497\n",
       "WALKING                  487\n",
       "WALKING_UPSTAIRS         461\n",
       "WALKING_DOWNSTAIRS       398"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.vc(\"activity\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:55.986960Z",
     "iopub.status.busy": "2020-11-13T07:01:55.986039Z",
     "iopub.status.idle": "2020-11-13T07:01:55.991049Z",
     "shell.execute_reply": "2020-11-13T07:01:55.990307Z"
    },
    "papermill": {
     "duration": 0.056045,
     "end_time": "2020-11-13T07:01:55.991179",
     "exception": false,
     "start_time": "2020-11-13T07:01:55.935134",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>activity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>SITTING</th>\n",
       "      <td>126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAYING</th>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STANDING</th>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WALKING</th>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WALKING_DOWNSTAIRS</th>\n",
       "      <td>95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WALKING_UPSTAIRS</th>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    activity\n",
       "SITTING                  126\n",
       "LAYING                   123\n",
       "STANDING                 115\n",
       "WALKING                  115\n",
       "WALKING_DOWNSTAIRS        95\n",
       "WALKING_UPSTAIRS          80"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "valid_df.vc(\"activity\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.040407,
     "end_time": "2020-11-13T07:01:56.072520",
     "exception": false,
     "start_time": "2020-11-13T07:01:56.032113",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Visualize data signal pattern"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.040936,
     "end_time": "2020-11-13T07:01:56.153750",
     "exception": false,
     "start_time": "2020-11-13T07:01:56.112814",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We have 561 input features at our disposal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:56.257862Z",
     "iopub.status.busy": "2020-11-13T07:01:56.252609Z",
     "iopub.status.idle": "2020-11-13T07:01:56.380280Z",
     "shell.execute_reply": "2020-11-13T07:01:56.379646Z"
    },
    "papermill": {
     "duration": 0.184588,
     "end_time": "2020-11-13T07:01:56.380401",
     "exception": false,
     "start_time": "2020-11-13T07:01:56.195813",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "561"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_nums = train_df.query(\"activity=='LAYING'\").values[:200,2:].shape[1]\n",
    "feature_nums"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:56.527630Z",
     "iopub.status.busy": "2020-11-13T07:01:56.510107Z",
     "iopub.status.idle": "2020-11-13T07:01:56.790661Z",
     "shell.execute_reply": "2020-11-13T07:01:56.790002Z"
    },
    "papermill": {
     "duration": 0.369811,
     "end_time": "2020-11-13T07:01:56.790802",
     "exception": false,
     "start_time": "2020-11-13T07:01:56.420991",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7efed5524750>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.imshow(train_df.query(\"activity=='LAYING'\").values[:200,2:].astype(np.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:56.922154Z",
     "iopub.status.busy": "2020-11-13T07:01:56.911790Z",
     "iopub.status.idle": "2020-11-13T07:01:57.161054Z",
     "shell.execute_reply": "2020-11-13T07:01:57.161733Z"
    },
    "papermill": {
     "duration": 0.326275,
     "end_time": "2020-11-13T07:01:57.161898",
     "exception": false,
     "start_time": "2020-11-13T07:01:56.835623",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7efed2b7aed0>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.imshow(train_df.query(\"activity=='STANDING'\").values[:200,2:].astype(np.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:57.273274Z",
     "iopub.status.busy": "2020-11-13T07:01:57.268299Z",
     "iopub.status.idle": "2020-11-13T07:01:57.541077Z",
     "shell.execute_reply": "2020-11-13T07:01:57.541796Z"
    },
    "papermill": {
     "duration": 0.332774,
     "end_time": "2020-11-13T07:01:57.541965",
     "exception": false,
     "start_time": "2020-11-13T07:01:57.209191",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7efed2af90d0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.imshow(train_df.query(\"activity=='WALKING_UPSTAIRS'\").values[:200,2:].astype(np.float32))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.050825,
     "end_time": "2020-11-13T07:01:57.644377",
     "exception": false,
     "start_time": "2020-11-13T07:01:57.593552",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Build up dataloader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:57.752617Z",
     "iopub.status.busy": "2020-11-13T07:01:57.751873Z",
     "iopub.status.idle": "2020-11-13T07:01:57.754607Z",
     "shell.execute_reply": "2020-11-13T07:01:57.755179Z"
    },
    "papermill": {
     "duration": 0.05955,
     "end_time": "2020-11-13T07:01:57.755368",
     "exception": false,
     "start_time": "2020-11-13T07:01:57.695818",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from torch.utils.data import DataLoader,Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.050991,
     "end_time": "2020-11-13T07:01:57.856911",
     "exception": false,
     "start_time": "2020-11-13T07:01:57.805920",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "We have a pretty balanced dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:57.969545Z",
     "iopub.status.busy": "2020-11-13T07:01:57.968436Z",
     "iopub.status.idle": "2020-11-13T07:01:57.972265Z",
     "shell.execute_reply": "2020-11-13T07:01:57.971498Z"
    },
    "papermill": {
     "duration": 0.063685,
     "end_time": "2020-11-13T07:01:57.972390",
     "exception": false,
     "start_time": "2020-11-13T07:01:57.908705",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "y_map = dict((v,k) for k,v in enumerate(train_df.vc(\"activity\").index))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.051035,
     "end_time": "2020-11-13T07:01:58.075113",
     "exception": false,
     "start_time": "2020-11-13T07:01:58.024078",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Target categories mapped to indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:58.184495Z",
     "iopub.status.busy": "2020-11-13T07:01:58.183604Z",
     "iopub.status.idle": "2020-11-13T07:01:58.187913Z",
     "shell.execute_reply": "2020-11-13T07:01:58.188667Z"
    },
    "papermill": {
     "duration": 0.062135,
     "end_time": "2020-11-13T07:01:58.188841",
     "exception": false,
     "start_time": "2020-11-13T07:01:58.126706",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'LAYING': 0,\n",
       " 'STANDING': 1,\n",
       " 'SITTING': 2,\n",
       " 'WALKING': 3,\n",
       " 'WALKING_UPSTAIRS': 4,\n",
       " 'WALKING_DOWNSTAIRS': 5}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:58.308924Z",
     "iopub.status.busy": "2020-11-13T07:01:58.308076Z",
     "iopub.status.idle": "2020-11-13T07:01:58.312019Z",
     "shell.execute_reply": "2020-11-13T07:01:58.311257Z"
    },
    "papermill": {
     "duration": 0.070952,
     "end_time": "2020-11-13T07:01:58.312163",
     "exception": false,
     "start_time": "2020-11-13T07:01:58.241211",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "class ArrayDs(Dataset):\n",
    "    def __init__(self, df, y_map=y_map):\n",
    "        self.df = df\n",
    "        self.y_map = y_map\n",
    "        self.Xs = self.df.values[:,2:].astype(np.float32)\n",
    "        if 'activity' in self.df.columns:\n",
    "            self.has_y = True\n",
    "            self.Ys = self.df['activity'].apply(lambda x:y_map[x]).values\n",
    "        else:\n",
    "            self.has_y = False\n",
    "        \n",
    "    def __len__(self): return len(self.df)\n",
    "    \n",
    "    def __getitem__(self,idx):\n",
    "        if self.has_y:\n",
    "            return self.Xs[idx],self.Ys[idx]\n",
    "        else:\n",
    "            return self.Xs[idx]\n",
    "    \n",
    "def get_data_dl(df: pd.DataFrame, batch_size: int=128, shuffle=False) -> DataLoader:\n",
    "    ds = ArrayDs(df)\n",
    "    return DataLoader(ds, shuffle=shuffle, batch_size=batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:58.433603Z",
     "iopub.status.busy": "2020-11-13T07:01:58.432709Z",
     "iopub.status.idle": "2020-11-13T07:01:58.692520Z",
     "shell.execute_reply": "2020-11-13T07:01:58.691779Z"
    },
    "papermill": {
     "duration": 0.327749,
     "end_time": "2020-11-13T07:01:58.692665",
     "exception": false,
     "start_time": "2020-11-13T07:01:58.364916",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([ 0.222  , -0.0427 , -0.123  , -0.254  , -0.0688 , -0.379  ,\n",
       "        -0.296  , -0.0765 , -0.38   , -0.117  , -0.211  , -0.394  ,\n",
       "         0.159  , -0.0318 ,  0.397  , -0.177  , -0.72   , -0.831  ,\n",
       "        -0.823  , -0.361  , -0.211  , -0.367  ,  0.272  ,  0.26   ,\n",
       "         0.237  , -0.48   ,  0.423  , -0.23   ,  0.198  , -0.333  ,\n",
       "         0.327  , -0.0314 ,  0.107  , -0.526  ,  0.558  , -0.516  ,\n",
       "         0.233  , -0.0192 , -0.0017 ,  0.331  ,  0.945  , -0.228  ,\n",
       "        -0.101  , -0.985  , -0.986  , -0.953  , -0.987  , -0.987  ,\n",
       "        -0.96   ,  0.876  , -0.246  , -0.0998 ,  0.959  , -0.204  ,\n",
       "        -0.115  , -0.183  ,  0.85   , -0.915  , -0.978  , -0.989  ,\n",
       "        -0.989  , -0.971  , -1.     , -1.     , -1.     , -0.301  ,\n",
       "         0.375  , -0.444  ,  0.507  , -0.0427 ,  0.0378 , -0.0945 ,\n",
       "         0.173  , -0.477  ,  0.535  , -0.593  ,  0.648  ,  0.534  ,\n",
       "        -0.835  , -0.681  , -0.283  ,  0.472  ,  0.0112 , -0.248  ,\n",
       "        -0.197  , -0.6    , -0.21   , -0.141  , -0.581  , -0.329  ,\n",
       "        -0.3    , -0.788  ,  0.404  ,  0.459  ,  0.509  , -0.277  ,\n",
       "        -0.714  , -0.673  , -0.915  , -0.116  , -0.178  , -0.614  ,\n",
       "         0.613  ,  0.648  ,  0.66   , -0.482  ,  0.319  , -0.155  ,\n",
       "         0.155  , -0.489  ,  0.378  , -0.172  ,  0.453  , -0.586  ,\n",
       "         0.447  , -0.304  , -0.0694 ,  0.186  ,  0.409  ,  0.16   ,\n",
       "         0.0504 , -0.173  ,  0.0399 , -0.468  , -0.502  , -0.332  ,\n",
       "        -0.537  , -0.52   , -0.323  , -0.231  , -0.656  , -0.265  ,\n",
       "         0.444  ,  0.638  ,  0.447  , -0.351  , -0.867  , -0.875  ,\n",
       "        -0.792  , -0.631  , -0.555  , -0.396  ,  0.44   ,  0.181  ,\n",
       "         0.0598 , -0.348  ,  0.181  ,  0.367  , -0.377  , -0.133  ,\n",
       "         0.0219 ,  0.319  , -0.0999 , -0.43   ,  0.429  , -0.379  ,\n",
       "         0.468  , -0.0756 , -0.166  ,  0.256  , -0.0772 , -0.0751 ,\n",
       "        -0.329  , -0.379  , -0.652  , -0.493  , -0.374  , -0.667  ,\n",
       "        -0.506  , -0.347  , -0.744  , -0.433  ,  0.55   ,  0.727  ,\n",
       "         0.747  , -0.549  , -0.806  , -0.939  , -0.869  , -0.359  ,\n",
       "        -0.721  , -0.568  ,  0.597  ,  0.664  ,  0.478  , -0.156  ,\n",
       "        -0.00726,  0.286  ,  0.293  , -0.0718 ,  0.0316 ,  0.2    ,\n",
       "         0.418  , -0.476  ,  0.302  , -0.158  , -0.095  ,  0.332  ,\n",
       "        -0.0828 , -0.161  , -0.168  , -0.337  , -0.41   , -0.278  ,\n",
       "        -0.774  , -0.168  , -0.65   , -0.45   ,  0.843  , -0.325  ,\n",
       "         0.313  , -0.183  , -0.0401 , -0.168  , -0.337  , -0.41   ,\n",
       "        -0.278  , -0.774  , -0.168  , -0.65   , -0.45   ,  0.843  ,\n",
       "        -0.325  ,  0.313  , -0.183  , -0.0401 , -0.269  , -0.349  ,\n",
       "        -0.343  , -0.367  , -0.204  , -0.269  , -0.736  , -0.4    ,\n",
       "         0.674  , -0.163  ,  0.141  , -0.231  ,  0.184  , -0.36   ,\n",
       "        -0.445  , -0.368  , -0.512  , -0.488  , -0.36   , -0.785  ,\n",
       "        -0.359  ,  0.838  , -0.0493 , -0.133  ,  0.439  , -0.435  ,\n",
       "        -0.547  , -0.614  , -0.625  , -0.665  , -0.56   , -0.547  ,\n",
       "        -0.901  , -0.645  ,  0.862  ,  0.175  , -0.444  ,  0.836  ,\n",
       "        -0.639  , -0.159  , -0.0488 , -0.467  , -0.294  , -0.139  ,\n",
       "        -0.38   , -0.179  , -0.06   , -0.365  , -0.5    , -0.527  ,\n",
       "        -0.545  , -0.593  , -0.54   , -0.879  , -0.101  , -0.72   ,\n",
       "        -0.561  , -0.8    , -0.12   , -0.371  , -0.647  ,  0.632  ,\n",
       "         0.464  ,  0.377  , -0.742  , -0.333  , -0.692  , -0.188  ,\n",
       "         0.0842 , -0.156  , -0.267  , -0.695  , -0.526  , -0.874  ,\n",
       "        -0.439  , -0.756  , -0.743  , -0.751  , -0.701  , -0.757  ,\n",
       "        -0.783  , -0.926  , -0.724  , -0.747  , -0.722  , -0.671  ,\n",
       "        -0.837  , -0.732  , -0.72   , -0.748  , -0.776  , -0.526  ,\n",
       "        -0.694  , -0.807  , -0.905  , -0.791  , -0.737  , -0.756  ,\n",
       "        -0.585  , -0.648  , -0.851  , -0.736  , -0.566  , -0.811  ,\n",
       "        -0.842  , -0.814  , -0.907  , -0.963  , -0.982  , -0.965  ,\n",
       "        -0.977  , -0.992  , -0.814  , -0.928  , -0.976  , -0.981  ,\n",
       "        -0.801  , -0.967  , -0.244  , -0.253  , -0.584  , -0.322  ,\n",
       "        -0.189  , -0.612  , -0.178  , -0.233  , -0.587  , -0.31   ,\n",
       "        -0.288  , -0.678  , -0.45   , -0.58   , -0.714  , -0.242  ,\n",
       "        -0.714  , -0.673  , -0.915  , -0.0721 , -0.379  , -0.662  ,\n",
       "         0.583  ,  0.578  ,  0.212  , -0.48   , -0.6    , -0.12   ,\n",
       "        -0.359  , -0.523  , -0.393  , -0.0892 , -0.442  , -0.215  ,\n",
       "        -0.719  , -0.289  , -0.771  , -0.803  , -0.697  , -0.796  ,\n",
       "        -0.79   , -0.91   , -0.893  , -0.894  , -0.956  , -0.718  ,\n",
       "        -0.746  , -0.895  , -0.888  , -0.696  , -0.775  , -0.835  ,\n",
       "        -0.643  , -0.769  , -0.816  , -0.857  , -0.912  , -0.883  ,\n",
       "        -0.938  , -0.638  , -0.745  , -0.846  , -0.89   , -0.635  ,\n",
       "        -0.826  , -0.875  , -0.848  , -0.907  , -0.963  , -0.981  ,\n",
       "        -0.961  , -0.958  , -0.931  , -0.826  , -0.934  , -0.974  ,\n",
       "        -0.952  , -0.859  , -0.967  , -0.354  , -0.508  , -0.313  ,\n",
       "        -0.504  , -0.501  , -0.399  , -0.375  , -0.487  , -0.344  ,\n",
       "        -0.485  , -0.633  , -0.441  , -0.817  , -0.812  , -0.698  ,\n",
       "        -0.373  , -0.858  , -0.875  , -0.775  , -0.328  , -0.532  ,\n",
       "        -0.528  ,  0.509  ,  0.447  ,  0.386  , -0.933  , -0.935  ,\n",
       "        -0.586  , -0.144  , -0.195  ,  0.0592 , -0.16   , -0.493  ,\n",
       "        -0.434  , -0.801  ,  0.105  , -0.174  , -0.911  , -0.655  ,\n",
       "        -0.838  , -0.968  , -0.915  , -0.923  , -0.953  , -0.989  ,\n",
       "        -0.865  , -0.857  , -0.91   , -0.969  , -0.859  , -0.951  ,\n",
       "        -0.89   , -0.917  , -0.966  , -0.983  , -0.982  , -0.936  ,\n",
       "        -0.934  , -0.993  , -0.872  , -0.963  , -0.972  , -0.951  ,\n",
       "        -0.865  , -0.979  , -0.804  , -0.948  , -0.869  , -0.965  ,\n",
       "        -0.925  , -0.932  , -0.832  , -0.854  , -0.796  , -0.859  ,\n",
       "        -0.926  , -0.842  , -0.78   , -0.953  , -0.28   , -0.475  ,\n",
       "        -0.359  , -0.603  , -0.861  , -0.28   , -0.776  , -0.409  ,\n",
       "         0.517  , -0.448  ,  0.0835 , -0.441  , -0.715  , -0.296  ,\n",
       "        -0.427  , -0.334  , -0.527  , -0.916  , -0.296  , -0.783  ,\n",
       "        -0.398  ,  0.369  , -0.873  , -0.122  , -0.35   , -0.683  ,\n",
       "        -0.481  , -0.516  , -0.417  , -0.598  , -0.945  , -0.481  ,\n",
       "        -0.845  , -0.387  ,  0.504  , -0.846  , -0.0575 , -0.518  ,\n",
       "        -0.808  , -0.615  , -0.64   , -0.604  , -0.726  , -0.964  ,\n",
       "        -0.615  , -0.925  , -0.652  ,  0.367  , -0.873  ,  0.0549 ,\n",
       "        -0.357  , -0.757  ,  0.292  ,  0.683  , -0.8    , -0.168  ,\n",
       "        -0.77   ,  0.24   ,  0.0918 ], dtype=float32),\n",
       " 3)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x, y = ArrayDs(train_df,)[6]\n",
    "\n",
    "x, y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.054057,
     "end_time": "2020-11-13T07:01:58.800232",
     "exception": false,
     "start_time": "2020-11-13T07:01:58.746175",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Training boiler template\n",
    "\n",
    "> You can also use sklearn at this part"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:01:58.911562Z",
     "iopub.status.busy": "2020-11-13T07:01:58.910806Z",
     "iopub.status.idle": "2020-11-13T07:02:07.494034Z",
     "shell.execute_reply": "2020-11-13T07:02:07.492803Z"
    },
    "papermill": {
     "duration": 8.641096,
     "end_time": "2020-11-13T07:02:07.494238",
     "exception": false,
     "start_time": "2020-11-13T07:01:58.853142",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "!pip install -q pytorch-lightning==1.0.4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:07.629569Z",
     "iopub.status.busy": "2020-11-13T07:02:07.628248Z",
     "iopub.status.idle": "2020-11-13T07:02:09.223090Z",
     "shell.execute_reply": "2020-11-13T07:02:09.226037Z"
    },
    "papermill": {
     "duration": 1.676633,
     "end_time": "2020-11-13T07:02:09.226237",
     "exception": false,
     "start_time": "2020-11-13T07:02:07.549604",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from pytorch_lightning import LightningDataModule, LightningModule\n",
    "import pytorch_lightning as pl\n",
    "\n",
    "class AllData(LightningDataModule):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        \n",
    "    def prepare_data(self):\n",
    "        self.total_df = read_data(\"train\")\n",
    "        # train/valid split\n",
    "        self.train_df, self.valid_df = split_df(\n",
    "            self.total_df, valid = VALID_RATIO)\n",
    "        \n",
    "        self.test_df = read_data(\"test\")\n",
    "    \n",
    "    def train_dataloader(self): return get_data_dl(\n",
    "        self.train_df, shuffle=True)\n",
    "    \n",
    "    def val_dataloader(self): return get_data_dl(self.valid_df)\n",
    "    \n",
