Testing_Time_Shifting_1000ms.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# We will test the Time-shift 1000\n",
    "- We will use the vanilla model-Baseline (No augmentation)\n",
    "- We will use the 1000ms augmentation model here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n",
      "/home/nesl/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:523: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint8 = np.dtype([(\"qint8\", np.int8, 1)])\n",
      "/home/nesl/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:524: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint8 = np.dtype([(\"quint8\", np.uint8, 1)])\n",
      "/home/nesl/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint16 = np.dtype([(\"qint16\", np.int16, 1)])\n",
      "/home/nesl/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_quint16 = np.dtype([(\"quint16\", np.uint16, 1)])\n",
      "/home/nesl/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  _np_qint32 = np.dtype([(\"qint32\", np.int32, 1)])\n",
      "/home/nesl/anaconda3/envs/py36/lib/python3.6/site-packages/tensorflow/python/framework/dtypes.py:532: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.\n",
      "  np_resource = np.dtype([(\"resource\", np.ubyte, 1)])\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import scipy.io\n",
    "\n",
    "from keras import optimizers\n",
    "from keras.optimizers import SGD\n",
    "from keras.optimizers import Adam\n",
    "from keras.metrics import categorical_crossentropy\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "from keras.layers.convolutional import *\n",
    "from keras.callbacks import Callback\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "\n",
    "\n",
    "import keras\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, MaxPooling2D, LSTM, Dense, Dropout, Flatten, Activation\n",
    "from keras.layers.core import Permute, Reshape\n",
    "from keras import backend as K\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "from PIL import Image as img_PIL\n",
    "\n",
    "\n",
    "# random seed.\n",
    "rand_seed = 2\n",
    "\n",
    "from numpy.random import seed\n",
    "seed(rand_seed)\n",
    "from tensorflow import set_random_seed\n",
    "set_random_seed(rand_seed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Loading the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def one_hot_encoding(y_data):\n",
    "    Mapping=dict()\n",
    "    sub_dirs=['downstair','upstair','run','jump','walk','handwashing','exercise']\n",
    "\n",
    "    categories=7\n",
    "\n",
    "    count=0\n",
    "    for i in sub_dirs:\n",
    "        Mapping[i]=count\n",
    "        count=count+1\n",
    "\n",
    "    y_features2=[]\n",
    "    for i in range(len(y_data)):\n",
    "        Type=y_data[i]\n",
    "        lab=Mapping[Type]\n",
    "        y_features2.append(lab)\n",
    "\n",
    "    y_features=np.array(y_features2)\n",
    "    y_features=y_features.reshape(y_features.shape[0],1)\n",
    "    from keras.utils import to_categorical\n",
    "    y_features = to_categorical(y_features)\n",
    "\n",
    "    return y_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "path='data/'\n",
    "\n",
    "\n",
    "def get_train_data(path=path):\n",
    "    Train_data=np.load(path+'train_5000.npz')\n",
    "    Features_imu=Train_data['arr_0']\n",
    "    Labels=Train_data['arr_1']\n",
    "    Features_sound=Train_data['arr_2']\n",
    "\n",
    "    Labels = one_hot_encoding(Labels)\n",
    "    Features_imu = Features_imu.reshape(Features_imu.shape[0],1, Features_imu.shape[1], Features_imu.shape[2]) \n",
    "\n",
    "    return Features_imu,Labels,Features_sound\n",
    "\n",
    "def get_valid_data(path=path):\n",
    "    Train_data=np.load(path+'valid_1000.npz')\n",
    "    Features_imu=Train_data['arr_0']\n",
    "    Labels=Train_data['arr_1']\n",
    "    Features_sound=Train_data['arr_2']\n",
    "\n",
    "    Labels = one_hot_encoding(Labels)\n",
    "    Features_imu = Features_imu.reshape(Features_imu.shape[0],1, Features_imu.shape[1], Features_imu.shape[2]) \n",
    "\n",
    "    return Features_imu,Labels,Features_sound\n",
    "\n",
    "def get_test_data(path=path):\n",
    "    Train_data=np.load(path+'test_1377.npz')\n",
    "    Features_imu=Train_data['arr_0']\n",
    "    Labels=Train_data['arr_1']\n",
