LSTM_keras.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dependencies\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "import time\n",
    "import math\n",
    "\n",
    "import preprocess"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### As you can see there are 33 time series data of dimension 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(33, 36900, 5)\n",
      "33\n"
     ]
    }
   ],
   "source": [
    "## Normal sEMG dataset\n",
    "file_path = \"./SEMG_DB1/N_TXT/\"\n",
    "\n",
    "input_array, output_array = preprocess.preprocess(file_path)\n",
    "print(input_array.shape)\n",
    "print(len(output_array))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### One-hot encode the output array\n",
    "### mar - march\n",
    "### sen - extension of the leg from sit position\n",
    "### pie - flexion of the knee stand up"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['mar', 'sen', 'pie', 'sen', 'pie', 'mar', 'pie', 'sen', 'pie', 'mar', 'pie', 'pie', 'pie', 'mar', 'sen', 'mar', 'pie', 'mar', 'sen', 'pie', 'sen', 'mar', 'sen', 'mar', 'pie', 'mar', 'sen', 'sen', 'sen', 'sen', 'mar', 'mar', 'pie']\n",
      "['mar' 'sen' 'pie' 'sen' 'pie' 'mar' 'pie' 'sen' 'pie' 'mar' 'pie' 'pie'\n",
      " 'pie' 'mar' 'sen' 'mar' 'pie' 'mar' 'sen' 'pie' 'sen' 'mar' 'sen' 'mar'\n",
      " 'pie' 'mar' 'sen' 'sen' 'sen' 'sen' 'mar' 'mar' 'pie']\n",
      "[0 2 1 2 1 0 1 2 1 0 1 1 1 0 2 0 1 0 2 1 2 0 2 0 1 0 2 2 2 2 0 0 1]\n",
      "[[1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]]\n"
     ]
    }
   ],
   "source": [
    "from numpy import array\n",
    "from numpy import argmax\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "\n",
    "## One Hot Encode Prediction\n",
    "\n",
    "print(output_array)\n",
    "# convert output_array to one hot encoded classes\n",
    "values = array(output_array)\n",
    "print(values)\n",
    "# integer encode\n",
    "label_encoder = LabelEncoder()\n",
    "integer_encoded = label_encoder.fit_transform(values)\n",
    "print(integer_encoded)\n",
    "# binary encode\n",
    "onehot_encoder = OneHotEncoder(sparse=False)\n",
    "integer_encoded = integer_encoded.reshape(len(integer_encoded), 1)\n",
    "onehot_encoded = onehot_encoder.fit_transform(integer_encoded)\n",
    "print(onehot_encoded)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout\n",
    "from keras.layers import Embedding\n",
    "from keras.layers import LSTM\n",
    "\n",
    "# Retrieve from input array shape\n",
    "time_steps = input_array.shape[1]\n",
    "n_features = input_array.shape[2]\n",
    "\n",
    "# LSTM with 100 neurons and dropout.\n",
    "\n",
    "model = Sequential()\n",
    "model.add(LSTM(100, return_sequences=False, input_shape=(time_steps, n_features)))\n",
    "model.add(Dropout(0.5))\n",
    "#model.add(LSTM(100)) dramatically worse results\n",
    "model.add(Dense(3, activation='sigmoid'))\n",
    "\n",
    "model.compile(loss='binary_crossentropy',\n",
    "              optimizer='rmsprop',\n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "3/3 [==============================] - 33s 11s/step - loss: 0.6897 - accuracy: 0.3030\n",
      "Epoch 2/10\n",
      "3/3 [==============================] - 32s 11s/step - loss: 0.6869 - accuracy: 0.3333\n",
      "Epoch 3/10\n",
      "3/3 [==============================] - 32s 11s/step - loss: 0.6717 - accuracy: 0.3939\n",
      "Epoch 4/10\n",
      "3/3 [==============================] - 84s 28s/step - loss: 0.6570 - accuracy: 0.3030\n",
      "Epoch 5/10\n",
      "3/3 [==============================] - 91s 30s/step - loss: 0.6539 - accuracy: 0.3030\n",
      "Epoch 6/10\n",
      "3/3 [==============================] - 88s 29s/step - loss: 0.6387 - accuracy: 0.3333\n",
      "Epoch 7/10\n",
      "3/3 [==============================] - 89s 30s/step - loss: 0.6627 - accuracy: 0.3333\n",
      "Epoch 8/10\n",
      "1/3 [=========>....................] - ETA: 0s - loss: 0.6661 - accuracy: 0.2500"
     ]
    }