    "    def test_dataloader(self): return get_data_dl(self.test_df)\n",
    "    \n",
    "\n",
    "class ltModule(LightningModule):\n",
    "    def __init__(self, base_model):\n",
    "        super().__init__()\n",
    "        self.base_model = base_model\n",
    "        self.crit = nn.CrossEntropyLoss()\n",
    "        self.accuracy = pl.metrics.Accuracy()\n",
    "            \n",
    "    def configure_optimizers(self):\n",
    "        opt = torch.optim.Adam(self.parameters(), lr=1e-3)\n",
    "        return opt\n",
    "    \n",
    "    def forward(self, x): return self.base_model(x)\n",
    "    \n",
    "    def forward_pass(self, batch):\n",
    "        x, y = batch\n",
    "        y_ = self(x)\n",
    "        loss = self.crit(y_,y)\n",
    "        acc = self.accuracy(y_.argmax(dim=-1),y)\n",
    "        return {'loss': loss, 'acc':acc}\n",
    "        \n",
    "    def training_step(self, batch, batch_idx): return self.forward_pass(batch)\n",
    "        \n",
    "    def validation_step(self, batch, batch_idx): return self.forward_pass(batch)\n",
    "    \n",
    "    def print_acc(self, outputs, phase):\n",
    "        avg_acc = torch.stack([x['acc'] for x in outputs]).mean()\n",
    "        print(f\"[{phase}]\\tAccuracy:\\t{int(avg_acc.item()*100)}%\", end=\"\\t\")\n",
    "    \n",
    "    def training_epoch_end(self, outputs):\n",
    "        self.print_acc(outputs, \"TRAIN\")\n",
    "\n",
    "    def validation_epoch_end(self, outputs):\n",
    "        self.print_acc(outputs, \"VALID\")\n",
    "\n",
    "def learn(base_model, max_epochs=5):\n",
    "    \"\"\"\n",
    "    Train the model automatially, the entire pipeline\n",
    "    \"\"\"\n",
    "    all_data = AllData()\n",
    "    module = ltModule(base_model)\n",
    "    trainer = pl.Trainer(max_epochs=max_epochs)\n",
    "    trainer.fit(model = module, datamodule=all_data, )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.053492,
     "end_time": "2020-11-13T07:02:09.332971",
     "exception": false,
     "start_time": "2020-11-13T07:02:09.279479",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Linear relation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:09.446803Z",
     "iopub.status.busy": "2020-11-13T07:02:09.445988Z",
     "iopub.status.idle": "2020-11-13T07:02:18.147948Z",
     "shell.execute_reply": "2020-11-13T07:02:18.148809Z"
    },
    "papermill": {
     "duration": 8.763325,
     "end_time": "2020-11-13T07:02:18.148991",
     "exception": false,
     "start_time": "2020-11-13T07:02:09.385666",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: False, used: False\n",
      "TPU available: False, using: 0 TPU cores\n",
      "\n",
      "  | Name       | Type             | Params\n",
      "------------------------------------------------\n",
      "0 | base_model | Linear           | 3 K   \n",
      "1 | crit       | CrossEntropyLoss | 0     \n",
      "2 | accuracy   | Accuracy         | 0     \n",
      "/opt/conda/lib/python3.7/site-packages/pytorch_lightning/utilities/distributed.py:45: UserWarning: The dataloader, val dataloader 0, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 4 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
      "  warnings.warn(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7eb3504ea27e40f6b897570e53850444",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validation sanity check', layout=Layout…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t27%\t"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/pytorch_lightning/utilities/distributed.py:45: UserWarning: The dataloader, train dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 4 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
      "  warnings.warn(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c237b172d9e64593a35770a0428752bd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Training', layout=Layout(flex='2'), max…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4dd8b700c2a34949a925205e7a865e9e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t68%\t[TRAIN]\tAccuracy:\t46%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fb370cebc66e499a959778a6b0e37ff5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t80%\t[TRAIN]\tAccuracy:\t76%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8cd4ec2d379d43b8b397f4fafdcfae9a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t82%\t[TRAIN]\tAccuracy:\t84%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9860fdecd65d48188f140e741977cdea",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t87%\t[TRAIN]\tAccuracy:\t86%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "16996736198a4afaa2b3fff7eab7807f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t87%\t[TRAIN]\tAccuracy:\t88%\t\n"
     ]
    }
   ],
   "source": [
    "# create linear model\n",
    "linear_model = nn.Linear(feature_nums,len(y_map))\n",
    "\n",
    "# learning process\n",
    "learn(linear_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.060826,
     "end_time": "2020-11-13T07:02:18.275687",
     "exception": false,
     "start_time": "2020-11-13T07:02:18.214861",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Break Linearity\n",
    "\n",
    "> Basic neural network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:18.412986Z",
     "iopub.status.busy": "2020-11-13T07:02:18.412148Z",
     "iopub.status.idle": "2020-11-13T07:02:21.806601Z",
     "shell.execute_reply": "2020-11-13T07:02:21.807331Z"
    },
    "papermill": {
     "duration": 3.469498,
     "end_time": "2020-11-13T07:02:21.807510",
     "exception": false,
     "start_time": "2020-11-13T07:02:18.338012",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: False, used: False\n",
      "TPU available: False, using: 0 TPU cores\n",
      "\n",
      "  | Name       | Type             | Params\n",
      "------------------------------------------------\n",
      "0 | base_model | Sequential       | 291 K \n",
      "1 | crit       | CrossEntropyLoss | 0     \n",
      "2 | accuracy   | Accuracy         | 0     \n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7d3c037f340e42e894f81dc7a767c5c4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validation sanity check', layout=Layout…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t19%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7355f79f75bf4dc2afab8fa6379546e8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Training', layout=Layout(flex='2'), max…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "42683a0567814919bd82863788412afd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t86%\t[TRAIN]\tAccuracy:\t84%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dcab492df8e146d18f472ce513b297f3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t89%\t[TRAIN]\tAccuracy:\t94%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b96e7b0094394ad9a0fe9c5f081e08d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t80%\t[TRAIN]\tAccuracy:\t96%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "56553975b2b84da7bf9f7bd34a512fe2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t91%\t[TRAIN]\tAccuracy:\t97%\t"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ccfc1b3c87624187b3816ffbb67e78ec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Validating', layout=Layout(flex='2'), m…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[VALID]\tAccuracy:\t94%\t[TRAIN]\tAccuracy:\t98%\t\n"
     ]
    }
   ],
   "source": [
    "HIDDEN_SIZE = 512\n",
    "\n",
    "# create model structure\n",
    "neural_network = nn.Sequential(\n",
    "    nn.Linear(feature_nums, HIDDEN_SIZE),\n",
    "    nn.BatchNorm1d(HIDDEN_SIZE),\n",
    "    nn.ReLU(),\n",
    "    nn.Linear(HIDDEN_SIZE, len(y_map))\n",
    ")\n",
    "\n",
    "# learning process\n",
    "learn(neural_network)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.068379,
     "end_time": "2020-11-13T07:02:21.949194",
     "exception": false,
     "start_time": "2020-11-13T07:02:21.880815",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Visualize"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.068523,
     "end_time": "2020-11-13T07:02:22.087005",
     "exception": false,
     "start_time": "2020-11-13T07:02:22.018482",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Weights structure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:22.240316Z",
     "iopub.status.busy": "2020-11-13T07:02:22.239302Z",
     "iopub.status.idle": "2020-11-13T07:02:22.242331Z",
     "shell.execute_reply": "2020-11-13T07:02:22.241712Z"
    },
    "papermill": {
     "duration": 0.079604,
     "end_time": "2020-11-13T07:02:22.242455",
     "exception": false,
     "start_time": "2020-11-13T07:02:22.162851",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def print_param_shape(model) -> pd.DataFrame:\n",
    "    \"\"\"print out the parameter shapes of a model\"\"\"\n",
    "    return pd.DataFrame(list({\"name\":k, \"shape\":tuple(p.T.shape)} \n",
    "        for k,p in model.named_parameters()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:22.388087Z",
     "iopub.status.busy": "2020-11-13T07:02:22.387228Z",
     "iopub.status.idle": "2020-11-13T07:02:22.402683Z",
     "shell.execute_reply": "2020-11-13T07:02:22.401906Z"
    },
    "papermill": {
     "duration": 0.090909,
     "end_time": "2020-11-13T07:02:22.402816",
     "exception": false,
     "start_time": "2020-11-13T07:02:22.311907",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>shape</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>weight</td>\n",
       "      <td>(561, 6)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>bias</td>\n",
       "      <td>(6,)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     name     shape\n",
       "0  weight  (561, 6)\n",
       "1    bias      (6,)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_param_shape(linear_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:22.561795Z",
     "iopub.status.busy": "2020-11-13T07:02:22.560864Z",
     "iopub.status.idle": "2020-11-13T07:02:22.565709Z",
     "shell.execute_reply": "2020-11-13T07:02:22.564957Z"
    },
    "papermill": {
     "duration": 0.093229,
     "end_time": "2020-11-13T07:02:22.565840",
     "exception": false,
     "start_time": "2020-11-13T07:02:22.472611",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>shape</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.weight</td>\n",
       "      <td>(561, 512)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.bias</td>\n",
       "      <td>(512,)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.weight</td>\n",
       "      <td>(512,)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.bias</td>\n",
       "      <td>(512,)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.weight</td>\n",
       "      <td>(512, 6)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3.bias</td>\n",
       "      <td>(6,)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       name       shape\n",
       "0  0.weight  (561, 512)\n",
       "1    0.bias      (512,)\n",
       "2  1.weight      (512,)\n",
       "3    1.bias      (512,)\n",
       "4  3.weight    (512, 6)\n",
       "5    3.bias        (6,)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print_param_shape(neural_network)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.070356,
     "end_time": "2020-11-13T07:02:22.707505",
     "exception": false,
     "start_time": "2020-11-13T07:02:22.637149",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Find influence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.06958,
     "end_time": "2020-11-13T07:02:22.847284",
     "exception": false,
     "start_time": "2020-11-13T07:02:22.777704",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Input features, we print out the first 50 features of our input columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:22.994032Z",
     "iopub.status.busy": "2020-11-13T07:02:22.993243Z",
     "iopub.status.idle": "2020-11-13T07:02:22.996731Z",
     "shell.execute_reply": "2020-11-13T07:02:22.997473Z"
    },
    "papermill": {
     "duration": 0.080141,
     "end_time": "2020-11-13T07:02:22.997660",
     "exception": false,
     "start_time": "2020-11-13T07:02:22.917519",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tBodyAcc.mean.X\ttBodyAcc.mean.Y\ttBodyAcc.mean.Z\ttBodyAcc.std.X\ttBodyAcc.std.Y\ttBodyAcc.std.Z\ttBodyAcc.mad.X\ttBodyAcc.mad.Y\ttBodyAcc.mad.Z\ttBodyAcc.max.X\ttBodyAcc.max.Y\ttBodyAcc.max.Z\ttBodyAcc.min.X\ttBodyAcc.min.Y\ttBodyAcc.min.Z\ttBodyAcc.sma\ttBodyAcc.energy.X\ttBodyAcc.energy.Y\ttBodyAcc.energy.Z\ttBodyAcc.iqr.X\ttBodyAcc.iqr.Y\ttBodyAcc.iqr.Z\ttBodyAcc.entropy.X\ttBodyAcc.entropy.Y\ttBodyAcc.entropy.Z\ttBodyAcc.arCoeff.X.1\ttBodyAcc.arCoeff.X.2\ttBodyAcc.arCoeff.X.3\ttBodyAcc.arCoeff.X.4\ttBodyAcc.arCoeff.Y.1\ttBodyAcc.arCoeff.Y.2\ttBodyAcc.arCoeff.Y.3\ttBodyAcc.arCoeff.Y.4\ttBodyAcc.arCoeff.Z.1\ttBodyAcc.arCoeff.Z.2\ttBodyAcc.arCoeff.Z.3\ttBodyAcc.arCoeff.Z.4\ttBodyAcc.correlation.X.Y\ttBodyAcc.correlation.X.Z\ttBodyAcc.correlation.Y.Z\ttGravityAcc.mean.X\ttGravityAcc.mean.Y\ttGravityAcc.mean.Z\ttGravityAcc.std.X\ttGravityAcc.std.Y\ttGravityAcc.std.Z\ttGravityAcc.mad.X\ttGravityAcc.mad.Y\ttGravityAcc.mad.Z\ttGravityAcc.max.X\n"
     ]
    }
   ],
   "source": [
    "x_map = np.array(total_df.columns[2:])\n",
    "print('\\t'.join(x_map[:50]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.076077,
     "end_time": "2020-11-13T07:02:23.145391",
     "exception": false,
     "start_time": "2020-11-13T07:02:23.069314",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Output categories"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:23.295765Z",
     "iopub.status.busy": "2020-11-13T07:02:23.294900Z",
     "iopub.status.idle": "2020-11-13T07:02:23.299547Z",
     "shell.execute_reply": "2020-11-13T07:02:23.298965Z"
    },
    "papermill": {
     "duration": 0.081409,
     "end_time": "2020-11-13T07:02:23.299710",
     "exception": false,
     "start_time": "2020-11-13T07:02:23.218301",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'LAYING': 0,\n",
       " 'STANDING': 1,\n",
       " 'SITTING': 2,\n",
       " 'WALKING': 3,\n",
       " 'WALKING_UPSTAIRS': 4,\n",
       " 'WALKING_DOWNSTAIRS': 5}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:23.452082Z",
     "iopub.status.busy": "2020-11-13T07:02:23.451286Z",
     "iopub.status.idle": "2020-11-13T07:02:23.466173Z",
     "shell.execute_reply": "2020-11-13T07:02:23.465489Z"
    },
    "papermill": {
     "duration": 0.09391,
     "end_time": "2020-11-13T07:02:23.466309",
     "exception": false,
     "start_time": "2020-11-13T07:02:23.372399",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([[-0.0289, -0.0342, -0.0063,  0.0535,  0.0173,  0.0131],\n",
       "         [ 0.0153,  0.0114,  0.0441, -0.0519, -0.0760, -0.0072],\n",
       "         [-0.0165,  0.0282, -0.0022, -0.0003, -0.0733, -0.0364],\n",
       "         ...,\n",
       "         [ 0.1058, -0.0874, -0.1345, -0.0538, -0.0384, -0.0784],\n",
       "         [-0.1187,  0.1129, -0.0106,  0.0588,  0.0483, -0.0284],\n",
       "         [-0.1122,  0.0559,  0.0169,  0.0593,  0.0422,  0.0154]]),\n",
       " torch.Size([561, 6]))"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "linear_weight = linear_model.weight.data.T\n",
    "linear_weight, linear_weight.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:23.618066Z",
     "iopub.status.busy": "2020-11-13T07:02:23.617284Z",
     "iopub.status.idle": "2020-11-13T07:02:23.625533Z",
     "shell.execute_reply": "2020-11-13T07:02:23.624968Z"
    },
    "papermill": {
     "duration": 0.087067,
     "end_time": "2020-11-13T07:02:23.625675",
     "exception": false,
     "start_time": "2020-11-13T07:02:23.538608",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['fBodyAcc.skewness.X', 'tBodyGyro.correlation.Y.Z',\n",
       "       'angle.X.gravityMean', 'tGravityAcc.sma', 'tGravityAcc.energy.X',\n",
       "       'tBodyAcc.correlation.X.Y', 'fBodyAcc.skewness.Z',\n",
       "       'tGravityAccMag.arCoeff1', 'tBodyGyroJerk.arCoeff.X.1',\n",
       "       'tBodyAcc.arCoeff.Y.4', 'tBodyGyro.correlation.X.Y',\n",
       "       'tBodyAccJerk.arCoeff.X.3', 'fBodyAccMag.meanFreq',\n",
       "       'tBodyAcc.correlation.Y.Z', 'tBodyGyroMag.entropy',\n",
       "       'fBodyAcc.kurtosis.X', 'tBodyGyroJerk.correlation.X.Z',\n",
       "       'angle.Y.gravityMean', 'tBodyAccMag.arCoeff2',\n",
       "       'fBodyAccJerk.mad.Y', 'tBodyGyro.correlation.X.Z',\n",
       "       'fBodyAcc.meanFreq.Z', 'tGravityAcc.arCoeff.Y.2',\n",
       "       'tBodyAccJerk.arCoeff.Z.3', 'tBodyGyroJerk.entropy.Y',\n",
       "       'tGravityAcc.mean.X', 'tGravityAcc.max.X', 'fBodyAcc.meanFreq.X',\n",
       "       'tBodyGyroJerk.arCoeff.Z.4', 'fBodyAccJerk.maxInds.X'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "most_influential_rank = linear_weight.abs().sum(-1).argsort(descending=True).numpy()\n",
    "\n",
    "x_map[most_influential_rank][:30]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:23.782416Z",
     "iopub.status.busy": "2020-11-13T07:02:23.781683Z",
     "iopub.status.idle": "2020-11-13T07:02:24.752213Z",
     "shell.execute_reply": "2020-11-13T07:02:24.751474Z"
    },
    "papermill": {
     "duration": 1.052874,
     "end_time": "2020-11-13T07:02:24.752363",
     "exception": false,
     "start_time": "2020-11-13T07:02:23.699489",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import plotly.express as px\n",
    "def plot_heat(w,xname,yname,x,y,x_axis_top=True, height=None):\n",
    "    pca = PCA(1,)\n",
    "    new_order = pca.fit_transform(w)[:,0].argsort()\n",
    "    new_w = w[new_order,:].T\n",
    "\n",
    "    fig = px.imshow(new_w,\n",
    "                    labels = dict(x=xname,y=yname),\n",
    "                    x=x[new_order],y=y)\n",
    "    if x_axis_top:\n",
    "        fig.update_xaxes(side=\"top\")\n",
    "    if height:\n",
    "        fig.update_layout(height=height)\n",
    "    fig.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.073074,
     "end_time": "2020-11-13T07:02:24.898755",
     "exception": false,
     "start_time": "2020-11-13T07:02:24.825681",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Linear visualize\n",
    "> We visualize the influences directly from ```input``` features to ```output``` categories\n",
    "\n",
    "The size of the weights are 561 columns x 6 labels\n",
    "\n",