    "    Features_sound=Train_data['arr_2']\n",
    "\n",
    "    Labels = one_hot_encoding(Labels)\n",
    "    Features_imu = Features_imu.reshape(Features_imu.shape[0],1, Features_imu.shape[1], Features_imu.shape[2]) \n",
    "\n",
    "    return Features_imu,Labels,Features_sound\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5000, 1, 40, 12) (5000, 7) (5000, 193)\n",
      "Train Classes distribution:  [760. 994. 546. 780. 532. 732. 656.]\n",
      "(1000, 1, 40, 12) (1000, 7) (1000, 193)\n",
      "Valid Classes distribution:  [150. 188. 136. 141.  98. 157. 130.]\n",
      "(1377, 1, 40, 12) (1377, 7) (1377, 193)\n",
      "Test Classes distribution:  [219. 268. 124. 146. 143. 234. 243.]\n"
     ]
    }
   ],
   "source": [
    "Features_imu,Labels,Features_sound = get_train_data()\n",
    "print(Features_imu.shape, Labels.shape, Features_sound.shape)\n",
    "print('Train Classes distribution: ',np.sum(Labels, axis =0))\n",
    "\n",
    "Features_imu2,Labels2,Features_sound2 = get_valid_data()\n",
    "print(Features_imu2.shape, Labels2.shape, Features_sound2.shape)\n",
    "print('Valid Classes distribution: ',np.sum(Labels2, axis =0))\n",
    "\n",
    "Features_imu3,Labels3,Features_sound3 = get_test_data()\n",
    "print(Features_imu3.shape, Labels3.shape, Features_sound3.shape)\n",
    "print('Test Classes distribution: ',np.sum(Labels3, axis =0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_path = 'data/IMU_Audio_Fusion_1000ms'\n",
    "\n",
    "from keras.models import load_model\n",
    "\n",
    "model = load_model(model_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1377/1377 [==============================] - 2s 1ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.10806767817294642, 0.9724037763253449]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Test Accuracy\n",
    "model.evaluate([Features_sound3,Features_imu3],Labels3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Observation: All of these test labels are continuous.\n",
    "### Time window used to create a sample: 2 Second."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Time shifting, our goal is to randomly insert the samples of different continuous windows in between which is to simulate the real-time deployment setting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# First preprocess the data to get the continuous acc and gyro windows\n",
    "- data ordering: 12 sensors each with 40 samples.\n",
    "- 12 sensors are: acc_right, gyro_right, acc_left, gyro_left"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_test_data2(path=path):\n",
    "    Train_data=np.load(path+'test_1377.npz')\n",
    "    Features_imu=Train_data['arr_0']\n",
    "    Labels=Train_data['arr_1']\n",
    "    Features_sound=Train_data['arr_2']\n",
    "    \n",
    "    Features_imu = Features_imu.reshape(Features_imu.shape[0],1, Features_imu.shape[1], Features_imu.shape[2]) \n",
    "\n",
    "    return Features_imu,Labels,Features_sound"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1377, 1, 40, 12) (1377,) (1377, 193)\n"
     ]
    }
   ],
   "source": [
    "Features_imu_t1,Labels_t1,Features_sound_t1 = get_test_data2()\n",
    "print(Features_imu_t1.shape, Labels_t1.shape, Features_sound_t1.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# We see all of the labels are continuous\n",
    "Labels_t1 = np.array(Labels_t1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair'],\n",
       "      dtype='<U11')"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Labels_t1[:100]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Getting windows of data together with num_windows_together in each"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For 10 second together\n",
    "num_windows_together = 5\n",
    "\n",
    "curr_indx = 0\n",
    "final_indx = Features_imu_t1.shape[0]\n",
    "\n",
    "data_together = []\n",
    "\n",
    "while curr_indx <= final_indx:\n",
    "    IMU_together   = Features_imu_t1[curr_indx:curr_indx+num_windows_together]\n",
    "    sound_together = Features_sound_t1[curr_indx:curr_indx+num_windows_together]\n",
    "    labels_together = Labels_t1[curr_indx:curr_indx+num_windows_together]\n",
    "    \n",
    "    data_together.append([IMU_together,sound_together,labels_together])\n",
    "    \n",