   ],
   "source": [
    "model.fit(input_array, onehot_encoded, batch_size=16, epochs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #7308 (got 1 columns instead of 5)\n",
      "    Line #7309 (got 1 columns instead of 5)\n",
      "    Line #7310 (got 1 columns instead of 5)\n",
      "    Line #7311 (got 1 columns instead of 5)\n",
      "    Line #7312 (got 1 columns instead of 5)\n",
      "    Line #7313 (got 1 columns instead of 5)\n",
      "    Line #7314 (got 1 columns instead of 5)\n",
      "    Line #7315 (got 1 columns instead of 5)\n",
      "    Line #7316 (got 1 columns instead of 5)\n",
      "    Line #7317 (got 1 columns instead of 5)\n",
      "    Line #7318 (got 1 columns instead of 5)\n",
      "    Line #7319 (got 1 columns instead of 5)\n",
      "    Line #7320 (got 1 columns instead of 5)\n",
      "    Line #7321 (got 1 columns instead of 5)\n",
      "    Line #7322 (got 1 columns instead of 5)\n",
      "    Line #7323 (got 1 columns instead of 5)\n",
      "    Line #7324 (got 1 columns instead of 5)\n",
      "    Line #7325 (got 1 columns instead of 5)\n",
      "    Line #7326 (got 1 columns instead of 5)\n",
      "    Line #7327 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #15608 (got 4 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #9547 (got 1 columns instead of 5)\n",
      "    Line #9548 (got 1 columns instead of 5)\n",
      "    Line #9549 (got 1 columns instead of 5)\n",
      "    Line #9550 (got 1 columns instead of 5)\n",
      "    Line #9551 (got 1 columns instead of 5)\n",
      "    Line #9552 (got 1 columns instead of 5)\n",
      "    Line #9553 (got 1 columns instead of 5)\n",
      "    Line #9554 (got 1 columns instead of 5)\n",
      "    Line #9555 (got 1 columns instead of 5)\n",
      "    Line #9556 (got 1 columns instead of 5)\n",
      "    Line #9557 (got 1 columns instead of 5)\n",
      "    Line #9558 (got 1 columns instead of 5)\n",
      "    Line #9559 (got 1 columns instead of 5)\n",
      "    Line #9560 (got 1 columns instead of 5)\n",
      "    Line #9561 (got 1 columns instead of 5)\n",
      "    Line #9562 (got 1 columns instead of 5)\n",
      "    Line #9563 (got 1 columns instead of 5)\n",
      "    Line #9564 (got 1 columns instead of 5)\n",
      "    Line #9565 (got 1 columns instead of 5)\n",
      "    Line #9566 (got 1 columns instead of 5)\n",
      "    Line #9567 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #6571 (got 1 columns instead of 5)\n",
      "    Line #6572 (got 1 columns instead of 5)\n",
      "    Line #6573 (got 1 columns instead of 5)\n",
      "    Line #6574 (got 1 columns instead of 5)\n",
      "    Line #6575 (got 1 columns instead of 5)\n",
      "    Line #6576 (got 1 columns instead of 5)\n",
      "    Line #6577 (got 1 columns instead of 5)\n",
      "    Line #6578 (got 1 columns instead of 5)\n",
      "    Line #6579 (got 1 columns instead of 5)\n",
      "    Line #6580 (got 1 columns instead of 5)\n",
      "    Line #6581 (got 1 columns instead of 5)\n",
      "    Line #6582 (got 1 columns instead of 5)\n",
      "    Line #6583 (got 1 columns instead of 5)\n",
      "    Line #6584 (got 1 columns instead of 5)\n",
      "    Line #6585 (got 1 columns instead of 5)\n",
      "    Line #6586 (got 1 columns instead of 5)\n",
      "    Line #6587 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #7530 (got 1 columns instead of 5)\n",
      "    Line #7531 (got 1 columns instead of 5)\n",
      "    Line #7532 (got 1 columns instead of 5)\n",
      "    Line #7533 (got 1 columns instead of 5)\n",
      "    Line #7534 (got 1 columns instead of 5)\n",
      "    Line #7535 (got 1 columns instead of 5)\n",
      "    Line #7536 (got 1 columns instead of 5)\n",
      "    Line #7537 (got 1 columns instead of 5)\n",
      "    Line #7538 (got 1 columns instead of 5)\n",
      "    Line #7539 (got 1 columns instead of 5)\n",
      "    Line #7540 (got 1 columns instead of 5)\n",