    "Here we pick the most influential 30 input features for visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:25.051339Z",
     "iopub.status.busy": "2020-11-13T07:02:25.050594Z",
     "iopub.status.idle": "2020-11-13T07:02:26.100426Z",
     "shell.execute_reply": "2020-11-13T07:02:26.099845Z"
    },
    "papermill": {
     "duration": 1.128636,
     "end_time": "2020-11-13T07:02:26.100555",
     "exception": false,
     "start_time": "2020-11-13T07:02:24.971919",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "        <script type=\"text/javascript\">\n",
       "        window.PlotlyConfig = {MathJaxConfig: 'local'};\n",
       "        if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}\n",
       "        if (typeof require !== 'undefined') {\n",
       "        require.undef(\"plotly\");\n",
       "        requirejs.config({\n",
       "            paths: {\n",
       "                'plotly': ['https://cdn.plot.ly/plotly-latest.min']\n",
       "            }\n",
       "        });\n",
       "        require(['plotly'], function(Plotly) {\n",
       "            window._Plotly = Plotly;\n",
       "        });\n",
       "        }\n",
       "        </script>\n",
       "        "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>                            <div id=\"ba6a60e6-fc56-4e72-a39d-b086d615c2be\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"ba6a60e6-fc56-4e72-a39d-b086d615c2be\")) {                    Plotly.newPlot(                        \"ba6a60e6-fc56-4e72-a39d-b086d615c2be\",                        [{\"coloraxis\": \"coloraxis\", \"hovertemplate\": \"features: %{x}<br>activities: %{y}<br>color: %{z}<extra></extra>\", \"type\": \"heatmap\", \"x\": [\"tGravityAcc.energy.X\", \"tBodyAcc.correlation.Y.Z\", \"angle.Y.gravityMean\", \"tGravityAcc.mean.X\", \"tGravityAcc.max.X\", \"tBodyAccJerk.arCoeff.X.3\", \"fBodyAcc.meanFreq.X\", \"tBodyGyro.correlation.X.Y\", \"fBodyAccJerk.maxInds.X\", \"fBodyAcc.skewness.Z\", \"tGravityAcc.arCoeff.Y.2\", \"fBodyAccMag.meanFreq\", \"tGravityAccMag.arCoeff1\", \"tBodyAccJerk.arCoeff.Z.3\", \"tBodyAcc.correlation.X.Y\", \"fBodyAcc.meanFreq.Z\", \"tBodyGyroJerk.arCoeff.Z.4\", \"fBodyAccJerk.mad.Y\", \"tBodyGyroMag.entropy\", \"tBodyGyroJerk.entropy.Y\", \"tBodyAccMag.arCoeff2\", \"tBodyGyro.correlation.X.Z\", \"tBodyGyroJerk.arCoeff.X.1\", \"tBodyGyro.correlation.Y.Z\", \"tBodyGyroJerk.correlation.X.Z\", \"tBodyAcc.arCoeff.Y.4\", \"fBodyAcc.kurtosis.X\", \"angle.X.gravityMean\", \"tGravityAcc.sma\", \"fBodyAcc.skewness.X\"], \"y\": [\"LAYING\", \"STANDING\", \"SITTING\", \"WALKING\", \"WALKING_UPSTAIRS\", \"WALKING_DOWNSTAIRS\"], \"z\": [[-0.12506404519081116, -0.08747697621583939, -0.11867627501487732, -0.09329769760370255, -0.05316799879074097, -0.05644785985350609, -0.10175898671150208, -0.09191207587718964, 0.005786261521279812, -0.04712550714612007, -0.08012846112251282, 0.001711255987174809, 0.06455017626285553, 0.07781440764665604, 0.02916008234024048, 0.023353856056928635, 0.037699244916439056, -0.05578716844320297, -0.061000071465969086, -0.045134808868169785, -0.05054891109466553, 0.06770746409893036, 0.0698365792632103, 0.0894642248749733, 0.09407395124435425, 0.07495778799057007, 0.08325956016778946, 0.10581513494253159, 0.11795762181282043, 0.11568257212638855], [0.13213802874088287, 0.09302938729524612, 0.11292312294244766, 0.054973360151052475, 0.09175565093755722, 0.10410808771848679, 0.05222642049193382, 0.09409697353839874, 0.05181366577744484, 0.08277490735054016, 0.05565440282225609, -0.04008062183856964, 0.01595372147858143, 0.07272207736968994, 0.06634240597486496, -0.014352160505950451, 0.10078015178442001, -0.0688856840133667, -0.04909716919064522, -0.08800720423460007, -0.03751015663146973, -0.09567337483167648, -0.10325893759727478, -0.06096098572015762, -0.11663317680358887, -0.08174449950456619, -0.05933407321572304, -0.08742713183164597, -0.09180673211812973, -0.10596295446157455], [0.09015067666769028, 0.08584014326334, -0.010577443987131119, 0.09876672178506851, 0.12585535645484924, -0.008018727414309978, 0.07378197461366653, 0.013913773000240326, 0.07898343354463577, 0.05402645841240883, 0.023496048524975777, 0.017080724239349365, 0.03457849472761154, -0.034661997109651566, -0.12879928946495056, -0.028677726164460182, -0.05542007461190224, -0.06637433916330338, -0.1052825078368187, -0.026350844651460648, -0.008771996945142746, 0.08296172320842743, 0.0911756232380867, -0.06467489153146744, 0.029325256124138832, 0.050604887306690216, -0.01669660210609436, -0.13450978696346283, -0.04257412627339363, -0.06721900403499603], [0.03413071110844612, 0.08689519017934799, 0.05876747518777847, 0.051809340715408325, 0.032269418239593506, 0.08961115777492523, 0.0009706061100587249, -0.07842012494802475, 0.06870652735233307, -0.03624861687421799, -0.09655920416116714, 0.10931248217821121, 0.09604904800653458, 0.062169257551431656, 0.06791498512029648, 0.11619175970554352, -0.0700753852725029, 0.1019294261932373, 0.093027263879776, 0.08273688703775406, -0.07861855626106262, 0.01643660105764866, -0.025690671056509018, 0.11149439960718155, 0.026136226952075958, -0.1042628064751625, -0.08044981956481934, -0.05378909036517143, -0.09420346468687057, -0.11786263436079025], [-0.014304901473224163, 0.022461744025349617, 0.048283565789461136, -0.011150428093969822, 0.023321667686104774, -0.029097285121679306, -0.06129834055900574, 0.05495767667889595, -0.06490529328584671, 0.09316761046648026, 0.08548198640346527, -0.0951932817697525, -0.09745271503925323, -0.08592412620782852, -0.07144114375114441, -0.08107370138168335, -0.05753544345498085, -0.013151616789400578, 0.023243457078933716, 0.06613173335790634, 0.09749399870634079, -0.06519018858671188, -0.05330369621515274, 0.11609309166669846, -0.1042892262339592, 0.03702179342508316, 0.10962673276662827, -0.038411352783441544, 0.07409635186195374, 0.08526137471199036], [0.064276784658432, -0.008561917580664158, -0.02844838984310627, 0.059077002108097076, 0.0424504280090332, -0.09945227950811386, -0.07398384809494019, 0.0680038258433342, -0.09040766954421997, -0.11034014075994492, 0.029970522969961166, -0.12214552611112595, -0.10173221677541733, 0.03789288550615311, -0.0925128236413002, 0.10994673520326614, 0.03925886005163193, 0.07098936289548874, 0.04876621440052986, 0.06266870349645615, 0.10463086515665054, 0.04631843417882919, -0.06156586483120918, -0.1054362803697586, -0.007687436882406473, 0.053758952766656876, 0.02924841083586216, -0.07840151339769363, -0.07486844062805176, 0.07361344993114471]]}],                        {\"coloraxis\": {\"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"margin\": {\"t\": 60}, \"template\": {\"data\": {\"bar\": [{\"error_x\": {\"color\": \"#2a3f5f\"}, \"error_y\": {\"color\": \"#2a3f5f\"}, \"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"bar\"}], \"barpolar\": [{\"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"barpolar\"}], \"carpet\": [{\"aaxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"baxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"type\": \"carpet\"}], \"choropleth\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"choropleth\"}], \"contour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"contour\"}], \"contourcarpet\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"contourcarpet\"}], \"heatmap\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmap\"}], \"heatmapgl\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmapgl\"}], \"histogram\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"histogram\"}], \"histogram2d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2d\"}], \"histogram2dcontour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2dcontour\"}], \"mesh3d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"mesh3d\"}], \"parcoords\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"parcoords\"}], \"pie\": [{\"automargin\": true, \"type\": \"pie\"}], \"scatter\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter\"}], \"scatter3d\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter3d\"}], \"scattercarpet\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattercarpet\"}], \"scattergeo\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergeo\"}], \"scattergl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergl\"}], \"scattermapbox\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattermapbox\"}], \"scatterpolar\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolar\"}], \"scatterpolargl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolargl\"}], \"scatterternary\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterternary\"}], \"surface\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"surface\"}], \"table\": [{\"cells\": {\"fill\": {\"color\": \"#EBF0F8\"}, \"line\": {\"color\": \"white\"}}, \"header\": {\"fill\": {\"color\": \"#C8D4E3\"}, \"line\": {\"color\": \"white\"}}, \"type\": \"table\"}]}, \"layout\": {\"annotationdefaults\": {\"arrowcolor\": \"#2a3f5f\", \"arrowhead\": 0, \"arrowwidth\": 1}, \"coloraxis\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"colorscale\": {\"diverging\": [[0, \"#8e0152\"], [0.1, \"#c51b7d\"], [0.2, \"#de77ae\"], [0.3, \"#f1b6da\"], [0.4, \"#fde0ef\"], [0.5, \"#f7f7f7\"], [0.6, \"#e6f5d0\"], [0.7, \"#b8e186\"], [0.8, \"#7fbc41\"], [0.9, \"#4d9221\"], [1, \"#276419\"]], \"sequential\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"sequentialminus\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"colorway\": [\"#636efa\", \"#EF553B\", \"#00cc96\", \"#ab63fa\", \"#FFA15A\", \"#19d3f3\", \"#FF6692\", \"#B6E880\", \"#FF97FF\", \"#FECB52\"], \"font\": {\"color\": \"#2a3f5f\"}, \"geo\": {\"bgcolor\": \"white\", \"lakecolor\": \"white\", \"landcolor\": \"#E5ECF6\", \"showlakes\": true, \"showland\": true, \"subunitcolor\": \"white\"}, \"hoverlabel\": {\"align\": \"left\"}, \"hovermode\": \"closest\", \"mapbox\": {\"style\": \"light\"}, \"paper_bgcolor\": \"white\", \"plot_bgcolor\": \"#E5ECF6\", \"polar\": {\"angularaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"radialaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"scene\": {\"xaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"yaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"zaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}}, \"shapedefaults\": {\"line\": {\"color\": \"#2a3f5f\"}}, \"ternary\": {\"aaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"baxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"caxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"title\": {\"x\": 0.05}, \"xaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}, \"yaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}}}, \"xaxis\": {\"constrain\": \"domain\", \"scaleanchor\": \"y\", \"side\": \"top\", \"title\": {\"text\": \"features\"}}, \"yaxis\": {\"autorange\": \"reversed\", \"constrain\": \"domain\", \"title\": {\"text\": \"activities\"}}},                        {\"responsive\": true}                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('ba6a60e6-fc56-4e72-a39d-b086d615c2be');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })                };                });            </script>        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_heat(\n",
    "    w=linear_weight[most_influential_rank][:30],\n",
    "    xname=\"features\",\n",
    "    yname=\"activities\",\n",
    "    x=x_map[most_influential_rank][:30],\n",
    "    y=np.array(list(y_map.keys()))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:26.300309Z",
     "iopub.status.busy": "2020-11-13T07:02:26.295167Z",
     "iopub.status.idle": "2020-11-13T07:02:26.333345Z",
     "shell.execute_reply": "2020-11-13T07:02:26.332714Z"
    },
    "papermill": {
     "duration": 0.157054,
     "end_time": "2020-11-13T07:02:26.333467",
     "exception": false,
     "start_time": "2020-11-13T07:02:26.176413",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/sklearn/decomposition/_pca.py:461: RuntimeWarning:\n",
      "\n",
      "invalid value encountered in true_divide\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>                            <div id=\"193601e9-fe40-4df8-ab40-b8cc4beb7b53\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"193601e9-fe40-4df8-ab40-b8cc4beb7b53\")) {                    Plotly.newPlot(                        \"193601e9-fe40-4df8-ab40-b8cc4beb7b53\",                        [{\"coloraxis\": \"coloraxis\", \"hovertemplate\": \"bias: %{x}<br>activities: %{y}<br>color: %{z}<extra></extra>\", \"type\": \"heatmap\", \"x\": [\"bias_layer\"], \"y\": [\"LAYING\", \"STANDING\", \"SITTING\", \"WALKING\", \"WALKING_UPSTAIRS\", \"WALKING_DOWNSTAIRS\"], \"z\": [[-0.027273820713162422], [0.006649511866271496], [-0.056083038449287415], [0.017834169790148735], [0.015856007114052773], [0.031830865889787674]]}],                        {\"coloraxis\": {\"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"margin\": {\"t\": 60}, \"template\": {\"data\": {\"bar\": [{\"error_x\": {\"color\": \"#2a3f5f\"}, \"error_y\": {\"color\": \"#2a3f5f\"}, \"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"bar\"}], \"barpolar\": [{\"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"barpolar\"}], \"carpet\": [{\"aaxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"baxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"type\": \"carpet\"}], \"choropleth\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"choropleth\"}], \"contour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"contour\"}], \"contourcarpet\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"contourcarpet\"}], \"heatmap\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmap\"}], \"heatmapgl\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmapgl\"}], \"histogram\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"histogram\"}], \"histogram2d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2d\"}], \"histogram2dcontour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2dcontour\"}], \"mesh3d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"mesh3d\"}], \"parcoords\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"parcoords\"}], \"pie\": [{\"automargin\": true, \"type\": \"pie\"}], \"scatter\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter\"}], \"scatter3d\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter3d\"}], \"scattercarpet\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattercarpet\"}], \"scattergeo\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergeo\"}], \"scattergl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergl\"}], \"scattermapbox\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattermapbox\"}], \"scatterpolar\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolar\"}], \"scatterpolargl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolargl\"}], \"scatterternary\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterternary\"}], \"surface\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"surface\"}], \"table\": [{\"cells\": {\"fill\": {\"color\": \"#EBF0F8\"}, \"line\": {\"color\": \"white\"}}, \"header\": {\"fill\": {\"color\": \"#C8D4E3\"}, \"line\": {\"color\": \"white\"}}, \"type\": \"table\"}]}, \"layout\": {\"annotationdefaults\": {\"arrowcolor\": \"#2a3f5f\", \"arrowhead\": 0, \"arrowwidth\": 1}, \"coloraxis\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"colorscale\": {\"diverging\": [[0, \"#8e0152\"], [0.1, \"#c51b7d\"], [0.2, \"#de77ae\"], [0.3, \"#f1b6da\"], [0.4, \"#fde0ef\"], [0.5, \"#f7f7f7\"], [0.6, \"#e6f5d0\"], [0.7, \"#b8e186\"], [0.8, \"#7fbc41\"], [0.9, \"#4d9221\"], [1, \"#276419\"]], \"sequential\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"sequentialminus\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"colorway\": [\"#636efa\", \"#EF553B\", \"#00cc96\", \"#ab63fa\", \"#FFA15A\", \"#19d3f3\", \"#FF6692\", \"#B6E880\", \"#FF97FF\", \"#FECB52\"], \"font\": {\"color\": \"#2a3f5f\"}, \"geo\": {\"bgcolor\": \"white\", \"lakecolor\": \"white\", \"landcolor\": \"#E5ECF6\", \"showlakes\": true, \"showland\": true, \"subunitcolor\": \"white\"}, \"hoverlabel\": {\"align\": \"left\"}, \"hovermode\": \"closest\", \"mapbox\": {\"style\": \"light\"}, \"paper_bgcolor\": \"white\", \"plot_bgcolor\": \"#E5ECF6\", \"polar\": {\"angularaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"radialaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"scene\": {\"xaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"yaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"zaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}}, \"shapedefaults\": {\"line\": {\"color\": \"#2a3f5f\"}}, \"ternary\": {\"aaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"baxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"caxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"title\": {\"x\": 0.05}, \"xaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}, \"yaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}}}, \"xaxis\": {\"constrain\": \"domain\", \"scaleanchor\": \"y\", \"side\": \"top\", \"title\": {\"text\": \"bias\"}}, \"yaxis\": {\"autorange\": \"reversed\", \"constrain\": \"domain\", \"title\": {\"text\": \"activities\"}}},                        {\"responsive\": true}                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('193601e9-fe40-4df8-ab40-b8cc4beb7b53');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })                };                });            </script>        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_heat(\n",
    "    w=linear_model.bias.data.numpy()[None,:],\n",
    "    xname=\"bias\",\n",
    "    yname=\"activities\",\n",
    "    x=np.array([\"bias_layer\",]),\n",
    "    y=np.array(list(y_map.keys()))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:26.495980Z",
     "iopub.status.busy": "2020-11-13T07:02:26.495121Z",
     "iopub.status.idle": "2020-11-13T07:02:26.499196Z",
     "shell.execute_reply": "2020-11-13T07:02:26.498453Z"
    },
    "papermill": {
     "duration": 0.088859,
     "end_time": "2020-11-13T07:02:26.499323",
     "exception": false,
     "start_time": "2020-11-13T07:02:26.410464",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(Linear(in_features=561, out_features=512, bias=True),\n",
       " Linear(in_features=512, out_features=6, bias=True))"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neural_network[0],neural_network[3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.076919,