    "    curr_indx = curr_indx+num_windows_together\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Now we randomly distribute the windows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "rand_seed = 4\n",
    "from numpy.random import seed\n",
    "seed(rand_seed)\n",
    "\n",
    "ordering = np.arange(len(data_together))\n",
    "np.random.shuffle(ordering)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([166, 225, 146, 215,  82,  46,  18, 113, 253,  47, 171, 275,   6,\n",
       "       123, 250, 232, 203, 124,  45, 133, 111,  24,   1, 159, 212, 101,\n",
       "       137, 128, 139, 256,  33,  11, 106,  68,  39,  69, 110,  43,  83,\n",
       "       108, 169,  64, 223, 189, 134,  88, 235,  93, 191, 260, 249, 271,\n",
       "       254, 100,  92,  71, 103,  63, 267, 144,  13,  42,  14, 268, 182,\n",
       "       140,  84,  48, 130,  55, 184, 152, 251, 178,  20, 175, 129, 266,\n",
       "       269, 218, 227, 193, 261,  91, 216, 162, 229, 116, 121,  34, 233,\n",
       "       112, 247, 231, 239, 153,  57,  81,  70,  26])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ordering[:100]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Now we again recreate the test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "indx = ordering[0]\n",
    "\n",
    "Features_imu_t1  = data_together[indx][0]\n",
    "Features_sound_t1 = data_together[indx][1]\n",
    "Labels_t1 = data_together[indx][2]\n",
    "\n",
    "\n",
    "#print(Features_imu_t1.shape,Labels_t1.shape,Features_sound_t1.shape)\n",
    "\n",
    "for i in range(1, len(data_together)):\n",
    "    indx = ordering[i]\n",
    "    IMU_together  = data_together[indx][0]\n",
    "    sound_together= data_together[indx][1]\n",
    "    labels_together=data_together[indx][2]\n",
    "    #print(indx, IMU_together.shape,sound_together.shape,labels_together.shape)\n",
    "    Features_imu_t1 = np.concatenate((Features_imu_t1,IMU_together),axis=0)\n",
    "    Labels_t1 = np.concatenate((Labels_t1,labels_together),axis=0)\n",
    "    Features_sound_t1 = np.concatenate((Features_sound_t1,sound_together),axis=0)\n",
    "    #print(indx, IMU_together.shape,sound_together.shape,labels_together.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1377, 1, 40, 12) (1377,) (1377, 193)\n"
     ]
    }
   ],
   "source": [
    "print(Features_imu_t1.shape, Labels_t1.shape, Features_sound_t1.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['walk', 'walk', 'walk', 'walk', 'walk', 'handwashing',\n",
       "       'handwashing', 'handwashing', 'handwashing', 'handwashing', 'jump',\n",
       "       'jump', 'jump', 'jump', 'jump', 'handwashing', 'handwashing',\n",
       "       'handwashing', 'handwashing', 'handwashing', 'upstair', 'upstair',\n",
       "       'upstair', 'upstair', 'upstair', 'upstair', 'upstair', 'upstair',\n",
       "       'upstair', 'upstair', 'downstair', 'downstair', 'downstair',\n",
       "       'downstair', 'downstair', 'run', 'run', 'run', 'run', 'run',\n",
       "       'exercise', 'exercise', 'exercise', 'exercise', 'exercise',\n",
       "       'upstair', 'upstair', 'upstair', 'upstair', 'upstair', 'walk',\n",
       "       'walk', 'walk', 'walk', 'walk', 'exercise', 'exercise',\n",
       "       'downstair', 'downstair', 'downstair', 'downstair', 'downstair',\n",
       "       'jump', 'jump', 'jump', 'jump', 'jump', 'exercise', 'exercise',\n",
       "       'exercise', 'exercise', 'exercise', 'exercise', 'exercise',\n",
       "       'exercise', 'exercise', 'exercise', 'handwashing', 'handwashing',\n",
       "       'handwashing', 'handwashing', 'handwashing', 'jump', 'jump',\n",
       "       'jump', 'jump', 'jump', 'upstair', 'upstair', 'upstair', 'upstair',\n",
       "       'upstair', 'jump', 'jump', 'jump', 'jump', 'jump', 'run', 'run',\n",
       "       'run'], dtype='<U11')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Labels_t1[:100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "Labels_t1 = one_hot_encoding(Labels_t1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Testing accuracy on Reshuffled data.\n",
    "## Note: not time errors are introduced so Accuracy should be same as the fusion model on original continuous test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1377/1377 [==============================] - 1s 986us/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.10806768087495676, 0.9724037763253449]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate([Features_sound_t1,Features_imu_t1],Labels_t1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Now aggregating the data and doing the timing errors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "acc_right=[]\n",