      "    Line #7541 (got 1 columns instead of 5)\n",
      "    Line #7542 (got 1 columns instead of 5)\n",
      "    Line #7543 (got 1 columns instead of 5)\n",
      "    Line #7544 (got 1 columns instead of 5)\n",
      "    Line #7545 (got 1 columns instead of 5)\n",
      "    Line #7546 (got 1 columns instead of 5)\n",
      "    Line #7547 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #9235 (got 1 columns instead of 5)\n",
      "    Line #9236 (got 1 columns instead of 5)\n",
      "    Line #9237 (got 1 columns instead of 5)\n",
      "    Line #9238 (got 1 columns instead of 5)\n",
      "    Line #9239 (got 1 columns instead of 5)\n",
      "    Line #9240 (got 1 columns instead of 5)\n",
      "    Line #9241 (got 1 columns instead of 5)\n",
      "    Line #9242 (got 1 columns instead of 5)\n",
      "    Line #9243 (got 1 columns instead of 5)\n",
      "    Line #9244 (got 1 columns instead of 5)\n",
      "    Line #9245 (got 1 columns instead of 5)\n",
      "    Line #9246 (got 1 columns instead of 5)\n",
      "    Line #9247 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #15948 (got 4 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #5689 (got 1 columns instead of 5)\n",
      "    Line #5690 (got 1 columns instead of 5)\n",
      "    Line #5691 (got 1 columns instead of 5)\n",
      "    Line #5692 (got 1 columns instead of 5)\n",
      "    Line #5693 (got 1 columns instead of 5)\n",
      "    Line #5694 (got 1 columns instead of 5)\n",
      "    Line #5695 (got 1 columns instead of 5)\n",
      "    Line #5696 (got 1 columns instead of 5)\n",
      "    Line #5697 (got 1 columns instead of 5)\n",
      "    Line #5698 (got 1 columns instead of 5)\n",
      "    Line #5699 (got 1 columns instead of 5)\n",
      "    Line #5700 (got 1 columns instead of 5)\n",
      "    Line #5701 (got 1 columns instead of 5)\n",
      "    Line #5702 (got 1 columns instead of 5)\n",
      "    Line #5703 (got 1 columns instead of 5)\n",
      "    Line #5704 (got 1 columns instead of 5)\n",
      "    Line #5705 (got 1 columns instead of 5)\n",
      "    Line #5706 (got 1 columns instead of 5)\n",
      "    Line #5707 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #24168 (got 4 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #5351 (got 1 columns instead of 5)\n",
      "    Line #5352 (got 1 columns instead of 5)\n",
      "    Line #5353 (got 1 columns instead of 5)\n",
      "    Line #5354 (got 1 columns instead of 5)\n",
      "    Line #5355 (got 1 columns instead of 5)\n",
      "    Line #5356 (got 1 columns instead of 5)\n",
      "    Line #5357 (got 1 columns instead of 5)\n",
      "    Line #5358 (got 1 columns instead of 5)\n",
      "    Line #5359 (got 1 columns instead of 5)\n",
      "    Line #5360 (got 1 columns instead of 5)\n",
      "    Line #5361 (got 1 columns instead of 5)\n",
      "    Line #5362 (got 1 columns instead of 5)\n",
      "    Line #5363 (got 1 columns instead of 5)\n",
      "    Line #5364 (got 1 columns instead of 5)\n",
      "    Line #5365 (got 1 columns instead of 5)\n",
      "    Line #5366 (got 1 columns instead of 5)\n",
      "    Line #5367 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #7392 (got 1 columns instead of 5)\n",
      "    Line #7393 (got 1 columns instead of 5)\n",
      "    Line #7394 (got 1 columns instead of 5)\n",
      "    Line #7395 (got 1 columns instead of 5)\n",
      "    Line #7396 (got 1 columns instead of 5)\n",
      "    Line #7397 (got 1 columns instead of 5)\n",
      "    Line #7398 (got 1 columns instead of 5)\n",
      "    Line #7399 (got 1 columns instead of 5)\n",
      "    Line #7400 (got 1 columns instead of 5)\n",
      "    Line #7401 (got 1 columns instead of 5)\n",
      "    Line #7402 (got 1 columns instead of 5)\n",
      "    Line #7403 (got 1 columns instead of 5)\n",
      "    Line #7404 (got 1 columns instead of 5)\n",
      "    Line #7405 (got 1 columns instead of 5)\n",