     "end_time": "2020-11-13T07:02:26.653477",
     "exception": false,
     "start_time": "2020-11-13T07:02:26.576558",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Break linearity\n",
    "> But linearity can not simulate a full spectrum of ```OR, AND, XOR``` logic gates, the weight $W_{ij}$ simply tells \"the bigger/smaller $X_{i}$ the better for lable $Y_{j}$\"\n",
    "\n",
    "Instead of $X_{i} \\times W_{ij} => Y_{j}$, we do $X_{i} \\times L1_{ik} => H_{k}, H_{k}\\times L2_{kj} => Y_{j}$\n",
    "\n",
    "> Neural network just using a linear model to fit input into hidden neurons (summarize input features into a middle layer), then use another linear model to fit hidden neurons into output. Hence breaking the linearity.\n",
    "\n",
    "### Visualize input to hidden layer\n",
    "> We pick first 80 input features and first 30 hidden neurons to visualize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:26.820803Z",
     "iopub.status.busy": "2020-11-13T07:02:26.819697Z",
     "iopub.status.idle": "2020-11-13T07:02:26.876309Z",
     "shell.execute_reply": "2020-11-13T07:02:26.876925Z"
    },
    "papermill": {
     "duration": 0.146964,
     "end_time": "2020-11-13T07:02:26.877099",
     "exception": false,
     "start_time": "2020-11-13T07:02:26.730135",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>                            <div id=\"69928805-2446-45f9-8abc-b2abe8cc47b9\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"69928805-2446-45f9-8abc-b2abe8cc47b9\")) {                    Plotly.newPlot(                        \"69928805-2446-45f9-8abc-b2abe8cc47b9\",                        [{\"coloraxis\": \"coloraxis\", \"hovertemplate\": \"features: %{x}<br>hidden_feature: %{y}<br>color: %{z}<extra></extra>\", \"type\": \"heatmap\", \"x\": [\"tGravityAcc.entropy.Z\", \"tBodyAcc.mean.X\", \"tGravityAcc.std.Y\", \"tBodyAcc.min.Z\", \"tGravityAcc.max.X\", \"tBodyAcc.min.X\", \"tGravityAcc.arCoeff.Y.3\", \"tGravityAcc.correlation.Y.Z\", \"tBodyAcc.mean.Z\", \"tBodyAcc.arCoeff.X.1\", \"tGravityAcc.mean.X\", \"tBodyAcc.correlation.X.Y\", \"tBodyAcc.correlation.X.Z\", \"tBodyAcc.min.Y\", \"tBodyAcc.arCoeff.Y.2\", \"tGravityAcc.arCoeff.Z.1\", \"tBodyAcc.arCoeff.Y.1\", \"tBodyAcc.mean.Y\", \"tGravityAcc.energy.X\", \"tGravityAcc.iqr.Y\", \"tBodyAcc.std.X\", \"tBodyAcc.iqr.Z\", \"tBodyAcc.arCoeff.Z.1\", \"tBodyAcc.arCoeff.X.4\", \"tBodyAcc.mad.Y\", \"tGravityAcc.mad.X\", \"tGravityAcc.iqr.X\", \"tGravityAcc.mad.Y\", \"tBodyAcc.sma\", \"tGravityAcc.arCoeff.Y.4\", \"tGravityAcc.entropy.X\", \"tGravityAcc.mean.Z\", \"tBodyAcc.iqr.Y\", \"tGravityAcc.arCoeff.Z.3\", \"tBodyAcc.entropy.Z\", \"tGravityAcc.arCoeff.Z.2\", \"tGravityAcc.correlation.X.Z\", \"tBodyAcc.arCoeff.Y.3\", \"tGravityAcc.arCoeff.Y.1\", \"tGravityAcc.min.X\", \"tGravityAcc.energy.Y\", \"tGravityAcc.sma\", \"tGravityAcc.arCoeff.X.3\", \"tGravityAcc.arCoeff.Y.2\", \"tBodyAcc.mad.X\", \"tGravityAcc.mad.Z\", \"tBodyAcc.energy.X\", \"tBodyAcc.max.Z\", \"tGravityAcc.max.Z\", \"tBodyAcc.arCoeff.Z.2\", \"tGravityAcc.arCoeff.X.1\", \"tBodyAcc.arCoeff.X.3\", \"tBodyAcc.iqr.X\", \"tBodyAcc.correlation.Y.Z\", \"tGravityAcc.std.Z\", \"tGravityAcc.arCoeff.X.4\", \"tGravityAcc.std.X\", \"tBodyAcc.max.X\", \"tBodyAcc.arCoeff.Z.3\", \"tGravityAcc.arCoeff.X.2\", \"tGravityAcc.min.Z\", \"tGravityAcc.energy.Z\", \"tBodyAcc.max.Y\", \"tGravityAcc.correlation.X.Y\", \"tBodyAcc.mad.Z\", \"tBodyAcc.arCoeff.Y.4\", \"tGravityAcc.iqr.Z\", \"tBodyAcc.energy.Z\", \"tBodyAcc.entropy.X\", \"tGravityAcc.arCoeff.Z.4\", \"tGravityAcc.min.Y\", \"tGravityAcc.entropy.Y\", \"tBodyAcc.entropy.Y\", \"tBodyAcc.energy.Y\", \"tBodyAcc.std.Y\", \"tBodyAcc.arCoeff.Z.4\", \"tBodyAcc.std.Z\", \"tBodyAcc.arCoeff.X.2\", \"tGravityAcc.mean.Y\", \"tGravityAcc.max.Y\"], \"y\": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29], \"z\": [[0.008419793099164963, 0.005119055975228548, 0.022322818636894226, 0.04343750700354576, 0.043932609260082245, 0.00015170089318417013, 0.016189921647310257, 0.02914193645119667, 0.040623825043439865, -0.02882143296301365, 0.049953289330005646, -0.037109069526195526, -0.048585787415504456, -0.008932586759328842, 0.037573374807834625, 0.00018476035620551556, 0.03413630649447441, 0.03498176857829094, 0.06223541125655174, 0.005569877568632364, 0.01965995691716671, -0.006544723641127348, 0.023677358403801918, -0.01148397009819746, 0.005451988894492388, 0.02727740816771984, 0.018991980701684952, 0.007132875733077526, -0.0222215186804533, 0.0007468624389730394, 0.010690080001950264, -0.015944385901093483, 0.03418427333235741, 0.029102664440870285, 0.029618913307785988, -0.016938719898462296, 0.019033534452319145, -0.010861219838261604, -0.019891560077667236, 0.040842242538928986, -0.010709572583436966, -0.019399432465434074, -0.019270306453108788, -0.019647950306534767, -0.002894457196816802, 0.03659532964229584, 0.026150278747081757, 0.02148914337158203, -0.02393462136387825, 0.0333644300699234, 0.03469480574131012, -0.027185242623090744, -0.01937750168144703, 0.003496469696983695, -0.033903125673532486, 0.017996646463871002, 0.022457342594861984, 0.0035125224385410547, -0.004869624972343445, 0.002930507529526949, -0.013253048993647099, -0.05917637422680855, 0.0013645474100485444, -0.01803744211792946, -0.013449656777083874, 0.05286485329270363, -0.03427046164870262, 0.016672711819410324, 0.006566652562469244, -0.011344286613166332, -0.037648145109415054, 0.026003357023000717, -0.019265098497271538, 0.03731296956539154, -0.03165147826075554, 0.018606774508953094, 0.010740585625171661, -0.007186455186456442, -0.005120011977851391, -0.02510991320014], [0.029197927564382553, 0.017353182658553123, 0.028752967715263367, -0.036017704755067825, -0.012122688814997673, 0.00792989693582058, 0.019994419068098068, -0.018156465142965317, 0.014408428221940994, 0.010772701352834702, -0.0058294557966291904, 0.005074128974229097, 0.009283547289669514, 0.032346952706575394, -0.01633227989077568, 0.00025056194863282144, -0.044703174382448196, -0.005141735076904297, -0.023993229493498802, 0.007877775467932224, 0.01599651388823986, 0.01881192810833454, -0.005044803023338318, -0.02738736756145954, -0.023155689239501953, -0.024918261915445328, -0.03170168772339821, 0.03512127324938774, 0.03575783222913742, -0.005978522356599569, -0.016167771071195602, 0.0157347172498703, -0.008653135970234871, 0.028959941118955612, -0.009657996706664562, -0.029964756220579147, 0.03789151459932327, -0.04682048410177231, -0.004873139783740044, 0.02008552849292755, 0.011296406388282776, -0.00948703195899725, 0.002938872901722789, 0.027551665902137756, -0.014492684043943882, -0.018758343532681465, -0.01192854531109333, -0.010536182671785355, -0.006377105135470629, 0.02053409442305565, -0.004356879275292158, -0.02300744690001011, 0.026788752526044846, 0.0017895152559503913, 0.02737886644899845, -0.0060204253531992435, -0.005432923790067434, 0.007489325944334269, -0.04869541525840759, 0.02088231034576893, 0.028487753123044968, 0.006709288340061903, -0.02626626007258892, -0.02192225307226181, -0.015167062170803547, 0.0213872492313385, -0.04553181678056717, -0.028306281194090843, -0.011918585747480392, 0.029708348214626312, -0.03282131999731064, 0.03236648440361023, -0.012213705107569695, -0.025651799514889717, 0.015367195941507816, 0.02489166148006916, -0.008569515310227871, -0.02539045922458172, 0.0046159024350345135, 0.029706832021474838], [-0.03401755541563034, -0.005660173017531633, -0.04992586374282837, -0.030466342344880104, -0.008123334497213364, -0.0344468392431736, -0.021233486011624336, 0.014214546419680119, -0.02935853973031044, -0.029880341142416, 0.020921599119901657, -0.012989349663257599, -0.04826957359910011, -0.039839938282966614, 0.014036335051059723, -0.03687916323542595, -0.0318121612071991, -0.02632071264088154, 0.05038031190633774, -0.015941746532917023, -0.018562424927949905, -0.031600773334503174, 0.022500040009617805, -0.00908862054347992, -0.009001464582979679, -0.02790112793445587, 0.010509192943572998, 0.00834701769053936, -0.030452623963356018, -0.03057226724922657, -0.005227651447057724, -0.005428890231996775, -0.016236860305070877, -0.03352610766887665, -0.008436903357505798, -0.02593224309384823, -0.0008454771596007049, -0.03529655560851097, -0.0176081545650959, 0.06824813038110733, -0.05680498853325844, -0.021931905299425125, -0.010632671415805817, -0.006295823026448488, -0.011622306890785694, 0.024442845955491066, -0.007341011427342892, 0.013295125216245651, -0.027463063597679138, 0.04187744855880737, 0.045691583305597305, 0.045024123042821884, 0.0015664269449189305, 0.023547900840640068, -0.000996186165139079, 0.008619294501841068, 0.02007896639406681, 0.026278164237737656, 0.006621980573982, 0.00873588863760233, 0.015610352158546448, 0.01920049451291561, 0.007197188679128885, 0.015505407005548477, 0.03940029814839363, 0.051147233694791794, 0.025083128362894058, -0.007395521737635136, 0.03389834612607956, 0.032671455293893814, 0.014265064150094986, 0.060021962970495224, 0.046026282012462616, 0.030699407681822777, 0.030485643073916435, 0.05447040870785713, 0.026031125336885452, 0.04343728348612785, 0.07068231701850891, 0.04749395698308945], [0.05482320487499237, 0.004750749096274376, 0.02347487211227417, -0.03367479145526886, 0.047555696219205856, 0.034103695303201675, -0.0018285138066858053, 0.00785896833986044, -0.04275102913379669, 0.03912537917494774, -0.013915925286710262, -0.015631552785634995, 0.032898932695388794, -0.03382347896695137, 0.011016097851097584, 0.013480178080499172, 0.010128012858331203, -0.0013963952660560608, -0.014917847700417042, -0.0167382899671793, 0.03305912762880325, -0.029574833810329437, 0.029577728360891342, 0.016102518886327744, 0.03061947040259838, -0.01877066306769848, -0.016293184831738472, 0.04954610764980316, 0.021668925881385803, -0.00576043501496315, -0.025689132511615753, 0.02271522395312786, -0.008069044910371304, -0.011513185687363148, 0.007222900632768869, -0.03717788681387901, 0.009091188199818134, 0.018879076465964317, -0.031032999977469444, 0.03588761389255524, 0.004388303030282259, -0.026888960972428322, 0.03208450973033905, -0.028516512364149094, -0.022943874821066856, -0.017376592382788658, -0.01515674777328968, 0.015695899724960327, -0.03272312507033348, -0.041151005774736404, 0.02418883703649044, -0.003951528109610081, -0.031135836616158485, -0.020088160410523415, -0.0359475202858448, 0.00913594663143158, 0.009159698151051998, -0.024126455187797546, -0.02496315725147724, -0.04448188841342926, -0.012882432900369167, -0.033144619315862656, 0.020068710669875145, 0.0023480167146772146, 0.03208056092262268, -0.012839610688388348, 0.024308670312166214, -0.04152379184961319, -0.0134751470759511, -0.02762826345860958, -0.017886556684970856, -0.02218972146511078, -0.03153608366847038, -0.01849399320781231, -0.032434623688459396, -0.023730238899588585, 0.013132167980074883, -0.023928750306367874, -0.04928987845778465, -0.020827241241931915], [0.002387294080108404, 0.017606323584914207, 0.0018683596281334758, 0.032758258283138275, -0.013818278908729553, 0.030784737318754196, 0.029887618497014046, 0.04527166485786438, 0.01924046128988266, -0.0029470184817910194, 0.03148050233721733, 0.0380452461540699, 0.023464517667889595, 0.026432331651449203, -0.0034053444396704435, -0.013831184245646, -0.03575019910931587, -0.0012628295226022601, -0.013179300352931023, 0.01618194207549095, -0.05251038819551468, 0.023104166612029076, 0.03298181667923927, -0.0074723996222019196, -0.04021986946463585, -0.018926652148365974, 0.02297075092792511, -0.044328413903713226, 0.005876847542822361, 0.03996637836098671, 0.01800764538347721, 0.013502588495612144, -0.020853452384471893, -0.037814680486917496, -0.0027809389866888523, -0.0008997954428195953, 0.010164637118577957, -0.013687635771930218, 0.01358850859105587, 0.0619412437081337, -0.031429972499608994, 0.014091458171606064, -0.016783615574240685, -0.03553660959005356, -0.004163308069109917, 0.013237674720585346, 0.0202510803937912, 0.0093249985948205, -0.007299480028450489, -0.00368566089309752, 0.003566236700862646, 0.04275134578347206, -0.03960571810603142, 0.011436842381954193, 0.01750308647751808, -0.014422446489334106, -0.027558820322155952, -0.01249293889850378, -0.032340534031391144, -0.04007979482412338, 0.016072865575551987, 0.004898718558251858, -0.043617215007543564, -0.04094221070408821, -0.015164557844400406, 0.026706183329224586, -0.038894765079021454, -0.028318166732788086, -0.036404337733983994, -0.0463910773396492, -0.05084238573908806, -0.013981969095766544, 0.024554600939154625, -0.027863439172506332, -0.026630127802491188, -0.03095388412475586, -0.0382356122136116, -0.03690604865550995, -0.01482831034809351, -0.04907691106200218], [0.027099305763840675, -0.04727502167224884, 0.023399775847792625, 0.006365541368722916, -0.03153866529464722, 0.02747746929526329, 0.0010147079592570662, -0.016920823603868484, -3.9820435631554574e-05, 0.004062382970005274, -0.021181458607316017, 0.0044080945663154125, 0.01810111291706562, 0.000518643471878022, 0.03965940698981285, -0.012326320633292198, 0.03217417374253273, -0.039687130600214005, 0.039134565740823746, -0.04411712661385536, 0.025809871032834053, 0.027692606672644615, -0.008449236862361431, -0.0004051874566357583, -0.02211005799472332, -0.013782867230474949, -0.0368092842400074, -0.03537043556571007, -0.012439907528460026, -0.0005031742621213198, 0.017404474318027496, -0.012773619033396244, 0.0013680456904694438, 0.023045038804411888, 0.004533306695520878, 0.01290006935596466, -0.006116893608123064, 0.041966743767261505, -0.012756364420056343, -0.00422325124964118, -0.014842184260487556, 0.027207037433981895, -0.023712661117315292, -0.02038775198161602, -0.007464883383363485, 0.016957813873887062, -0.010288815945386887, 0.024649253115057945, -0.017399245873093605, -0.030511396005749702, 0.03307906538248062, -0.021531539037823677, 0.0066640982404351234, 0.03975412994623184, -0.039407309144735336, 0.027905410155653954, -0.023996200412511826, -0.03365834057331085, 0.021434279158711433, -0.032624922692775726, -0.023940902203321457, -0.0352163165807724, 0.027374964207410812, -0.02266068384051323, 0.04726025462150574, -0.02410920336842537, -0.024865208193659782, -0.005661298520863056, 0.010540656745433807, 0.052610378712415695, -0.0020212873350828886, -0.024871980771422386, 0.032550688832998276, 0.04541488364338875, 0.01521726232022047, 0.00953441672027111, -0.024970808997750282, -0.028208283707499504, 0.023106476292014122, -0.024070562794804573], [0.0025011226534843445, 0.03772978484630585, -0.0010987239656969905, 0.006374535616487265, 0.05179310217499733, 0.010380097664892673, 0.006030859891325235, 0.03515387699007988, -0.004737880080938339, 0.002263173460960388, 0.019723575562238693, 0.020130716264247894, 0.032545872032642365, 0.008605602197349072, 0.05310356244444847, 0.002916162135079503, 0.020356111228466034, 0.041565876454114914, 0.03453252464532852, -0.02569326013326645, 0.0013425389770418406, 0.010479310527443886, -0.016049327328801155, 0.03214683383703232, -0.004956105723977089, -0.003802481573075056, 0.03477124124765396, 0.009643021039664745, 0.016706785187125206, 0.030038608238101006, 0.0025044253561645746, 0.003037108574062586, -0.0020585660822689533, 0.036384399980306625, 0.03162368759512901, 0.0125497505068779, 0.004265417344868183, -0.03885253518819809, 0.012486059218645096, -0.0005290124681778252, -0.022287312895059586, 0.03705828636884689, 0.0194727573543787, 0.03440907225012779, 0.00922976341098547, 0.011504245921969414, -0.029510118067264557, 0.015985233709216118, -0.008183464407920837, 0.012509732507169247, -0.006452670320868492, -0.006918944884091616, -0.012696762569248676, 0.01700301095843315, 0.023284925147891045, -0.012436572462320328, -0.032648369669914246, -0.033853206783533096, -0.00639109592884779, 0.014037667773663998, -0.003901038086041808, -0.011001293547451496, -0.03788292780518532, 0.009229145012795925, -0.03507891297340393, -0.023699240759015083, -0.03126513212919235, -0.01777310110628605, -0.02685128152370453, -0.00270496541634202, -0.005834416951984167, -0.019882116466760635, 0.004486748483031988, 0.025088567286729813, -0.01984383910894394, -0.02102203294634819, -0.02707185037434101, -0.012413178570568562, -0.04629313945770264, -0.022761376574635506], [0.004811696242541075, 0.01921481266617775, 0.0177506934851408, 0.026273218914866447, -0.011256316676735878, 0.03666676580905914, 0.03311483934521675, 0.0023613141383975744, -0.045243605971336365, 0.021301042288541794, -0.05052763223648071, 0.004330682568252087, 0.012718915939331055, 0.01623593457043171, -0.0021327922586351633, -0.013437135145068169, -0.029194841161370277, 0.029938610270619392, 0.01067628525197506, 0.014952920377254486, 0.021913394331932068, 0.009984255768358707, 0.01898398995399475, 0.00413094274699688, -0.019399968907237053, 0.01189558394253254, 0.02200254611670971, 0.011808362789452076, -0.02282235585153103, 0.023744909092783928, -0.011026205494999886, -0.020994985476136208, -0.0024247700348496437, -0.026512298732995987, 0.003102427115663886, -0.04002491012215614, -0.01303571555763483, 0.03674043342471123, -0.018596429377794266, -0.018294891342520714, 0.02015092223882675, -0.05800871551036835, -0.03750251606106758, 0.028112098574638367, 0.04269716143608093, 0.002049538539722562, 0.026771849021315575, -0.041076552122831345, 0.017639078199863434, 0.04388701170682907, -0.009439749643206596, 0.03449122980237007, 0.04341858997941017, 0.003240567399188876, -0.03810097277164459, -0.009256791323423386, -0.010582616552710533, 0.015149325132369995, -0.03636671230196953, -0.015876885503530502, -0.027694465592503548, 0.006016007158905268, -0.022071925923228264, -0.02620784007012844, -0.0065389638766646385, -0.013301823288202286, -0.015455571003258228, -0.020382702350616455, 0.0035983251873403788, 0.004557281732559204, -0.03332528471946716, -0.01280075404793024, -0.008837418630719185, -0.02801748551428318, -0.027676869183778763, -0.044568780809640884, -0.04590759426355362, -0.03149840980768204, -0.05648988112807274, -0.0714450553059578], [0.039378631860017776, 0.033195074647665024, 0.016992023214697838, 0.025872381404042244, -0.016170086339116096, 0.006545848213136196, 0.027785366401076317, -0.03231287747621536, -0.0035838226322084665, 0.0315442755818367, 0.011684952303767204, 0.03311890736222267, -0.006655113771557808, -0.009579055942595005, 0.013681294396519661, 0.03150805085897446, -0.015802206471562386, 0.02812267653644085, -0.023141823709011078, 0.015163845382630825, -0.008336542174220085, -0.004585461225360632, 