    "gyro_right=[]\n",
    "acc_left=[]\n",
    "gyro_left=[]\n",
    "\n",
    "for i in range(Features_imu_t1.shape[0]):\n",
    "    acc_right.append([Features_imu_t1[i,0,:,0],Features_imu_t1[i,0,:,1],Features_imu_t1[i,0,:,2]])\n",
    "    gyro_right.append([Features_imu_t1[i,0,:,3],Features_imu_t1[i,0,:,4],Features_imu_t1[i,0,:,5]])\n",
    "    acc_left.append([Features_imu_t1[i,0,:,6],Features_imu_t1[i,0,:,7],Features_imu_t1[i,0,:,8]])\n",
    "    gyro_left.append([Features_imu_t1[i,0,:,9],Features_imu_t1[i,0,:,10],Features_imu_t1[i,0,:,11]])\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "acc_right=np.array(acc_right)\n",
    "gyro_right=np.array(gyro_right)\n",
    "acc_left=np.array(acc_left)\n",
    "gyro_left=np.array(gyro_left)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1377, 3, 40)\n",
      "(1377, 3, 40)\n",
      "(1377, 3, 40)\n",
      "(1377, 3, 40)\n"
     ]
    }
   ],
   "source": [
    "print(acc_right.shape)\n",
    "print(gyro_right.shape)\n",
    "print(acc_left.shape)\n",
    "print(gyro_left.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "acc_right_cont= acc_right[0,]\n",
    "gyro_right_cont= gyro_right[0,]\n",
    "\n",
    "acc_left_cont= acc_left[0,]\n",
    "gyro_left_cont= gyro_left[0,]\n",
    "\n",
    "\n",
    "for i in range(1, Features_imu_t1.shape[0]):\n",
    "    #print(i)\n",
    "    acc_right_cont=np.hstack((acc_right_cont,acc_right[i,]))\n",
    "    gyro_right_cont=np.hstack((gyro_right_cont,gyro_right[i,]))\n",
    "    acc_left_cont=np.hstack((acc_left_cont,acc_left[i,]))\n",
    "    gyro_left_cont=np.hstack((gyro_left_cont,gyro_left[i,]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3, 55080)\n",
      "(3, 55080)\n",
      "(3, 55080)\n",
      "(3, 55080)\n"
     ]
    }
   ],
   "source": [
    "print(acc_right_cont.shape)\n",
    "print(gyro_right_cont.shape)\n",
    "print(acc_left_cont.shape)\n",
    "print(gyro_left_cont.shape)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Now shifting the samples"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Defining function to do the evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_data_sample_shift(shift_samples = 1):\n",
    "    sample_size = 40 #need to be 40, as decided during training\n",
    "    total_samples = acc_right_cont.shape[1]\n",
    "    \n",
    "    #print(total_samples)\n",
    "    \n",
    "    current_cursor = shift_samples\n",
    "\n",
    "    i = 0\n",
    "\n",
    "    acc_right_pro= np.array(acc_right_cont[np.newaxis,:,current_cursor:current_cursor+sample_size])\n",
    "\n",
    "    #print(acc_right_pro.shape)\n",
    "\n",
    "    gyro_right_pro= np.array(gyro_right_cont[np.newaxis,:,current_cursor:current_cursor+sample_size])\n",
    "\n",
    "    acc_left_pro= np.array(acc_left_cont[np.newaxis,:,current_cursor:current_cursor+sample_size])\n",
    "\n",
    "    gyro_left_pro= np.array(gyro_left_cont[np.newaxis,:,current_cursor:current_cursor+sample_size])\n",
    "\n",
    "    #print(np.array(acc_right_cont[np.newaxis,:,current_cursor:current_cursor+sample_size]).shape)\n",
    "\n",
    "    while current_cursor<=(total_samples-2*sample_size):\n",
    "        current_cursor = current_cursor + sample_size\n",
    "        #print(current_cursor,\" : \", i)\n",
    "        a=acc_right_pro\n",
    "        b=np.array(acc_right_cont[np.newaxis,:,current_cursor:current_cursor+sample_size])\n",
    "        acc_right_pro = np.concatenate((a,b),axis=0)\n",
    "\n",
    "\n",
    "        a=gyro_right_pro\n",
    "        b=np.array(gyro_right_cont[np.newaxis,:,current_cursor:current_cursor+sample_size])\n",
    "        gyro_right_pro = np.concatenate((a,b),axis=0)\n",
    "\n",
    "\n",
    "        a=acc_left_pro\n",
    "        b=np.array(acc_left_cont[np.newaxis,:,current_cursor:current_cursor+sample_size])\n",
    "        #print(a.shape,b.shape)\n",
    "        acc_left_pro = np.concatenate((a,b),axis=0)\n",
    "\n",
    "        a=gyro_left_pro\n",
    "        b=np.array(gyro_left_cont[np.newaxis,:,current_cursor:current_cursor+sample_size])\n",
