experimentalResults.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a href=\"https://colab.research.google.com/github/NahuelCostaCortez/DTW-Li-ion-Diagnosis/blob/main/experimentalResults.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import utils\n",
    "from joblib import load\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "chemistry = 'LFP' # 'LFP' or 'NMC' or 'NCA'\n",
    "\n",
    "path = './mat/'+chemistry\n",
    "size = utils.SIZE\n",
    "min_val = utils.LFP_MIN # utils.NMC_MIN or utils.NCA_MIN\n",
    "max_val = utils.LFP_MAX # utils.NMC_MAX or utils.NCA_MAX\n",
    "Q = utils.read_mat('./mat/Q.mat')['Qnorm'].flatten()\n",
    "y_test = utils.read_mat(path+'/y_test.mat')['y_test']\n",
    "\n",
    "x_test_1 = utils.read_mat(path+'/x_test_1.mat')['x_test'].T\n",
    "x_test_2 = utils.read_mat(path+'/x_test_2.mat')['x_test'].T\n",
    "x_test_3 = utils.read_mat(path+'/x_test_3.mat')['x_test'].T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot some data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "duty = np.random.randint(0, y_test.shape[0])\n",
    "cycle = np.random.randint(0, y_test.shape[1])\n",
    "\n",
    "datasets = [x_test_1, x_test_2, x_test_3]\n",
    "labels = ['x_test_1', 'x_test_2', 'x_test_3']\n",
    "\n",
    "plt.title(\"Voltage curve\")\n",
    "for i, data in enumerate(datasets):\n",
    "    plt.plot(Q, data[duty, cycle], label=labels[i])\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# convert data to IC\n",
    "plt.title(\"IC curve\")\n",
    "for i, data in enumerate(datasets):\n",
    "    ui, dqi = utils.IC(data[duty,cycle], Q)\n",
    "    plt.plot(utils.reduce_size(ui, dqi, size), label=labels[i])\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convert to input data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_test = y_test/100\n",
    "y_test = y_test.reshape(-1, y_test.shape[2])\n",
    "\n",
    "processed_test_datasets = []\n",
    "\n",
    "# x_test_0 is part of the training set, so we don't need it\n",
    "for dataset in datasets:\n",
    "    dataset = dataset.reshape(-1, dataset.shape[2])\n",
    "    dataset = utils.convert_to_input_data(dataset, Q, size-1, chemistry)\n",
    "    dataset = utils.normalise_data(dataset, min_val, max_val)\n",
    "    processed_test_datasets.append(dataset)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/nahuel/.local/lib/python3.6/site-packages/sklearn/base.py:315: UserWarning: Trying to unpickle estimator DecisionTreeRegressor from version 0.23.1 when using version 0.24.2. This might lead to breaking code or invalid results. Use at your own risk.\n",
      "  UserWarning)\n",
      "/home/nahuel/.local/lib/python3.6/site-packages/sklearn/base.py:315: UserWarning: Trying to unpickle estimator RandomForestRegressor from version 0.23.1 when using version 0.24.2. This might lead to breaking code or invalid results. Use at your own risk.\n",
      "  UserWarning)\n"
     ]
    }
   ],
   "source": [
    "# Models\n",
    "model_RF = load('./saved/'+chemistry+'/model-RF.joblib')\n",
    "model_FFN = tf.keras.models.load_model('./saved/'+chemistry+'/model-MLP.h5')\n",
    "model_CNN = tf.keras.models.load_model('./saved/'+chemistry+'/model-CNN.h5')\n",
    "model_DTW = tf.keras.models.load_model('./saved/'+chemistry+'/model-DTWCNN.h5')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# FFN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>1.897386</td>\n",
       "      <td>1.937745</td>\n",
       "      <td>2.004651</td>\n",
       "      <td>1.820726</td>\n",
       "      <td>1.672401</td>\n",
       "      <td>3.916146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>2.533825</td>\n",
       "      <td>2.903456</td>\n",
       "      <td>3.098658</td>\n",
       "      <td>3.285525</td>\n",
       "      <td>3.583620</td>\n",
       "      <td>11.118142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>2.303765</td>\n",
       "      <td>2.328015</td>\n",
       "      <td>2.321713</td>\n",
       "      <td>2.103639</td>\n",
       "      <td>2.159653</td>\n",
       "      <td>6.319307</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    1.897386  1.937745  2.004651  1.820726  1.672401   3.916146\n",
       "LAMPE  2.533825  2.903456  3.098658  3.285525  3.583620  11.118142\n",