0.0360763780772686, -0.03369646891951561, 0.01692231558263302, 0.012526364997029305, 0.006831384263932705, 0.02521163411438465, -0.04430801048874855, -0.018390975892543793, 0.002200956456363201, 0.016064485535025597, -0.044106025248765945, 0.0024971303064376116, -0.026707498356699944, 0.025855887681245804, 0.040829211473464966, -0.023562166839838028, -0.01223859004676342, -0.010936492122709751, 0.007756573148071766, 0.007048866245895624, -0.01596755161881447, -0.04114373400807381, -0.02329174242913723, -0.007228977046906948, -0.024921059608459473, -0.01827472634613514, -0.019536027684807777, -0.038639407604932785, 0.0005218615988269448, -0.039367642253637314, 0.015108603052794933, -0.007536846213042736, 0.032099880278110504, 0.021194124594330788, 0.014923378825187683, 0.023726407438516617, 0.007912296801805496, 0.039636652916669846, -0.030145511031150818, 0.003565557301044464, 0.005439072847366333, 0.05316469073295593, -0.04731645807623863, 0.020670855417847633, 0.014598988927900791, 0.031999506056308746, 0.01073169894516468, -0.035750217735767365, -0.01651325449347496, 0.01841021329164505, -0.01125295925885439, -0.04578745365142822, -0.030444392934441566, -0.003188003320246935, 0.031206803396344185, -0.041666168719530106, 0.0159793458878994, 0.024826813489198685], [0.03344966471195221, -0.022646557539701462, -0.0051019699312746525, -0.0471191368997097, 0.01964651793241501, 0.025540295988321304, -0.00695036118850112, -0.01195467822253704, -0.04713590070605278, 0.025363633409142494, 0.020716041326522827, -0.00941488891839981, 0.028764594346284866, 0.010876849293708801, -0.0005389402504079044, -0.03256549313664436, -0.029683230444788933, -0.01085666287690401, 0.016684871166944504, -0.027393074706196785, 0.02480979450047016, 0.044460855424404144, -0.00661525409668684, 0.002009049989283085, 0.005076996050775051, 0.019590172916650772, 0.015179544687271118, 0.005453904625028372, 0.03478217497467995, -0.018141059204936028, 0.034979984164237976, -0.04526381194591522, 0.043561987578868866, -0.03162926062941551, -0.018869010731577873, -0.007631742395460606, 0.008174830116331577, 0.03020390123128891, 0.01106699276715517, 0.01775495335459709, -0.016751667484641075, 0.008851890452206135, -0.026447029784321785, 0.036449216306209564, -0.028306733816862106, 0.005599597003310919, 0.0006406237371265888, 0.030779050663113594, 0.009051651693880558, -0.00029552585328929126, -0.018306629732251167, 0.009342746809124947, -0.023518113419413567, -0.02848098799586296, -0.01675274968147278, 0.042042896151542664, 0.022792482748627663, 0.004666600376367569, 0.0010228449245914817, 0.009377584792673588, -0.03578916937112808, -0.01480929832905531, -0.005290340166538954, 0.03159373998641968, -0.034635964781045914, -0.0006760809337720275, -0.03775674104690552, -0.00953019130975008, 0.020306015387177467, -0.02432030625641346, 0.02679600939154625, 0.007065820507705212, 0.044664300978183746, -0.008571703918278217, 0.01990176923573017, 0.004112578462809324, 0.037717003375291824, -0.01210184395313263, -0.03913087770342827, -0.0013241994893178344], [-0.033000584691762924, 0.019880443811416626, -0.03469261899590492, -0.015607738867402077, -0.02122040092945099, -0.01024304423481226, -0.004673215094953775, 0.02685869112610817, -0.03349827229976654, 0.0234085526317358, -0.014239462092518806, 0.05920170247554779, 0.016396401450037956, -0.008390569128096104, 0.022999543696641922, 0.004817050416022539, -0.024836301803588867, 0.02549934573471546, -0.0024881712161004543, -0.024445245042443275, -0.014081251807510853, -0.015718283131718636, 0.04259457811713219, 0.02461508847773075, -0.015536773018538952, -0.01584978587925434, -0.024758679792284966, 0.04090948402881622, -0.012035300023853779, -0.04584532231092453, -0.054940808564424515, 0.03786887601017952, -0.01936924085021019, 0.026404978707432747, 0.03639303892850876, -0.05410125106573105, -0.017627887427806854, 0.016600191593170166, -0.017271099612116814, -0.007486558053642511, 0.04547428339719772, -0.01992993801832199, -0.030333444476127625, 0.004074727650731802, 0.017423594370484352, 0.0333775132894516, 0.022672167047858238, -0.010812482796609402, 0.00432508485391736, -0.017494821920990944, 0.04669656977057457, 0.015669597312808037, 0.010696710087358952, -0.00438659405335784, -0.04078524187207222, -0.017193960025906563, -0.004479837138205767, -0.00010618512169457972, -0.013067997992038727, 0.010509101673960686, 0.04536900669336319, 0.012200933881103992, -0.019639743492007256, -0.006756077986210585, 0.023500345647335052, 0.00706885801628232, -0.02933391183614731, -0.009395609609782696, 0.02281978540122509, 0.016197122633457184, -0.01314709521830082, 0.010147515684366226, 0.01895475760102272, 0.011916089802980423, 0.02043761871755123, -0.04562489315867424, 0.03604757785797119, 0.031161002814769745, 0.04494215175509453, 0.03956848755478859], [0.04568146541714668, 0.03218835964798927, 0.013182317838072777, 0.04555736854672432, 0.011184624396264553, -0.016000060364603996, -0.00939561054110527, 0.023720894008874893, -0.010766097344458103, -0.031011944636702538, 0.05229910835623741, 0.012243891134858131, -0.0242216344922781, 0.0011907638981938362, 0.040910400450229645, 0.014848173595964909, -0.006800197530537844, 0.012677996419370174, -0.012679614126682281, 0.010544167831540108, -0.04698992893099785, -0.03307758644223213, -0.0007466587703675032, -0.013138985261321068, 0.01669168844819069, -0.0012832748470827937, 0.0015525422058999538, -0.0021098293364048004, 0.006256511434912682, 0.017084341496229172, -0.008504819124937057, -0.01999039016664028, 0.025700023397803307, -0.04967525601387024, 0.01707646995782852, 0.0034562863875180483, 0.011345980688929558, 0.010843086987733841, 0.016462799161672592, 0.006523163989186287, -0.05937277153134346, -0.04074879363179207, 0.004985992331057787, 0.05303017050027847, 0.01660769060254097, 0.048452019691467285, -0.04233647882938385, 0.029525356367230415, -0.022154733538627625, -0.029839536175131798, 0.0028560510836541653, -0.01248947437852621, -0.0067426785826683044, 0.054176583886146545, 0.042351070791482925, -0.017489507794380188, -0.020350083708763123, -0.03967610001564026, -0.03268921747803688, -0.01867290586233139, -0.01037315558642149, 0.01193111389875412, -0.03914377838373184, 0.005474317818880081, 0.00012504911865107715, -0.037492893636226654, -0.02544126659631729, 0.0006833241204731166, -0.005230623763054609, 0.012080714106559753, -0.03940508887171745, 0.02687404304742813, -0.011500418186187744, -0.041807033121585846, 0.02752562239766121, 0.02806815691292286, -0.04137225076556206, 0.012821956537663937, 0.011368848383426666, 0.014775266870856285], [0.007057772949337959, 0.027171336114406586, 0.010029343888163567, -0.002480119466781616, 0.03504098579287529, -0.022191232070326805, -0.04128522053360939, -0.03201638534665108, -0.008709825575351715, -0.011952333152294159, 0.03246263042092323, 0.004268269520252943, 0.02589568682014942, -0.02200464904308319, 0.019398674368858337, -0.022161053493618965, -0.014768870547413826, -0.010220161639153957, -0.011615394614636898, 0.013827760703861713, 0.008063854649662971, 0.010707189328968525, -0.015086651779711246, -0.005562812089920044, 0.004389332607388496, 0.007418474648147821, -0.03230433166027069, -0.009772855788469315, 0.006169180851429701, -0.005515906028449535, -0.030475035309791565, 0.013424710370600224, 0.04336010664701462, 0.029754048213362694, -0.05007928982377052, 0.026199763640761375, -0.0037804043386131525, 0.0354221947491169, -0.010350611060857773, 0.037324946373701096, -0.0064066508784890175, 0.023979928344488144, 0.025337576866149902, 0.010743838734924793, -0.036707840859889984, -0.018084751442074776, -0.0293436236679554, -0.04063744470477104, -0.030720485374331474, 0.02951483614742756, 0.01981591247022152, -0.034149471670389175, 0.01802981086075306, -0.017535744234919548, -0.030079413205385208, -0.041644930839538574, 0.02195558324456215, -0.041101668030023575, 0.027228204533457756, 0.01483901310712099, 0.0035172614734619856, -0.058989908546209335, -0.02941795065999031, -0.017624711617827415, -0.013305364176630974, -0.029024649411439896, 0.006352197378873825, 0.02389180287718773, -0.011845158413052559, -0.034695036709308624, -0.025638272985816002, -0.06949789822101593, 0.028780575841665268, -0.039753563702106476, -0.0038277374114841223, -0.05919530615210533, -0.009697332046926022, -0.03874126821756363, -0.0013221389381214976, -0.076441191136837], [-0.012279737740755081, -0.03700114041566849, 0.003579768119379878, 0.044815268367528915, 0.01159351971000433, -0.04675168916583061, 0.009963982738554478, -0.0018300417577847838, -0.010038331151008606, 0.003586104139685631, -0.004966351669281721, 0.02382960356771946, -0.010888779535889626, -0.004015771206468344, 0.02331607975065708, 0.01896188221871853, -0.043554652482271194, 0.003984745591878891, -0.007091572508215904, -0.01664997823536396, -0.03141988068819046, -0.010113004595041275, -0.015739252790808678, 0.008224071003496647, 0.0304382536560297, -0.0569864884018898, -0.06437747925519943, 0.002681296318769455, -0.038243066519498825, 0.04141971841454506, 0.03383787348866463, -0.0141359968110919, 0.020450130105018616, -0.03549864888191223, 0.0036135055124759674, 0.04564634710550308, 0.045855894684791565, 0.001592070097103715, -0.005183303263038397, 0.03000570647418499, 0.06584082543849945, 0.037267714738845825, 0.048510655760765076, -0.002812903141602874, -0.012387527152895927, -0.05334658920764923, 0.008150769397616386, 0.042437002062797546, -0.0068869865499436855, -0.0007132330792956054, -0.011311153881251812, 0.016499515622854233, -0.010604669339954853, 0.008066126145422459, -0.041772082448005676, 0.010545410215854645, -0.015436223708093166, -0.01621302030980587, -0.013240690343081951, -0.055837929248809814, 0.033456962555646896, 0.026830438524484634, 0.024225490167737007, -0.00038028403650969267, -0.017829429358243942, -0.0026836178731173277, 0.008013982325792313, -0.012614239938557148, 0.015705430880188942, -0.033626846969127655, -0.01760956272482872, 0.04979715496301651, -0.002755152527242899, 0.020847633481025696, 0.021627794951200485, -0.018235202878713608, -0.03347427770495415, 0.005487985443323851, -0.008878495544195175, 0.0035959931556135416], [0.010188599117100239, 5.344587407307699e-05, -0.03905678540468216, 0.012578106485307217, -0.0003604100493248552, 0.017089737579226494, 0.007884006947278976, -0.021139176562428474, -0.0098055275157094, -0.02571180835366249, -0.01884797029197216, -0.010303832590579987, 0.014973982237279415, -0.019306056201457977, 0.04808010533452034, -0.02211320400238037, -0.04842998832464218, -0.04083970934152603, 0.004459049552679062, 0.025593088939785957, 0.01554154884070158, 0.023745503276586533, -0.044065628200769424, -0.017841048538684845, 0.02996731363236904, -0.027030272409319878, -0.05089498683810234, -0.05551852285861969, 0.027554186061024666, 0.033009327948093414, -0.04230191186070442, 0.017663590610027313, 0.007966313511133194, 0.004631504882127047, 0.014324414543807507, 0.030606383457779884, -0.010516480542719364, 0.010056200437247753, 0.042217738926410675, -0.011774099431931973, 0.018001899123191833, -0.005114024970680475, 0.010405407287180424, 0.006242615636438131, -0.01196720078587532, -0.05833010748028755, -0.051085010170936584, -0.035302095115184784, 0.026090843603014946, -0.02059522643685341, 0.027346301823854446, -0.03866301476955414, 0.03383004665374756, 0.026066657155752182, -0.03723565489053726, -0.019497627392411232, -0.06453479081392288, 0.016374429687857628, 0.01611802726984024, -0.06262417882680893, 0.015671653673052788, 0.019671736285090446, -0.007096069399267435, 0.029949938878417015, -0.034850653260946274, 0.0015265154652297497, -0.008003692142665386, 0.040416065603494644, 0.012935658916831017, 0.008412294089794159, -0.019981535151600838, -0.04128990322351456, 0.009852457791566849, 0.009357665665447712, 0.030178865417838097, -0.006883010268211365, 0.03425116464495659, -0.02174011990427971, 0.00827708188444376, -0.03458455950021744], [0.05100039765238762, 0.04311395436525345, 0.020581066608428955, 0.018863171339035034, 0.04075154662132263, 0.01716168411076069, 0.04227723181247711, 0.033916547894477844, 0.027228500694036484, 0.02874690294265747, 0.05820620432496071, 0.02057201787829399, -0.02006077766418457, 0.006904394831508398, 0.0359865203499794, 0.03002023510634899, 0.027531849220395088, -0.014005390927195549, 0.05881977826356888, 0.021139688789844513, 0.01792590133845806, 0.020769212394952774, 0.026361726224422455, 0.036150652915239334, 0.030301544815301895, -0.002323759254068136, 0.032074593007564545, 0.0025727422907948494, -0.005575620103627443, -0.010512717068195343, 0.03403586894273758, 0.0003847808693535626, 0.017011936753988266, 0.023719845339655876, 0.0009898231364786625, 0.020396577194333076, -0.020541490986943245, -0.005286825355142355, 0.022351110354065895, 0.01326825376600027, 0.005723666865378618, 0.0019989556167274714, -0.042430900037288666, 0.011116480454802513, -0.025673622265458107, -0.021376825869083405, -0.008119533769786358, -0.013152024708688259, 0.0244684386998415, -0.005110922735184431, 0.02346503548324108, -0.00025562132941558957, -0.03861933574080467, 0.018111124634742737, -0.019622480496764183, -0.007210624869912863, -0.031800609081983566, 0.04245381057262421, -0.0005091429920867085, 0.017351511865854263, -0.03560870885848999, 0.01307671982795, 0.0015627272659912705, -0.022738581523299217, -0.01299265306442976, -0.021293893456459045, -0.0007290267385542393, -0.049877382814884186, 0.012401667423546314, -0.03192080557346344, -0.027561647817492485, 0.0078995106741786, -0.0332430936396122, -0.012605422176420689, 0.006995460018515587, -0.024301040917634964, -0.040470920503139496, -0.02545551210641861, -0.030149297788739204, -0.057411011308431625], [-0.016376253217458725, 0.014918378554284573, -0.03181956335902214, 0.02314526028931141, -0.02926541119813919, -0.043969351798295975, 0.01227852888405323, -0.007821192033588886, 0.01577857881784439, 0.01732807792723179, 0.009636645205318928, -0.028828216716647148, 0.036168184131383896, 0.005400415044277906, -0.02166178822517395, 0.03875428065657616, 0.035013895481824875, -0.01961999014019966, 0.029172508046030998, 0.0018265210092067719, 0.03810766711831093, -0.0011198298307135701, -0.007798737846314907, 0.011086557991802692, -0.003404365386813879, -0.04944977909326553, -0.0010298284469172359, 0.015453596599400043, -0.01650438830256462, 0.01697009801864624, -0.01145090814679861, -0.016766885295510292, -0.05022808909416199, 0.032589592039585114, 0.0051085203886032104, 0.0051764799281954765, -0.005205453373491764, -0.027406688779592514, -0.006252264603972435, -0.024476295337080956, 0.0001418737811036408, -0.006541733630001545, -0.016439447179436684, 0.010387030430138111, -0.01457472238689661, 0.029099436476826668, 0.01950111798942089, -0.011654071509838104, -0.010577775537967682, 0.022014614194631577, 0.010765810497105122, -0.026966776698827744, 0.011267813853919506, -0.03595717251300812, -0.01220628246665001, 0.030872123315930367, -0.04478039592504501, -0.01143447495996952, 0.017860963940620422, 0.023713331669569016, 0.045405834913253784, -0.0006730981403961778, 0.042189255356788635, -0.0008371592848561704, 0.01833309233188629, 0.013129981234669685, -0.01997246965765953, -0.04016987979412079, 0.02279444970190525, 0.0003797561803366989, 0.019452625885605812, -0.047525666654109955, 0.048247482627630234, 0.007736089173704386, 0.027996812015771866, 0.013780564069747925, -0.004765941761434078, -0.035279806703329086, -0.019256403669714928, 0.03463440015912056], [-0.024867409840226173, 0.025829428806900978, 0.02024780586361885, -0.019386086612939835, -0.04511575400829315, -0.02225605584681034, 0.04513605684041977, 0.03184987232089043, 0.024696826934814453, 0.0232918169349432, -0.06402435153722763, 0.018836693838238716, 0.041814617812633514, -0.021629922091960907, -0.014474031515419483, -0.022860992699861526, 0.02680905908346176, -0.01463227067142725, -0.04839247092604637, -0.04961107671260834, 0.004994198679924011, 0.0010728873312473297, -0.007472333498299122, 0.020307861268520355, -0.009618070907890797, 0.007212487049400806, -0.002817300148308277, 0.01563916727900505, 0.02288096398115158, -0.05245443433523178, -0.019747015088796616, 0.0061571188271045685, -0.03873163089156151, 0.04515351727604866, -0.010926978662610054, 0.03238814324140549, 0.006611377466470003, -0.0008187585044652224, -0.019222650676965714, -0.05209814012050629, -0.006876643747091293, -0.024980615824460983, 0.006967483554035425, -0.017425471916794777, -0.018628384917974472, 0.02520453743636608, -0.06078796833753586, -0.04015269875526428, -0.006676200311630964, 0.021093858405947685, -0.016425788402557373, 0.0083609102293849, -0.01729467697441578, 0.007063830271363258, 0.02711879462003708, -0.023205740377306938, 0.027523184195160866, -0.00908849947154522, 0.027845870703458786, 0.030143504962325096, -0.009113403037190437, 0.038933344185352325, 0.002999026793986559, -0.02019025944173336, -0.00312636187300086, -0.042903125286102295, -0.03680922091007233, -0.005301241297274828, -0.0009639656054787338, 0.015276285819709301, 0.03485735133290291, -0.021358061581850052, -0.013200170360505581, -0.05238475278019905, -0.00427471287548542, -0.022048458456993103, -0.013569478876888752, 0.0005809927242808044, 0.03712430223822594, -0.023557033389806747], [0.017823893576860428, 0.02527431771159172, 0.015841979533433914, 0.0026668424252420664, -0.02140965685248375, -0.03418499603867531, -0.02045951969921589, -0.021549636498093605, -0.004949764348566532, 0.010094324126839638, 0.010057933628559113, 0.0034097584430128336, -0.0172469150274992, 0.030252506956458092, -0.01589096710085869, -0.0175615344196558, 0.01678607054054737, 0.018625594675540924, 0.031160449609160423, 0.04097608104348183, -0.01864270679652691, -0.020993828773498535, -0.006260541267693043, -0.020295443013310432, 0.03737766295671463, -0.022138478234410286, -0.0389975905418396, 0.02827322855591774, 0.014717373996973038, -0.014403823763132095, 0.00906879361718893, 0.02818247862160206, -0.006216841284185648, 0.008535197004675865, 0.012794600799679756, 0.01890740357339382, -0.016739390790462494, -0.03888906538486481, -0.01402259711176157, 0.012180664576590061, 0.0024177939631044865, 0.017343668267130852, -0.025722626596689224, 0.02041611075401306, 0.02087627351284027, -0.035534799098968506, 0.03876861557364464, 0.02254084311425686, 0.019807325676083565, -0.011213360354304314, -0.024046223610639572, 0.017457932233810425, -0.009843366220593452, -0.03760914504528046, 