      "    Line #7406 (got 1 columns instead of 5)\n",
      "    Line #7407 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['mar' 'sen' 'pie' 'sen' 'pie' 'mar' 'pie' 'sen' 'pie' 'mar' 'pie' 'pie'\n",
      " 'pie' 'mar' 'sen' 'mar' 'pie' 'mar' 'sen' 'pie' 'sen' 'mar' 'sen' 'mar'\n",
      " 'pie' 'mar' 'sen' 'sen' 'sen' 'sen' 'mar' 'mar' 'pie']\n",
      "[0 2 1 2 1 0 1 2 1 0 1 1 1 0 2 0 1 0 2 1 2 0 2 0 1 0 2 2 2 2 0 0 1]\n",
      "[[1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 1. 0.]\n",
      " [0. 0. 1.]\n",
      " [1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [0. 0. 1.]\n",
      " [1. 0. 0.]\n",
      " [1. 0. 0.]\n",
      " [0. 1. 0.]]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #6930 (got 1 columns instead of 5)\n",
      "    Line #6931 (got 1 columns instead of 5)\n",
      "    Line #6932 (got 1 columns instead of 5)\n",
      "    Line #6933 (got 1 columns instead of 5)\n",
      "    Line #6934 (got 1 columns instead of 5)\n",
      "    Line #6935 (got 1 columns instead of 5)\n",
      "    Line #6936 (got 1 columns instead of 5)\n",
      "    Line #6937 (got 1 columns instead of 5)\n",
      "    Line #6938 (got 1 columns instead of 5)\n",
      "    Line #6939 (got 1 columns instead of 5)\n",
      "    Line #6940 (got 1 columns instead of 5)\n",
      "    Line #6941 (got 1 columns instead of 5)\n",
      "    Line #6942 (got 1 columns instead of 5)\n",
      "    Line #6943 (got 1 columns instead of 5)\n",
      "    Line #6944 (got 1 columns instead of 5)\n",
      "    Line #6945 (got 1 columns instead of 5)\n",
      "    Line #6946 (got 1 columns instead of 5)\n",
      "    Line #6947 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n",
      "/Users/hobinkang/Desktop/EMG _Leg_ Classification/preprocess.py:27: ConversionWarning: Some errors were detected !\n",
      "    Line #5853 (got 1 columns instead of 5)\n",
      "    Line #5854 (got 1 columns instead of 5)\n",
      "    Line #5855 (got 1 columns instead of 5)\n",
      "    Line #5856 (got 1 columns instead of 5)\n",
      "    Line #5857 (got 1 columns instead of 5)\n",
      "    Line #5858 (got 1 columns instead of 5)\n",
      "    Line #5859 (got 1 columns instead of 5)\n",
      "    Line #5860 (got 1 columns instead of 5)\n",
      "    Line #5861 (got 1 columns instead of 5)\n",
      "    Line #5862 (got 1 columns instead of 5)\n",
      "    Line #5863 (got 1 columns instead of 5)\n",
      "    Line #5864 (got 1 columns instead of 5)\n",
      "    Line #5865 (got 1 columns instead of 5)\n",
      "    Line #5866 (got 1 columns instead of 5)\n",
      "    Line #5867 (got 1 columns instead of 5)\n",
      "  self.array = np.genfromtxt(self.path, skip_header=7, invalid_raise=False)\n"
     ]
    }
   ],
   "source": [
    "## Abnormal sEMG dataset\n",
    "abnormal_path = \"./SEMG_DB1/A_TXT/\"\n",
    "\n",
    "abnormal_input_array, abnormal_output_array = preprocess.preprocess(file_path)\n",
    "\n",
    "# convert output_array to one hot encoded classes\n",
    "values = array(output_array)\n",
    "print(values)\n",
    "# integer encode\n",
    "label_encoder = LabelEncoder()\n",
    "integer_encoded = label_encoder.fit_transform(values)\n",
    "print(integer_encoded)\n",
    "# binary encode\n",
    "onehot_encoder = OneHotEncoder(sparse=False)\n",
    "integer_encoded = integer_encoded.reshape(len(integer_encoded), 1)\n",
    "abnormal_onehot_encoded = onehot_encoder.fit_transform(integer_encoded)\n",
    "print(abnormal_onehot_encoded)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "33/33 [==============================] - 13s 379ms/step\n"
     ]
    }
   ],
   "source": [
    "# test on abnormal dataset\n",
    "score = model.evaluate(abnormal_input_array, abnormal_onehot_encoded, batch_size=16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.6438265081607935, 0.6767676870028178]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score"
   ]
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}