       "LAMNE  2.303765  2.328015  2.321713  2.103639  2.159653   6.319307"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>2.065159</td>\n",
       "      <td>2.166340</td>\n",
       "      <td>2.234195</td>\n",
       "      <td>1.811590</td>\n",
       "      <td>1.558170</td>\n",
       "      <td>3.679649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>3.302188</td>\n",
       "      <td>2.941344</td>\n",
       "      <td>2.948705</td>\n",
       "      <td>2.735347</td>\n",
       "      <td>3.465126</td>\n",
       "      <td>11.320685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>3.413675</td>\n",
       "      <td>3.294412</td>\n",
       "      <td>3.289092</td>\n",
       "      <td>2.778900</td>\n",
       "      <td>2.350681</td>\n",
       "      <td>6.191301</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    2.065159  2.166340  2.234195  1.811590  1.558170   3.679649\n",
       "LAMPE  3.302188  2.941344  2.948705  2.735347  3.465126  11.320685\n",
       "LAMNE  3.413675  3.294412  3.289092  2.778900  2.350681   6.191301"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>1.451380</td>\n",
       "      <td>1.931783</td>\n",
       "      <td>1.889349</td>\n",
       "      <td>1.684941</td>\n",
       "      <td>1.733579</td>\n",
       "      <td>4.029704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>2.272202</td>\n",
       "      <td>3.145731</td>\n",
       "      <td>3.404697</td>\n",
       "      <td>3.524634</td>\n",
       "      <td>3.784329</td>\n",
       "      <td>11.310488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>3.047854</td>\n",
       "      <td>3.068850</td>\n",
       "      <td>2.984351</td>\n",
       "      <td>2.648276</td>\n",
       "      <td>2.443732</td>\n",
       "      <td>6.378584</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    1.451380  1.931783  1.889349  1.684941  1.733579   4.029704\n",
       "LAMPE  2.272202  3.145731  3.404697  3.524634  3.784329  11.310488\n",
       "LAMNE  3.047854  3.068850  2.984351  2.648276  2.443732   6.378584"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.get_pred(model_FFN, processed_test_datasets, y_test, False, False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>6.326812</td>\n",
       "      <td>5.696067</td>\n",
       "      <td>4.940409</td>\n",
       "      <td>3.629421</td>\n",
       "      <td>3.233855</td>\n",
       "      <td>9.211487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>5.896490</td>\n",
       "      <td>5.132423</td>\n",
       "      <td>4.262228</td>\n",
       "      <td>3.153127</td>\n",
       "      <td>5.164388</td>\n",
       "      <td>9.136563</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>7.002441</td>\n",
       "      <td>6.064502</td>\n",
       "      <td>5.020779</td>\n",
       "      <td>3.820403</td>\n",
       "      <td>6.249853</td>\n",
       "      <td>11.837801</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    6.326812  5.696067  4.940409  3.629421  3.233855   9.211487\n",
       "LAMPE  5.896490  5.132423  4.262228  3.153127  5.164388   9.136563\n",
       "LAMNE  7.002441  6.064502  5.020779  3.820403  6.249853  11.837801"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>6.326812</td>\n",
       "      <td>5.696067</td>\n",
       "      <td>4.940409</td>\n",
       "      <td>3.642654</td>\n",
       "      <td>3.148818</td>\n",
       "      <td>9.222205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>5.896490</td>\n",
       "      <td>5.132423</td>\n",
       "      <td>4.262228</td>\n",
       "      <td>3.172645</td>\n",
       "      <td>4.971949</td>\n",
       "      <td>9.793606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>7.002441</td>\n",
       "      <td>6.064502</td>\n",
       "      <td>5.020779</td>\n",
       "      <td>3.826087</td>\n",
       "      <td>6.381696</td>\n",
       "      <td>11.558122</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    6.326812  5.696067  4.940409  3.642654  3.148818   9.222205\n",
       "LAMPE  5.896490  5.132423  4.262228  3.172645  4.971949   9.793606\n",
       "LAMNE  7.002441  6.064502  5.020779  3.826087  6.381696  11.558122"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>6.326812</td>\n",
       "      <td>5.696067</td>\n",
       "      <td>4.940409</td>\n",
       "      <td>3.629421</td>\n",
       "      <td>3.201966</td>\n",