0.018929826095700264, -0.023248130455613136, 0.033273205161094666, 0.03280654549598694, -0.01959765888750553, -0.021147551015019417, 0.004263641312718391, -0.06339448690414429, 0.01936080865561962, 0.02050480619072914, 0.022488810122013092, -0.021616414189338684, -0.036865491420030594, 0.005192728713154793, -0.02156379260122776, -0.000894820608664304, -0.010788694955408573, 0.0012659472413361073, -0.01129241194576025, -0.0087211262434721, -0.030939940363168716, -0.0195162296295166, -0.00887875072658062, 0.0182338859885931, -0.06263546645641327, -0.07421140372753143], [0.019975436851382256, 0.05014113709330559, -0.0207224078476429, 0.0014060343382880092, -0.026208503171801567, 0.02065690979361534, 0.008779391646385193, 0.038724903017282486, 0.04416827857494354, -0.0027126779314130545, 0.032548870891332626, 0.030441809445619583, 0.04783693328499794, -0.013083287514746189, 0.010581767186522484, 0.04204661771655083, -0.021506723016500473, -0.002690656576305628, 0.0022371115628629923, -0.02070464752614498, 0.026633350178599358, 0.006071494892239571, -0.023325040936470032, -0.014412321150302887, 0.021336939185857773, 0.025217335671186447, -0.008548570796847343, 0.006597965024411678, -0.016611197963356972, -0.03224930539727211, -0.008614965714514256, 0.029289839789271355, 0.00594312185421586, 0.020025718957185745, 0.013988586142659187, 0.025117507204413414, 0.014400956220924854, -0.014301913790404797, 0.014834350906312466, 0.0013904194347560406, -0.012987569905817509, 0.01664113439619541, 0.05595235154032707, -0.028746141120791435, 0.011458680965006351, -0.04925486817955971, -0.02865404263138771, 0.021676531061530113, -0.03990801051259041, -0.018963966518640518, -0.009275341406464577, 0.012629054486751556, -0.0075411926954984665, -0.011903715319931507, -0.03774462267756462, -0.04290062189102173, -0.0516153946518898, 0.01317768543958664, -0.006210953462868929, -0.05421651899814606, -0.024261029437184334, -0.011260217987000942, -0.03131568059325218, 0.033061906695365906, 0.003287913976237178, -0.027670027688145638, -0.012975265271961689, -0.038122814148664474, 0.034575629979372025, 0.01692233979701996, -0.004858074244111776, 0.007613299414515495, -0.024498872458934784, -0.014096359722316265, -0.014458823949098587, -0.008248613215982914, 0.0025320120621472597, -0.03276152163743973, 0.006647006142884493, 0.030207879841327667], [0.040607187896966934, -0.004004291724413633, 0.027764959260821342, -0.010942467488348484, -0.018749872222542763, 0.0008192723616957664, -0.026265626773238182, -0.021534087136387825, 0.028782468289136887, -0.007580674719065428, -0.017948191612958908, -0.016622405499219894, -0.004605850670486689, -0.020779259502887726, 0.042213909327983856, -0.02052224427461624, 0.02059180662035942, 0.02073550969362259, 0.00018730897863861173, 0.011564225889742374, 0.00024102175666484982, -0.03815208375453949, 0.008307542651891708, 0.0011976405512541533, -0.046951521188020706, -0.03974815830588341, 0.03600218892097473, 0.018994048237800598, 0.017091380432248116, -0.02156139723956585, 0.023283421993255615, -0.015529775992035866, -0.06281067430973053, -0.018450690433382988, 0.02984272502362728, 0.011145994067192078, -0.036622054874897, -0.024503832682967186, 0.03164549544453621, 0.016976280137896538, -0.016659822314977646, -0.004022936802357435, 0.01806514337658882, -0.01836630515754223, 0.00026395500754006207, 0.01837754249572754, 0.00790803786367178, -0.030870653688907623, 0.04616880044341087, -0.012741513550281525, 0.03098992072045803, -0.04486247897148132, 0.0012380063999444246, -0.032776057720184326, 0.011669396422803402, 0.016768943518400192, 0.01638820581138134, 0.045993924140930176, 0.0013235056539997458, 0.02321389876306057, -0.02167987823486328, -0.03543759509921074, -0.00879888329654932, 0.008098259568214417, -0.020786048844456673, 0.03757373243570328, -0.018701817840337753, 0.023347627371549606, -0.0411975271999836, 0.02375112846493721, 0.025631412863731384, -0.03285316377878189, -0.03369353711605072, -0.036230653524398804, -0.02159304730594158, 0.01728297583758831, -0.008386998437345028, -0.002244365867227316, 0.0200254637748003, 0.004627497401088476], [-0.027733037248253822, -0.024002889171242714, -0.056383196264505386, -0.01612393744289875, 0.02200372703373432, 0.0015347679145634174, -0.005049678031355143, 0.005708459764719009, -0.013352650217711926, 8.465501014143229e-05, 0.013721565715968609, 0.045667845755815506, 0.016354814171791077, -0.0027071721851825714, 0.01398528553545475, -0.03585083410143852, -0.0197207760065794, -0.01625213213264942, 0.034816592931747437, -0.005171562545001507, -0.0054767075926065445, -0.005823215935379267, -0.029736613854765892, 0.00042160513112321496, 0.04166198894381523, -0.020847279578447342, 0.007452191784977913, -0.04982766881585121, -0.027245743200182915, 0.015557686798274517, 0.024141950532794, 0.0013661959674209356, 0.007072723936289549, -0.018870046362280846, -0.013742126524448395, -0.006950761657208204, -0.00785005558282137, -0.006746916100382805, 0.023722048848867416, -0.03900139406323433, 0.05715268850326538, 0.01769462414085865, -0.011012797243893147, 0.020851727575063705, -0.011929106898605824, 0.013605969026684761, -0.004961726721376181, -0.015508780255913734, -0.027340339496731758, 0.02330050617456436, -0.015278486534953117, 0.05325079709291458, 0.015340382233262062, -0.015382206998765469, -0.03261095657944679, -0.012165537104010582, -0.007979102432727814, -0.036407385021448135, 0.0053952415473759174, 0.005603662226349115, 0.0001558634830871597, 0.008577175438404083, -0.009220696985721588, -0.02443940006196499, -0.03275061398744583, 0.002849970478564501, -0.04770514369010925, -0.002754746936261654, 0.04331311956048012, -0.023882877081632614, 0.025972586125135422, -0.02999156340956688, -0.04234123229980469, -0.018506916239857674, -0.00868876464664936, 0.01638440228998661, 0.016512122005224228, 0.016312595456838608, -0.018598906695842743, -0.057101380079984665], [-0.011453486979007721, -0.06528507173061371, 0.018847430124878883, -0.008716662414371967, 0.006707124412059784, 0.018259547650814056, -0.029081998392939568, 0.025077026337385178, 0.010818473994731903, 0.014230790548026562, 0.006336873397231102, 0.049259014427661896, 0.018183736130595207, 0.03592366352677345, 0.013872017152607441, 0.03637036308646202, 0.015117325820028782, -0.013234824873507023, 0.0023183312732726336, 0.04657890275120735, 0.01663503237068653, 0.03082611970603466, 0.034055039286613464, -0.003372220555320382, -0.023587189614772797, 0.015603245235979557, 0.010544252581894398, 0.011410228908061981, -0.037990085780620575, -0.03538324311375618, -0.00717422179877758, 0.02979378029704094, -0.031180337071418762, 0.006934483535587788, -0.012219826690852642, 0.008762695826590061, -0.009051118977367878, 0.019761811941862106, -0.013444699347019196, -0.052693553268909454, 0.028143683448433876, -0.021984538063406944, 0.03241316229104996, -0.0024816838558763266, -0.04096851870417595, -0.01690838299691677, 0.01544168684631586, -0.02997272089123726, -0.02094598114490509, 0.03578425571322441, 0.036228474229574203, 0.03666409105062485, 0.0009883390739560127, -0.015301146544516087, -0.003481048159301281, 0.0056392960250377655, -0.011067559942603111, -0.01799740269780159, -0.02310417778789997, 0.01253846101462841, 0.04006265848875046, -0.016756350174546242, -0.03415055572986603, 0.019145378842949867, 0.04004889726638794, -0.008777491748332977, 0.01055123284459114, -0.01802867464721203, -0.020692182704806328, -0.007042312528938055, 0.0217039342969656, 0.016286388039588928, -0.023232875391840935, 0.021397771313786507, 0.03608845919370651, -0.005423869006335735, 0.024768101051449776, -0.0061196559108793736, 6.404049781849608e-05, 0.010895990766584873], [-0.016171123832464218, -0.048487868160009384, -0.02750767022371292, 0.043565794825553894, 0.0048269000835716724, -0.05137094855308533, -0.021195460110902786, 0.01499813050031662, -0.04629078879952431, 0.017198050394654274, -0.017573010176420212, -0.018806112930178642, -0.011587406508624554, -0.011169311590492725, 0.02320854365825653, 0.01761781983077526, 0.024992696940898895, -0.018763070926070213, -0.0341467447578907, -0.037646085023880005, 0.01014502439647913, 0.022880330681800842, -0.034271400421857834, -0.014774634502828121, 0.017786243930459023, -0.057714998722076416, -0.04313123598694801, 0.0104365861043334, 0.03414563462138176, 0.008944532833993435, -0.01646796055138111, -0.021700967103242874, 0.043091047555208206, -0.012942061759531498, -0.013927542604506016, -0.011392142623662949, 0.0333172082901001, 0.04523078352212906, 0.025133877992630005, -0.02715792879462242, -0.010865876451134682, -0.017290452495217323, 0.02768171951174736, 0.017280761152505875, -0.01693982630968094, -0.043232694268226624, -0.04311876371502876, 0.03862666338682175, 0.030147843062877655, -0.027372954413294792, -0.00018985880888067186, 0.018560392782092094, -0.041247107088565826, -0.004257052671164274, 0.023609153926372528, -0.017202626913785934, -0.0478605255484581, 0.021896829828619957, -0.02833244949579239, 0.02925196848809719, -0.042441993951797485, 0.01925286091864109, 0.028968915343284607, 0.023302191868424416, 0.047649312764406204, 0.029546819627285004, 0.001073875231668353, 0.01970692351460457, 0.014533421024680138, 0.00805207435041666, -0.03650715947151184, 0.021236766129732132, -0.03195622190833092, -0.03621061146259308, -0.019811782985925674, 0.03337712958455086, 0.026863915845751762, -0.015491891652345657, -0.04136670380830765, -0.04366380721330643], [-0.0420515239238739, -0.05923710763454437, -0.013074052520096302, 0.0032093876507133245, -0.021312864497303963, -0.013509167358279228, 0.0019074873998761177, -0.03983910009264946, -0.07848550379276276, -0.045976318418979645, 0.021045923233032227, -0.027914581820368767, 0.04533501714468002, 0.012510335072875023, -0.026970291510224342, 0.0028479767497628927, -0.0209379643201828, -0.014035006985068321, -0.03309935703873634, 0.025432338938117027, 0.03863804042339325, 0.0019492488354444504, 0.0049585625529289246, 0.019069645553827286, -0.012019325979053974, -0.009467637166380882, 0.023722801357507706, -0.05114573612809181, 0.03499029949307442, 0.02999146096408367, -0.0026363444048911333, 0.00697320606559515, 0.028970913961529732, 0.010314698331058025, -0.0612027645111084, 0.04142541065812111, 0.019796567037701607, -0.0016898015746846795, 0.004820054862648249, 0.017343970015645027, 0.04976120591163635, -4.704411549028009e-05, -0.014644619077444077, 0.042683571577072144, -0.004283578135073185, -0.023897988721728325, -0.007696727756410837, 0.02730581909418106, 0.026360202580690384, 0.011641860008239746, 0.005724360700696707, -0.008886795490980148, 0.0015162300551310182, 0.008681397885084152, 0.025671906769275665, -0.00821202527731657, -0.00014527067833114415, 0.021876320242881775, -0.01058410108089447, -0.0026966542936861515, 0.015665527433156967, -0.013777018524706364, -0.017733311280608177, 0.0194451455026865, 0.0023856288753449917, 0.013309828005731106, -0.023226333782076836, -0.012520557269454002, 0.045349933207035065, -0.02634064480662346, -0.023196320980787277, -0.046685826033353806, -0.005366622935980558, 0.035058021545410156, 0.040496692061424255, -0.017899099737405777, -0.0064037335105240345, 0.053165048360824585, 0.03687513619661331, 0.014970925636589527], [-0.036290790885686874, -0.04012379050254822, -0.0111270472407341, -0.04606785625219345, -0.013518981635570526, 0.0020050194580107927, -0.019328374415636063, -0.014179726131260395, -0.04419546574354172, -0.00899600051343441, -0.013102278113365173, -0.03841623291373253, -0.0286426842212677, -0.03676828369498253, 0.0067377956584095955, -0.025924736633896828, -0.03321165591478348, 0.021657051518559456, -0.021613623946905136, 0.014757579192519188, -0.02168164774775505, 0.03629022836685181, -0.03835425153374672, -0.006260300520807505, 0.004391225520521402, 0.021907243877649307, 0.0044603655114769936, -0.007162881549447775, -0.031281452625989914, 0.048696208745241165, 0.025098510086536407, -0.010026432573795319, 0.00869716890156269, 0.014876226894557476, 0.017117977142333984, 0.03160722181200981, -0.029203645884990692, 0.002914899494498968, -0.012918447144329548, 0.03217693790793419, -0.032674677670001984, 0.04042518883943558, -0.01982985995709896, -0.005538012366741896, 0.016541672870516777, -0.018094975501298904, 0.001737052807584405, 0.015215359628200531, 0.010753653943538666, -0.00928227137774229, -0.026129258796572685, 0.025199411436915398, -0.028548043221235275, 0.035280708223581314, -0.02336364984512329, -0.02780831605195999, 0.004559437744319439, 0.041650716215372086, 0.010733506642282009, 0.012474958784878254, -0.038855332881212234, -0.027991535142064095, 0.0038561939727514982, 0.03797175735235214, -0.006003826856613159, -0.00835469365119934, 0.009415717795491219, 0.033605750650167465, 0.016832107678055763, -0.005729274358600378, -0.03171266242861748, 0.024135813117027283, 0.04875125735998154, -0.002951457630842924, 0.017257975414395332, 0.001278805430047214, 0.049352504312992096, 0.04027069732546806, -1.3535618563764729e-06, -0.012098891660571098], [7.883959187893197e-05, -0.007955729030072689, 0.03611557558178902, -0.0010201374534517527, -0.06247490271925926, 0.043386299163103104, 0.04287507012486458, -0.002704766346141696, -0.002697542542591691, 0.017366250976920128, 0.008639157749712467, -0.04060947522521019, 0.020033851265907288, 0.00580646563321352, -0.03642181307077408, 0.0029958493541926146, 0.01838936097919941, -0.010800128802657127, -0.04497149959206581, -0.015413975343108177, 0.025938749313354492, 0.014787903055548668, 0.01002697367221117, 0.04608909785747528, 0.022004032507538795, 0.016933254897594452, -0.0037561848293989897, -0.03141486644744873, -0.007233592215925455, 0.011421328410506248, 0.00982891209423542, 0.013386509381234646, -0.0014675222337245941, -0.01774008385837078, -0.025914065539836884, -0.019633252173662186, 0.03307247906923294, -0.028353461995720863, -0.02949431724846363, -0.03872569277882576, 0.011180253699421883, -0.03096945397555828, -0.006249772384762764, -0.0016858992166817188, 0.026252198964357376, 0.03439486771821976, -0.03616734594106674, 0.039259523153305054, 0.006555718835443258, 0.023661699146032333, -0.008091200143098831, 0.0010447257664054632, -0.01792147196829319, -0.03830504044890404, -0.02947288565337658, -0.02121175266802311, 0.011660340242087841, 0.03648902475833893, 0.011592142283916473, -0.0172190573066473, 0.04430913180112839, 0.04232577234506607, 0.02337939850986004, -0.006776624824851751, 0.0037028510123491287, 0.01739330403506756, -0.019500799477100372, 0.002532050246372819, -0.034668270498514175, -0.01881936378777027, -0.023546578362584114, -0.019480673596262932, -0.010775703005492687, -0.0014230605447664857, -0.02991478331387043, -0.014853745698928833, -0.0021840210538357496, -0.011126913130283356, -0.02775375172495842, 0.012761401012539864], [0.03256092965602875, -0.0031930110417306423, 0.030036354437470436, 0.018588891252875328, 0.028929751366376877, 0.0027825061697512865, 0.026147784665226936, 0.023763880133628845, -0.004661043640226126, -0.006944763939827681, -0.0026606025639921427, 0.013818487524986267, -0.03169454634189606, 0.017713837325572968, -0.024713201448321342, 0.012278451584279537, 0.05459573492407799, -0.02624337747693062, 0.02327267825603485, 0.020437171682715416, 0.022143874317407608, 0.03140665218234062, 0.010150518268346786, 0.05621233209967613, -0.017864443361759186, -0.005009342450648546, 0.038361188024282455, 0.032133713364601135, -0.032131023705005646, -0.001968255965039134, -0.03999881073832512, -0.01756199821829796, 0.002386166714131832, -0.009081031195819378, -0.029729487374424934, 0.028671974316239357, -0.015059875324368477, 0.04891538992524147, -0.01587671786546707, -0.037443604320287704, -0.027764106169342995, 0.011570297181606293, -0.032116495072841644, -0.031102819368243217, 0.006070851814001799, 0.036944132298231125, 0.024788541719317436, 0.029241692274808884, 0.015889469534158707, 0.01295932661741972, -0.013674219138920307, -0.012941901572048664, 0.008223016746342182, -0.0049587092362344265, -0.024278514087200165, 0.009127509780228138, 0.03349979221820831, -0.01065688394010067, 0.00749559560790658, 0.006387852132320404, 0.024465039372444153, 0.03452841937541962, -0.031675565987825394, -7.70509650465101e-05, 0.006861744448542595, 0.04182368516921997, 0.01969854347407818, -0.00974168349057436, 0.03592842072248459, 0.027389641851186752, 0.029014453291893005, -0.021240796893835068, -0.02197938971221447, 0.030156325548887253, 0.013967524282634258, -0.0011947302846238017, -0.03693675622344017, -0.0392993688583374, 0.0014993924414739013, 0.0035131045151501894], [-0.024597356095910072, -0.03483647480607033, 0.011291280388832092, 0.017137926071882248, -0.0012932912213727832, -0.0032242201268672943, 0.0049193259328603745, 0.009770152159035206, 0.013371530920267105, -0.007368910126388073, 0.02685084380209446, 0.045021265745162964, -0.028225695714354515, 0.0065462179481983185, 0.021344881504774094, 0.010515406727790833, 0.00651269918307662, -0.015147640369832516, -0.019495319575071335, 0.017293807119131088, 0.030650988221168518, -0.039346419274806976, -0.021664023399353027, -0.0045525808818638325, -0.03984079882502556, 0.0001959626388270408, 0.0013399521121755242, 0.004955027252435684, -0.032836101949214935, -0.014604594558477402, 0.003964463248848915, 0.03824467211961746, 0.006236635148525238, -0.021133197471499443, -0.00973488762974739, 0.04674519598484039, 0.009419449605047703, 0.010968354530632496, -0.04244507476687431, 0.02457592636346817, 0.02449231781065464, -0.030019370838999748, 0.01506743486970663, -0.002316680271178484, -0.03286181762814522, 0.04341918230056763, 0.008770575746893883, -0.00249960133805871, -0.023599039763212204, -0.011841206811368465, -0.028983639553189278, 0.03249518573284149, 0.017651261761784554, -0.004689761437475681, -0.033971235156059265, -0.006966674700379372, -0.006565593183040619, -0.04218783602118492, -0.025812847539782524, -0.0007784918998368084, -0.023914098739624023, -0.0038292095996439457, -0.04070267826318741, -0.016291428357362747, 0.010472223162651062, 0.016314642503857613, 0.026676354929804802, -0.04753563180565834, -0.04188584163784981, -0.01658652350306511, -0.006515377666801214, -0.04810449853539467, -0.010919628664851189, -0.007825931534171104, 0.0059630051255226135, -0.031740717589855194, -0.027714502066373825, 0.028975581750273705, -0.004933635704219341, 0.023092415183782578], [0.01337728463113308, 0.002629949478432536, -0.015612374059855938, 0.011312675662338734, -0.006523948162794113, -0.04470944404602051, 0.012489127926528454, 0.04292282089591026, -0.03546230122447014, 0.02866130881011486, 0.012354866601526737, -0.020197536796331406, 0.023175057023763657, 0.026500023901462555, -0.012791622430086136, 0.021316492930054665, -0.028276413679122925, 0.011955768801271915, 0.011434475891292095, 0.03280289098620415, 0.03953249752521515, 0.018644148483872414, 0.009624609723687172, -0.00931023433804512, -0.015657885000109673, 0.032225534319877625, -0.027272745966911316, 0.03595569729804993, 0.048502117395401, -0.010553639382123947, 0.03669204190373421, -0.031263288110494614, -0.026117807254195213, -0.005847756750881672, 0.0159215796738863, -0.016255773603916168, 0.01637611910700798, -0.0076580955646932125, -0.010826620273292065, 0.04247354343533516, -0.012570650316774845, 0.017045650631189346, -0.01331584993749857, -0.011763269081711769, -0.01478840783238411, 0.040747903287410736, -0.02159062586724758, 0.035712022334337234, -0.00928544718772173, -0.029009392485022545, -0.03195814788341522, -0.02515578456223011, 0.04598073288798332, -0.031394265592098236, -0.014950929209589958, -0.007734905928373337, 0.025000110268592834, 0.000544744310900569, 0.01176623534411192, -0.019274704158306122, 0.0366835854947567, -0.023165304213762283, 0.033494073897600174, 0.011131266131997108, 0.018647393211722374, 0.0326278880238533, -0.04050171747803688, 0.004332876298576593, 0.00997436884790659, 0.004762817639857531, -0.035301607102155685, -0.03210917115211487, 0.04340789094567299, -0.02847307361662388, -0.000691383087541908, 0.010040812194347382, 0.01593669131398201, 0.025139354169368744, -0.039108265191316605, -0.044794704765081406]]}],                        {\"coloraxis\": {\"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"margin\": {\"t\": 60}, \"template\": {\"data\": {\"bar\": [{\"error_x\": {\"color\": \"#2a3f5f\"}, \"error_y\": {\"color\": \"#2a3f5f\"}, \"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"bar\"}], \"barpolar\": [{\"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"barpolar\"}], \"carpet\": [{\"aaxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"baxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"type\": \"carpet\"}], \"choropleth\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"choropleth\"}], \"contour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"contour\"}], \"contourcarpet\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"contourcarpet\"}], \"heatmap\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmap\"}], \"heatmapgl\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmapgl\"}], \"histogram\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"histogram\"}], \"histogram2d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2d\"}], \"histogram2dcontour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2dcontour\"}], \"mesh3d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"mesh3d\"}], \"parcoords\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"parcoords\"}], \"pie\": [{\"automargin\": true, \"type\": \"pie\"}], \"scatter\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter\"}], \"scatter3d\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter3d\"}], \"scattercarpet\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattercarpet\"}], \"scattergeo\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergeo\"}], \"scattergl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergl\"}], \"scattermapbox\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattermapbox\"}], \"scatterpolar\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolar\"}], \"scatterpolargl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolargl\"}], \"scatterternary\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterternary\"}], \"surface\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"surface\"}], \"table\": [{\"cells\": {\"fill\": {\"color\": \"#EBF0F8\"}, \"line\": {\"color\": \"white\"}}, \"header\": {\"fill\": {\"color\": \"#C8D4E3\"}, \"line\": {\"color\": \"white\"}}, \"type\": \"table\"}]}, \"layout\": {\"annotationdefaults\": {\"arrowcolor\": \"#2a3f5f\", \"arrowhead\": 0, \"arrowwidth\": 1}, \"coloraxis\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"colorscale\": {\"diverging\": [[0, \"#8e0152\"], [0.1, \"#c51b7d\"], [0.2, \"#de77ae\"], [0.3, \"#f1b6da\"], [0.4, \"#fde0ef\"], [0.5, \"#f7f7f7\"], [0.6, \"#e6f5d0\"], [0.7, \"#b8e186\"], [0.8, \"#7fbc41\"], [0.9, \"#4d9221\"], [1, \"#276419\"]], \"sequential\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"sequentialminus\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"colorway\": [\"#636efa\", \"#EF553B\", \"#00cc96\", \"#ab63fa\", \"#FFA15A\", \"#19d3f3\", \"#FF6692\", \"#B6E880\", \"#FF97FF\", \"#FECB52\"], \"font\": {\"color\": \"#2a3f5f\"}, \"geo\": {\"bgcolor\": \"white\", \"lakecolor\": \"white\", \"landcolor\": \"#E5ECF6\", \"showlakes\": true, \"showland\": true, \"subunitcolor\": \"white\"}, \"hoverlabel\": {\"align\": \"left\"}, \"hovermode\": \"closest\", \"mapbox\": {\"style\": \"light\"}, \"paper_bgcolor\": \"white\", \"plot_bgcolor\": \"#E5ECF6\", \"polar\": {\"angularaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"radialaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"scene\": {\"xaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"yaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"zaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}}, \"shapedefaults\": {\"line\": {\"color\": \"#2a3f5f\"}}, \"ternary\": {\"aaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"baxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"caxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"title\": {\"x\": 0.05}, \"xaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}, \"yaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}}}, \"xaxis\": {\"constrain\": \"domain\", \"scaleanchor\": \"y\", \"side\": \"top\", \"title\": {\"text\": \"features\"}}, \"yaxis\": {\"autorange\": \"reversed\", \"constrain\": \"domain\", \"title\": {\"text\": \"hidden_feature\"}}},                        {\"responsive\": true}                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('69928805-2446-45f9-8abc-b2abe8cc47b9');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })                };                });            </script>        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_heat(\n",
    "    w=neural_network[0].weight.data.T[:80,:30],\n",
    "    xname=\"features\",\n",
    "    yname=\"hidden_feature\",\n",
    "    x=x_map[:80],\n",
    "    y=np.arange(30)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "papermill": {
     "duration": 0.079325,
     "end_time": "2020-11-13T07:02:27.036416",
     "exception": false,
     "start_time": "2020-11-13T07:02:26.957091",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Visualize hidden layer to output\n",
    "\n",
    "> We pick first 30 hidden neurons to all our output categories for visualize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2020-11-13T07:02:27.205070Z",
     "iopub.status.busy": "2020-11-13T07:02:27.204243Z",
     "iopub.status.idle": "2020-11-13T07:02:27.252895Z",
     "shell.execute_reply": "2020-11-13T07:02:27.252134Z"
    },
    "papermill": {
     "duration": 0.137447,
     "end_time": "2020-11-13T07:02:27.253023",
     "exception": false,
     "start_time": "2020-11-13T07:02:27.115576",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>                            <div id=\"c8505b44-4802-419e-9be3-b722708b4916\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>            <script type=\"text/javascript\">                require([\"plotly\"], function(Plotly) {                    window.PLOTLYENV=window.PLOTLYENV || {};                                    if (document.getElementById(\"c8505b44-4802-419e-9be3-b722708b4916\")) {                    Plotly.newPlot(                        \"c8505b44-4802-419e-9be3-b722708b4916\",                        [{\"coloraxis\": \"coloraxis\", \"hovertemplate\": \"hidden_feature: %{x}<br>activity: %{y}<br>color: %{z}<extra></extra>\", \"type\": \"heatmap\", \"x\": [\"H1\", \"H23\", \"H13\", \"H14\", \"H10\", \"H5\", \"H21\", \"H17\", \"H16\", \"H24\", \"H26\", \"H25\", \"H27\", \"H9\", \"H22\", \"H2\", \"H19\", \"H7\", \"H29\", \"H8\", \"H12\", \"H20\", \"H0\", \"H28\", \"H3\", \"H4\", \"H11\", \"H6\", \"H18\", \"H15\"], \"y\": [\"LAYING\", \"STANDING\", \"SITTING\", \"WALKING\", \"WALKING_UPSTAIRS\", \"WALKING_DOWNSTAIRS\"], \"z\": [[0.012006659992039204, 0.02504749596118927, -0.05054490268230438, -0.0036123255267739296, 0.06033864989876747, -0.0440806970000267, -0.025289511308073997, 0.0627039298415184, -0.004913497716188431, 0.052361857146024704, 0.06477506458759308, -0.034241218119859695, -0.003959754016250372, -0.05881046503782272, 0.012120100669562817, -0.05373965576291084, -0.04720263555645943, 0.035281844437122345, -0.03211548924446106, 0.039818838238716125, 0.0201825313270092, -0.03665798157453537, -0.03300793841481209, 0.03343433514237404, -0.03660904988646507, -0.03988990932703018, -0.01602700538933277, -0.0350475087761879, -0.05114364996552467, -0.06774324178695679], [-0.06125747784972191, -0.03732733428478241, -0.051597196608781815, -0.024721939116716385, 0.007751080673187971, -0.014020700938999653, -0.02073495090007782, -0.050311129540205, -0.044722944498062134, 0.0006589772528968751, -0.05292832851409912, -0.05024425685405731, -0.049835748970508575, -0.008216852322220802, 0.011130420491099358, -0.02803882770240307, -0.013878344558179379, 0.03797111287713051, 0.010043672285974026, 0.015337511897087097, 0.06263827532529831, -0.02496456727385521, 0.01701401360332966, 0.03849151358008385, 0.056184593588113785, 0.015764269977808, 0.025702467188239098, 0.047959912568330765, 0.06515630334615707, 0.06308808922767639], [-0.05495259165763855, -0.01148668397217989, -0.027145780622959137, -0.010227535851299763, -0.049063604325056076, -0.04624645784497261, -0.04808076471090317, -0.030315522104501724, 0.014036009088158607, 0.01793554052710533, -0.030498169362545013, -0.05081673711538315, -0.039235617965459824, -0.027593964710831642, -0.017459958791732788, 0.08189254999160767, -0.05595911294221878, -0.04089134931564331, -0.03164071962237358, -0.026570675894618034, 9.02572282939218e-05, 0.015786798670887947, 0.04890076071023941, 0.03619087114930153, -0.03565608710050583, 0.01783084124326706, 0.05181039497256279, 0.008910383097827435, -0.05790971964597702, -0.0036520930007100105], [0.029661258682608604, 0.007440941873937845, 0.030792737379670143, 0.05314113199710846, 0.06039180979132652, 0.03774460032582283, -0.01424690056592226, 0.04523853212594986, 0.03032110631465912, -0.006950369570404291, 0.002785508055239916, -0.020791953429579735, -0.023315884172916412, 0.007804925553500652, -0.004358592908829451, 0.007768447510898113, 0.012682277709245682, -0.008494826965034008, -0.010153580456972122, -0.02868613600730896, 0.02124001830816269, -0.03692319989204407, 0.00039957466651685536, -0.03227231651544571, 0.0012812624918296933, -0.0008404504042118788, -0.04108739644289017, 0.008321442641317844, -0.0424710288643837, -0.0472957044839859], [0.038512974977493286, 0.06755295395851135, 0.06353450566530228, 0.04472433775663376, -0.01862558163702488, 0.05386057868599892, 0.05466281995177269, -0.06192125752568245, 0.02481723017990589, 0.06531190872192383, -0.04413173720240593, 0.0013788373908028007, -0.03059893101453781, 0.031056107953190804, 0.010892660357058048, 0.0331558883190155, -0.02075352519750595, 0.0072134132497012615, 0.012314808554947376, -0.042961955070495605, -0.0013533738674595952, -0.02482442744076252, -0.013257885351777077, -0.016505224630236626, -0.024107886478304863, -0.06684505194425583, -0.024816716089844704, -0.0530557818710804, -0.024168221279978752, -0.019488954916596413], [0.026941783726215363, -0.0404992401599884, -0.06315364688634872, -0.005862339399755001, -0.07651369273662567, 0.035588640719652176, -0.04553576558828354, 0.011525302194058895, 0.054614484310150146, -0.005222603678703308, 0.027117857709527016, -0.005794068798422813, 0.04178189858794212, 0.021515950560569763, -0.057243116199970245, -0.05369376018643379, -0.01112290471792221, 0.022647736594080925, 0.04639025777578354, -0.008388206362724304, -0.030627017840743065, 0.04818353056907654, -0.03195653855800629, -0.00026771464035846293, -0.01482978742569685, -0.01548593770712614, -0.02461121417582035, 0.0245036818087101, -0.0002664542698767036, 0.023916903883218765]]}],                        {\"coloraxis\": {\"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"margin\": {\"t\": 60}, \"template\": {\"data\": {\"bar\": [{\"error_x\": {\"color\": \"#2a3f5f\"}, \"error_y\": {\"color\": \"#2a3f5f\"}, \"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"bar\"}], \"barpolar\": [{\"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"barpolar\"}], \"carpet\": [{\"aaxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"baxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"type\": \"carpet\"}], \"choropleth\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"choropleth\"}], \"contour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"contour\"}], \"contourcarpet\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"contourcarpet\"}], \"heatmap\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmap\"}], \"heatmapgl\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmapgl\"}], \"histogram\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"histogram\"}], \"histogram2d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2d\"}], \"histogram2dcontour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2dcontour\"}], \"mesh3d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"mesh3d\"}], \"parcoords\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"parcoords\"}], \"pie\": [{\"automargin\": true, \"type\": \"pie\"}], \"scatter\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter\"}], \"scatter3d\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter3d\"}], \"scattercarpet\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattercarpet\"}], \"scattergeo\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergeo\"}], \"scattergl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergl\"}], \"scattermapbox\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattermapbox\"}], \"scatterpolar\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolar\"}], \"scatterpolargl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolargl\"}], \"scatterternary\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterternary\"}], \"surface\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"surface\"}], \"table\": [{\"cells\": {\"fill\": {\"color\": \"#EBF0F8\"}, \"line\": {\"color\": \"white\"}}, \"header\": {\"fill\": {\"color\": \"#C8D4E3\"}, \"line\": {\"color\": \"white\"}}, \"type\": \"table\"}]}, \"layout\": {\"annotationdefaults\": {\"arrowcolor\": \"#2a3f5f\", \"arrowhead\": 0, \"arrowwidth\": 1}, \"coloraxis\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"colorscale\": {\"diverging\": [[0, \"#8e0152\"], [0.1, \"#c51b7d\"], [0.2, \"#de77ae\"], [0.3, \"#f1b6da\"], [0.4, \"#fde0ef\"], [0.5, \"#f7f7f7\"], [0.6, \"#e6f5d0\"], [0.7, \"#b8e186\"], [0.8, \"#7fbc41\"], [0.9, \"#4d9221\"], [1, \"#276419\"]], \"sequential\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"sequentialminus\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"colorway\": [\"#636efa\", \"#EF553B\", \"#00cc96\", \"#ab63fa\", \"#FFA15A\", \"#19d3f3\", \"#FF6692\", \"#B6E880\", \"#FF97FF\", \"#FECB52\"], \"font\": {\"color\": \"#2a3f5f\"}, \"geo\": {\"bgcolor\": \"white\", \"lakecolor\": \"white\", \"landcolor\": \"#E5ECF6\", \"showlakes\": true, \"showland\": true, \"subunitcolor\": \"white\"}, \"hoverlabel\": {\"align\": \"left\"}, \"hovermode\": \"closest\", \"mapbox\": {\"style\": \"light\"}, \"paper_bgcolor\": \"white\", \"plot_bgcolor\": \"#E5ECF6\", \"polar\": {\"angularaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"radialaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"scene\": {\"xaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"yaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"zaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}}, \"shapedefaults\": {\"line\": {\"color\": \"#2a3f5f\"}}, \"ternary\": {\"aaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"baxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"caxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"title\": {\"x\": 0.05}, \"xaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}, \"yaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}}}, \"xaxis\": {\"constrain\": \"domain\", \"scaleanchor\": \"y\", \"side\": \"top\", \"title\": {\"text\": \"hidden_feature\"}}, \"yaxis\": {\"autorange\": \"reversed\", \"constrain\": \"domain\", \"title\": {\"text\": \"activity\"}}},                        {\"responsive\": true}                    ).then(function(){\n",
       "                            \n",
       "var gd = document.getElementById('c8505b44-4802-419e-9be3-b722708b4916');\n",
       "var x = new MutationObserver(function (mutations, observer) {{\n",
       "        var display = window.getComputedStyle(gd).display;\n",
       "        if (!display || display === 'none') {{\n",
       "            console.log([gd, 'removed!']);\n",
       "            Plotly.purge(gd);\n",
       "            observer.disconnect();\n",
       "        }}\n",
       "}});\n",
       "\n",
       "// Listen for the removal of the full notebook cells\n",
       "var notebookContainer = gd.closest('#notebook-container');\n",
       "if (notebookContainer) {{\n",
       "    x.observe(notebookContainer, {childList: true});\n",
       "}}\n",
       "\n",
       "// Listen for the clearing of the current output cell\n",
       "var outputEl = gd.closest('.output');\n",
       "if (outputEl) {{\n",
       "    x.observe(outputEl, {childList: true});\n",
       "}}\n",
       "\n",
       "                        })                };                });            </script>        </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_heat(\n",
    "    w=neural_network[3].weight.data.T[:30,:],\n",
    "    xname=\"hidden_feature\",\n",
    "    yname=\"activity\",\n",
    "    x=np.array(list(map(lambda i:f\"H{i}\",range(30)))),\n",
    "    y=np.array(list(y_map.keys())),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "papermill": {
     "duration": 0.081202,
     "end_time": "2020-11-13T07:02:27.415454",