       "      <td>9.135190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>5.896490</td>\n",
       "      <td>5.132423</td>\n",
       "      <td>4.262228</td>\n",
       "      <td>3.153127</td>\n",
       "      <td>5.071271</td>\n",
       "      <td>9.375053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>7.002441</td>\n",
       "      <td>6.064502</td>\n",
       "      <td>5.020779</td>\n",
       "      <td>3.820403</td>\n",
       "      <td>6.185985</td>\n",
       "      <td>11.665670</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    6.326812  5.696067  4.940409  3.629421  3.201966   9.135190\n",
       "LAMPE  5.896490  5.132423  4.262228  3.153127  5.071271   9.375053\n",
       "LAMNE  7.002441  6.064502  5.020779  3.820403  6.185985  11.665670"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.get_pred(model_RF, processed_test_datasets, y_test, False, False)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1DConv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>1.180207</td>\n",
       "      <td>0.951202</td>\n",
       "      <td>0.733230</td>\n",
       "      <td>1.065122</td>\n",
       "      <td>1.689569</td>\n",
       "      <td>3.216216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>1.901536</td>\n",
       "      <td>1.238535</td>\n",
       "      <td>1.805410</td>\n",
       "      <td>2.801486</td>\n",
       "      <td>3.385904</td>\n",
       "      <td>10.737505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>1.182502</td>\n",
       "      <td>1.336198</td>\n",
       "      <td>1.271986</td>\n",
       "      <td>1.719755</td>\n",
       "      <td>2.833926</td>\n",
       "      <td>6.605924</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    1.180207  0.951202  0.733230  1.065122  1.689569   3.216216\n",
       "LAMPE  1.901536  1.238535  1.805410  2.801486  3.385904  10.737505\n",
       "LAMNE  1.182502  1.336198  1.271986  1.719755  2.833926   6.605924"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>0.630395</td>\n",
       "      <td>0.596814</td>\n",
       "      <td>0.860183</td>\n",
       "      <td>1.113128</td>\n",
       "      <td>1.621577</td>\n",
       "      <td>3.155063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>0.418775</td>\n",
       "      <td>1.289180</td>\n",
       "      <td>2.760915</td>\n",
       "      <td>2.629113</td>\n",
       "      <td>3.504882</td>\n",
       "      <td>10.856303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>2.050356</td>\n",
       "      <td>1.832850</td>\n",
       "      <td>2.030776</td>\n",
       "      <td>2.364677</td>\n",
       "      <td>2.864634</td>\n",
       "      <td>6.581638</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    0.630395  0.596814  0.860183  1.113128  1.621577   3.155063\n",
       "LAMPE  0.418775  1.289180  2.760915  2.629113  3.504882  10.856303\n",
       "LAMNE  2.050356  1.832850  2.030776  2.364677  2.864634   6.581638"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>1.950257</td>\n",
       "      <td>0.899970</td>\n",
       "      <td>0.604320</td>\n",
       "      <td>0.969911</td>\n",
       "      <td>1.752043</td>\n",
       "      <td>3.351155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>2.085784</td>\n",
       "      <td>1.158399</td>\n",
       "      <td>2.012010</td>\n",
       "      <td>2.957311</td>\n",
       "      <td>3.447960</td>\n",
       "      <td>10.860979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>2.867554</td>\n",
       "      <td>1.979410</td>\n",
       "      <td>1.591945</td>\n",
       "      <td>2.074955</td>\n",
       "      <td>2.930547</td>\n",
       "      <td>6.616024</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400        1000\n",
       "LLI    1.950257  0.899970  0.604320  0.969911  1.752043   3.351155\n",
       "LAMPE  2.085784  1.158399  2.012010  2.957311  3.447960  10.860979\n",
       "LAMNE  2.867554  1.979410  1.591945  2.074955  2.930547   6.616024"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.get_pred(model_CNN, processed_test_datasets, y_test, True, False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DTW-CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "V_references = utils.read_mat(path+'/V_references.mat')['V_references']\n",
    "\n",
    "processed_test_datasets_DTW = []\n",
    "\n",
    "# this usually takes some minutes, be patient!\n",
    "for V, x_test in zip(V_references, processed_test_datasets):\n",
    "    # get IC reference\n",
    "    ui, dqi = utils.IC(V, Q, utils.UI_STEP, utils.MIN_V_LFP, utils.MAX_V_LFP) # change to NCA/NMC if needed\n",
    "    new_sample = utils.reduce_size(ui, dqi, size-1)\n",
    "    IC_reference = utils.normalise_data(new_sample, min_val, max_val)\n",
    "    # get DTW images\n",
    "    processed_test_datasets_DTW.append(utils.get_DTWImages(x_test, IC_reference, size))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>0.147196</td>\n",
       "      <td>0.533658</td>\n",
       "      <td>0.728286</td>\n",
       "      <td>1.161802</td>\n",
       "      <td>1.312178</td>\n",
       "      <td>2.474986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>0.968427</td>\n",
       "      <td>0.983055</td>\n",
       "      <td>1.826342</td>\n",
       "      <td>2.673091</td>\n",
       "      <td>3.591306</td>\n",
       "      <td>8.643713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>0.177229</td>\n",
       "      <td>0.706995</td>\n",
       "      <td>1.408829</td>\n",
       "      <td>1.984942</td>\n",
       "      <td>1.932055</td>\n",
       "      <td>3.867265</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400       1000\n",
       "LLI    0.147196  0.533658  0.728286  1.161802  1.312178  2.474986\n",
       "LAMPE  0.968427  0.983055  1.826342  2.673091  3.591306  8.643713\n",
       "LAMNE  0.177229  0.706995  1.408829  1.984942  1.932055  3.867265"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>0.449007</td>\n",
       "      <td>0.842166</td>\n",
       "      <td>0.916717</td>\n",
       "      <td>1.186617</td>\n",
       "      <td>1.325558</td>\n",
       "      <td>2.155543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>0.785202</td>\n",
       "      <td>2.067852</td>\n",
       "      <td>2.765922</td>\n",
       "      <td>3.220181</td>\n",
       "      <td>3.926521</td>\n",
       "      <td>8.893262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>0.216502</td>\n",
       "      <td>0.574732</td>\n",
       "      <td>0.800012</td>\n",
       "      <td>1.117781</td>\n",
       "      <td>1.418098</td>\n",
       "      <td>4.011069</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400       1000\n",
       "LLI    0.449007  0.842166  0.916717  1.186617  1.325558  2.155543\n",
       "LAMPE  0.785202  2.067852  2.765922  3.220181  3.926521  8.893262\n",
       "LAMNE  0.216502  0.574732  0.800012  1.117781  1.418098  4.011069"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>10</th>\n",
       "      <th>50</th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>400</th>\n",
       "      <th>1000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LLI</th>\n",
       "      <td>0.800950</td>\n",
       "      <td>0.568149</td>\n",
       "      <td>0.565061</td>\n",
       "      <td>0.953905</td>\n",
       "      <td>1.120239</td>\n",
       "      <td>2.588985</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMPE</th>\n",
       "      <td>2.309712</td>\n",
       "      <td>1.322498</td>\n",
       "      <td>2.038502</td>\n",
       "      <td>2.723597</td>\n",
       "      <td>3.678208</td>\n",
       "      <td>8.632853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LAMNE</th>\n",
       "      <td>0.599813</td>\n",
       "      <td>0.553767</td>\n",
       "      <td>1.003389</td>\n",
       "      <td>1.433580</td>\n",
       "      <td>1.640689</td>\n",
       "      <td>3.940036</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           10        50        100       200       400       1000\n",
       "LLI    0.800950  0.568149  0.565061  0.953905  1.120239  2.588985\n",
       "LAMPE  2.309712  1.322498  2.038502  2.723597  3.678208  8.632853\n",
       "LAMNE  0.599813  0.553767  1.003389  1.433580  1.640689  3.940036"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "utils.get_pred(model_DTW, processed_test_datasets_DTW, y_test, False, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.6.12 (default, Oct 23 2020, 13:36:31) \n[GCC 4.8.5 20150623 (Red Hat 4.8.5-44)]"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "0f867e4773a5d549f6b33974196549f54f953cbb6ad5c913ab3ce9ac160f6abd"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}