     "exception": false,
     "start_time": "2020-11-13T07:02:27.334252",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.6"
  },
  "papermill": {
   "duration": 53.942858,
   "end_time": "2020-11-13T07:02:28.565313",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2020-11-13T07:01:34.622455",
   "version": "2.1.0"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "00d2615e32fa4d6482ea6415cee6eb91": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "02ebfcf4bfd24675ad4e463f79469caa": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "06aa2a53e1ce41e7be855b1e5a466bf8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "06f4b646b34f487e9ffb27e35fbce8f5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "08a19eefc3474cc2a434acf4bf165868": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_8395485002ff451abebc74fc6b7d0516",
       "placeholder": "​",
       "style": "IPY_MODEL_ec179c85b8cb4449a19e0e281b8be9e2",
       "value": " 6/6 [00:00&lt;00:00, 100.82it/s]"
      }
     },
     "0a090f00de49436d96bd94e0f665a2d1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "0acbd1b4e4cd4620a730b290114416d7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "0b8675e13b1a40daa98bf7386b7ddd15": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "0df3c8ba929048279f678d820a8cb8b4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "0e356f6f85024a6aa2d1498b129d1e5e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_b922e791aa10492db62e7e9d3f118823",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_4604d1c979324ff3a649aef266802659",
       "value": 1.0
      }
     },
     "0fa6ccd54e2548f78388e53298c5ab00": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_5c8413f5297045c8bccfd11a1b7f86e2",
       "placeholder": "​",
       "style": "IPY_MODEL_f3e9ce426ac141ae903b048b5a41e110",
       "value": " 2/2 [00:00&lt;00:00, 21.53it/s]"
      }
     },
     "12852d406891495787ba28949c2e9550": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "16996736198a4afaa2b3fff7eab7807f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_6a3ea4136e77491bb403bc5036202cfa",
        "IPY_MODEL_08a19eefc3474cc2a434acf4bf165868"
       ],
       "layout": "IPY_MODEL_06aa2a53e1ce41e7be855b1e5a466bf8"
      }
     },
     "184349886c9c4ac3afcdaf6d4155c49b": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "185f2a3eb192441e9b8617e66ec4aaf0": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_e6a0dc64f68045b881cc2732391fb5a6",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_4416580a4ecd4f4098d1738216f25a45",
       "value": 1.0
      }
     },
     "1bb1199db29b4175955ef5465d9cc997": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "26e86b7a0bde4e20a863f115c33e66c7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "3170465016e24dbb98b95db52a9c4df8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_a921dd3c957c44e2ac43152f31c8f3ad",
       "placeholder": "​",
       "style": "IPY_MODEL_8362f7b666f44966903f1e46ae68a77f",
       "value": " 6/6 [00:00&lt;00:00, 82.81it/s]"
      }
     },
     "34b0dbd49f6a4e3287a91ea2da5d2417": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "389670afdf5343549f9c7e416c7d1449": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_986d47a43dcb4fc4acee1d5ef59c6705",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_0a090f00de49436d96bd94e0f665a2d1",
       "value": 1.0
      }
     },
     "40694da38f2647fcbb48d825b3129841": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "407fd31d41e44289a15ee3902a92b2b3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "41d88b65a53b40f5b727575df089f6b9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_fee822ac550e413d9c62e608ad492601",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_dca507ded0c742ed80f8f4adc1e18c64",
       "value": 1.0
      }
     },
     "42421c021bf34e7393e6976de5b530e1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "42683a0567814919bd82863788412afd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_763676ab06b34d08afd31e16312a1dff",
        "IPY_MODEL_d97c8e023e954894b2965877aadb177d"
       ],
       "layout": "IPY_MODEL_c7b246463e98433d99074ccf43f5cf29"
      }
     },
     "4416580a4ecd4f4098d1738216f25a45": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "44dd10001cdc4958b29573248de88db5": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "4604d1c979324ff3a649aef266802659": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "4776362bf1e3446493283e189336fe6e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_184349886c9c4ac3afcdaf6d4155c49b",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_1bb1199db29b4175955ef5465d9cc997",
       "value": 1.0
      }
     },
     "4afd386ae4134484b1eb0d4c023e9a52": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "Epoch 4: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_66b34ab8bc8848aea5fd272649ce663a",
       "max": 29.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_6d2252f6515a41dca4eb44aeeab5fb87",
       "value": 29.0
      }
     },
     "4dd8b700c2a34949a925205e7a865e9e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_41d88b65a53b40f5b727575df089f6b9",
        "IPY_MODEL_d14944237ce74baca24c4bddddad6a77"
       ],
       "layout": "IPY_MODEL_be2bcb0b512546cea326406a2a47819e"
      }
     },
     "514a98793e934ff39c308ca45db6f522": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validation sanity check: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_9d13c2cd52e04716b0ac4b34a39392b3",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_a7fec28c90654d93ac9009dd08bd0126",
       "value": 1.0
      }
     },
     "544f4af8fd674a5e8ccd5fc66f1d4590": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_c7d63dda029149e2944cbf055cadad1d",
       "placeholder": "​",
       "style": "IPY_MODEL_b00889defe4f43b9acce033b9f3fb321",
       "value": " 6/6 [00:00&lt;00:00, 89.99it/s]"
      }
     },
     "54b669be86ce44d8aed2e53c870e3207": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "56553975b2b84da7bf9f7bd34a512fe2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_389670afdf5343549f9c7e416c7d1449",
        "IPY_MODEL_3170465016e24dbb98b95db52a9c4df8"
       ],
       "layout": "IPY_MODEL_57aef5e10d1e40c1ac71b458e516a6da"
      }
     },
     "57747b20ac4a45618c579c1e69d3f910": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "57aef5e10d1e40c1ac71b458e516a6da": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "59b4cd15431542789e75b114f4d685e5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_99c54709e90a4478a050b2ffddddc3ef",
       "placeholder": "​",
       "style": "IPY_MODEL_b3ff41784e8b471baed88e53ade4f855",
       "value": " 29/29 [00:02&lt;00:00, 11.78it/s, loss=0.069, v_num=1]"
      }
     },
     "5c8413f5297045c8bccfd11a1b7f86e2": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "66b34ab8bc8848aea5fd272649ce663a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "68b58c9bb65844d0af328b3a20eede06": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "6a3ea4136e77491bb403bc5036202cfa": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_42421c021bf34e7393e6976de5b530e1",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_7a212a942bfb4826a46e150b9ecaea85",
       "value": 1.0
      }
     },
     "6b5e31a0cf7d497383cf44909ee0a8c5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_fe17c0450c744da6a96307dc8a111dd4",
       "placeholder": "​",
       "style": "IPY_MODEL_0b8675e13b1a40daa98bf7386b7ddd15",
       "value": " 2/2 [00:00&lt;00:00, 38.75it/s]"
      }
     },
     "6d2252f6515a41dca4eb44aeeab5fb87": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "6fa3857d6b204760a1059e98fb89e123": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "7198f009ed004c45bbd3beb2cc5bf4be": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "7355f79f75bf4dc2afab8fa6379546e8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_e0c55e94da0041b99a3091b239ce8db1",
        "IPY_MODEL_59b4cd15431542789e75b114f4d685e5"
       ],
       "layout": "IPY_MODEL_44dd10001cdc4958b29573248de88db5"
      }
     },
     "763676ab06b34d08afd31e16312a1dff": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_02ebfcf4bfd24675ad4e463f79469caa",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_57747b20ac4a45618c579c1e69d3f910",
       "value": 1.0
      }
     },
     "76e0ba51ac8648fea374e308a9f942b0": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "7a212a942bfb4826a46e150b9ecaea85": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "7d3c037f340e42e894f81dc7a767c5c4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_514a98793e934ff39c308ca45db6f522",
        "IPY_MODEL_6b5e31a0cf7d497383cf44909ee0a8c5"
       ],
       "layout": "IPY_MODEL_ccf7719923fa4c17b0ca7e9c09be9a76"
      }
     },
     "7d657ced09c7422da4504564052ff313": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "7eb3504ea27e40f6b897570e53850444": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_9951340eed9f4b52b70b3d4ffbefebcf",
        "IPY_MODEL_0fa6ccd54e2548f78388e53298c5ab00"
       ],
       "layout": "IPY_MODEL_12852d406891495787ba28949c2e9550"
      }
     },
     "8362f7b666f44966903f1e46ae68a77f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "8395485002ff451abebc74fc6b7d0516": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "8406b58dbda64d70a40b0c91cf544b7d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_c4df8c12157b4c30a73bb76ddd0a3da4",
       "placeholder": "​",
       "style": "IPY_MODEL_d3c7693400fb457984c3d3ed831808b1",
       "value": " 6/6 [00:00&lt;00:00, 99.07it/s]"
      }
     },
     "8658a3792bf04b228e8872a24cdbdc42": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "8cd4ec2d379d43b8b397f4fafdcfae9a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_ab53d10ad6994da985b49e74d72be55f",
        "IPY_MODEL_cde826428127433bb4b7ad628cfd7178"
       ],
       "layout": "IPY_MODEL_76e0ba51ac8648fea374e308a9f942b0"
      }
     },
     "8e75470ae36d4d7b86263bfcbcc9d516": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "9860fdecd65d48188f140e741977cdea": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_d03aab69c5b74441984ead0c8d5d16e9",
        "IPY_MODEL_c5971cdea7e44599b6751db89c31dba6"
       ],
       "layout": "IPY_MODEL_7d657ced09c7422da4504564052ff313"
      }
     },
     "986d47a43dcb4fc4acee1d5ef59c6705": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "9951340eed9f4b52b70b3d4ffbefebcf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validation sanity check: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_26e86b7a0bde4e20a863f115c33e66c7",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_40694da38f2647fcbb48d825b3129841",
       "value": 1.0
      }
     },
     "99c54709e90a4478a050b2ffddddc3ef": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "9aa2a4396e2a407280c9653a8ebc6fd9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_68b58c9bb65844d0af328b3a20eede06",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_d0aedf67555b4965b91da2df088f9a63",
       "value": 1.0
      }
     },
     "9d13c2cd52e04716b0ac4b34a39392b3": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "9e94f504265c41cfae0ffe085fad2b0d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "a21ace2c03ef4980a1ad15c662189ae7": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "a7fec28c90654d93ac9009dd08bd0126": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "a921dd3c957c44e2ac43152f31c8f3ad": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "ab1d4e8eca164086b666e2a2b3f03d8a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "ab53d10ad6994da985b49e74d72be55f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_06f4b646b34f487e9ffb27e35fbce8f5",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_bcdacb71b69540aa9759d865eae3b2b9",
       "value": 1.0
      }
     },
     "b00889defe4f43b9acce033b9f3fb321": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "b0691a14ea594d23949955e2ec7244bb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "b36f402dd33a413eabfe5b88acb2e082": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "b3ff41784e8b471baed88e53ade4f855": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "b6219a342d254cc8a13d1788516f1e13": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_34b0dbd49f6a4e3287a91ea2da5d2417",
       "placeholder": "​",
       "style": "IPY_MODEL_ab1d4e8eca164086b666e2a2b3f03d8a",
       "value": " 29/29 [00:00&lt;00:00, 110.91it/s, loss=0.559, v_num=0]"
      }
     },
     "b922e791aa10492db62e7e9d3f118823": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "b96e7b0094394ad9a0fe9c5f081e08d2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_185f2a3eb192441e9b8617e66ec4aaf0",
        "IPY_MODEL_f762350a72f54929a040dadb33dbd163"
       ],
       "layout": "IPY_MODEL_ee039645d5a84d3fafc9b1ab7c8277d9"
      }
     },
     "bcdacb71b69540aa9759d865eae3b2b9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "be2bcb0b512546cea326406a2a47819e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "bee4a50a3f214a2abeacd68a82bd35cb": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "bee9bb67e49d4403bbd5a8bee652f731": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "c237b172d9e64593a35770a0428752bd": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_4afd386ae4134484b1eb0d4c023e9a52",
        "IPY_MODEL_b6219a342d254cc8a13d1788516f1e13"
       ],
       "layout": "IPY_MODEL_00d2615e32fa4d6482ea6415cee6eb91"
      }
     },
     "c4df8c12157b4c30a73bb76ddd0a3da4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "c51f93ceb7a94cce86257bb7d3a612b8": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "c5971cdea7e44599b6751db89c31dba6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_7198f009ed004c45bbd3beb2cc5bf4be",
       "placeholder": "​",
       "style": "IPY_MODEL_b36f402dd33a413eabfe5b88acb2e082",
       "value": " 6/6 [00:00&lt;00:00, 93.26it/s]"
      }
     },
     "c7b246463e98433d99074ccf43f5cf29": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "c7d63dda029149e2944cbf055cadad1d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "c86c4f81b89040849714372208f1ad65": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_0df3c8ba929048279f678d820a8cb8b4",
       "placeholder": "​",
       "style": "IPY_MODEL_bee9bb67e49d4403bbd5a8bee652f731",
       "value": " 6/6 [00:00&lt;00:00, 85.26it/s]"
      }
     },
     "c98f3da7b54b45468a4d13acd5d0f856": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "ccf7719923fa4c17b0ca7e9c09be9a76": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "ccfc1b3c87624187b3816ffbb67e78ec": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_4776362bf1e3446493283e189336fe6e",
        "IPY_MODEL_544f4af8fd674a5e8ccd5fc66f1d4590"
       ],
       "layout": "IPY_MODEL_bee4a50a3f214a2abeacd68a82bd35cb"
      }
     },
     "cde826428127433bb4b7ad628cfd7178": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_0acbd1b4e4cd4620a730b290114416d7",
       "placeholder": "​",
       "style": "IPY_MODEL_c98f3da7b54b45468a4d13acd5d0f856",
       "value": " 6/6 [00:00&lt;00:00, 80.17it/s]"
      }
     },
     "d03aab69c5b74441984ead0c8d5d16e9": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "info",
       "description": "Validating: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_a21ace2c03ef4980a1ad15c662189ae7",
       "max": 1.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_6fa3857d6b204760a1059e98fb89e123",
       "value": 1.0
      }
     },
     "d0aedf67555b4965b91da2df088f9a63": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "d14944237ce74baca24c4bddddad6a77": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_54b669be86ce44d8aed2e53c870e3207",
       "placeholder": "​",
       "style": "IPY_MODEL_407fd31d41e44289a15ee3902a92b2b3",
       "value": " 6/6 [00:00&lt;00:00, 94.83it/s]"
      }
     },
     "d32a690e4b194ae6b756ae6c43e6da83": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "d3c7693400fb457984c3d3ed831808b1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "d97c8e023e954894b2965877aadb177d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_c51f93ceb7a94cce86257bb7d3a612b8",
       "placeholder": "​",
       "style": "IPY_MODEL_9e94f504265c41cfae0ffe085fad2b0d",
       "value": " 6/6 [00:00&lt;00:00, 79.60it/s]"
      }
     },
     "dca507ded0c742ed80f8f4adc1e18c64": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": "initial"
      }
     },
     "dcab492df8e146d18f472ce513b297f3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_0e356f6f85024a6aa2d1498b129d1e5e",
        "IPY_MODEL_c86c4f81b89040849714372208f1ad65"
       ],
       "layout": "IPY_MODEL_b0691a14ea594d23949955e2ec7244bb"
      }
     },
     "e0c55e94da0041b99a3091b239ce8db1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "Epoch 4: 100%",
       "description_tooltip": null,
       "layout": "IPY_MODEL_8658a3792bf04b228e8872a24cdbdc42",
       "max": 29.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_d32a690e4b194ae6b756ae6c43e6da83",
       "value": 29.0
      }
     },
     "e6a0dc64f68045b881cc2732391fb5a6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "ec179c85b8cb4449a19e0e281b8be9e2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "ee039645d5a84d3fafc9b1ab7c8277d9": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "f3e9ce426ac141ae903b048b5a41e110": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "f762350a72f54929a040dadb33dbd163": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_8e75470ae36d4d7b86263bfcbcc9d516",
       "placeholder": "​",
       "style": "IPY_MODEL_f8345ecfe8ef4951a92c63495a286628",
       "value": " 6/6 [00:00&lt;00:00, 83.24it/s]"
      }
     },
     "f785b43282444de1a27bed7ed3315475": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": "inline-flex",
       "flex": null,
       "flex_flow": "row wrap",
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": "100%"
      }
     },
     "f8345ecfe8ef4951a92c63495a286628": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "fb370cebc66e499a959778a6b0e37ff5": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_9aa2a4396e2a407280c9653a8ebc6fd9",
        "IPY_MODEL_8406b58dbda64d70a40b0c91cf544b7d"
       ],
       "layout": "IPY_MODEL_f785b43282444de1a27bed7ed3315475"
      }
     },
     "fe17c0450c744da6a96307dc8a111dd4": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "fee822ac550e413d9c62e608ad492601": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": "2",
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}