    "        gyro_left_pro = np.concatenate((a,b),axis=0)\n",
    "        i = i+1\n",
    "        \n",
    "    IMU_processed = np.concatenate((acc_right_pro,gyro_right_pro,acc_left_pro,gyro_left_pro),axis=1)\n",
    "    IMU_processed = IMU_processed[:,np.newaxis,:,:]\n",
    "    IMU_processed = np.swapaxes(IMU_processed,2,3)\n",
    "        \n",
    "    size = IMU_processed.shape[0]\n",
    "        \n",
    "    return Features_sound_t1[:size], IMU_processed ,Labels_t1[:size]\n",
    "        \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "Accuracy_vanilla = []\n",
    "Accuracy_1000ms = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_path1 = 'data/Baseline_IMU_Audio_Fusion'\n",
    "model_path2 = 'data/IMU_Audio_Fusion_1000ms'\n",
    "\n",
    "from keras.models import load_model\n",
    "model1 = load_model(model_path1)\n",
    "model2 = load_model(model_path2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1377/1377 [==============================] - 2s 1ms/step\n",
      "1377/1377 [==============================] - 2s 1ms/step\n",
      "Shift is: 0 0.9607843137254902 0.9724037763253449\n",
      "1376/1376 [==============================] - 1s 945us/step\n",
      "1376/1376 [==============================] - 1s 919us/step\n",
      "Shift is: 1 0.9593023255813954 0.970203488372093\n",
      "1376/1376 [==============================] - 1s 895us/step\n",
      "1376/1376 [==============================] - 1s 920us/step\n",
      "Shift is: 2 0.9476744186046512 0.970203488372093\n",
      "1376/1376 [==============================] - 1s 886us/step\n",
      "1376/1376 [==============================] - 1s 926us/step\n",
      "Shift is: 3 0.9454941860465116 0.967296511627907\n",
      "1376/1376 [==============================] - 1s 877us/step\n",
      "1376/1376 [==============================] - 1s 936us/step\n",
      "Shift is: 4 0.9367732558139535 0.970203488372093\n",
      "1376/1376 [==============================] - 1s 910us/step\n",
      "1376/1376 [==============================] - 1s 868us/step\n",
      "Shift is: 5 0.934593023255814 0.9643895348837209\n",
      "1376/1376 [==============================] - 1s 878us/step\n",
      "1376/1376 [==============================] - 1s 865us/step\n",
      "Shift is: 6 0.9338662790697675 0.9622093023255814\n",
      "1376/1376 [==============================] - 1s 929us/step\n",
      "1376/1376 [==============================] - 1s 998us/step\n",
      "Shift is: 7 0.9265988372093024 0.9607558139534884\n",
      "1376/1376 [==============================] - 1s 931us/step\n",
      "1376/1376 [==============================] - 1s 958us/step\n",
      "Shift is: 8 0.9280523255813954 0.9607558139534884\n",
      "1376/1376 [==============================] - 1s 923us/step\n",
      "1376/1376 [==============================] - 1s 907us/step\n",
      "Shift is: 9 0.9200581395348837 0.9578488372093024\n",
      "1376/1376 [==============================] - 1s 901us/step\n",
      "1376/1376 [==============================] - 1s 952us/step\n",
      "Shift is: 10 0.9164244186046512 0.9593023255813954\n",
      "1376/1376 [==============================] - 1s 887us/step\n",
      "1376/1376 [==============================] - 1s 965us/step\n",
      "Shift is: 11 0.9127906976744186 0.9593023255813954\n",
      "1376/1376 [==============================] - 1s 920us/step\n",
      "1376/1376 [==============================] - 1s 985us/step\n",
      "Shift is: 12 0.9055232558139535 0.9578488372093024\n",
      "1376/1376 [==============================] - 1s 929us/step\n",
      "1376/1376 [==============================] - 1s 964us/step\n",
      "Shift is: 13 0.9084302325581395 0.9571220930232558\n",
      "1376/1376 [==============================] - 1s 933us/step\n",
      "1376/1376 [==============================] - 1s 907us/step\n",
      "Shift is: 14 0.8968023255813954 0.9549418604651163\n",
      "1376/1376 [==============================] - 1s 888us/step\n",
      "1376/1376 [==============================] - 1s 970us/step\n",
      "Shift is: 15 0.8938953488372093 0.9491279069767442\n",
      "1376/1376 [==============================] - 1s 895us/step\n",
      "1376/1376 [==============================] - 1s 985us/step\n",
      "Shift is: 16 0.8938953488372093 0.9469476744186046\n",
      "1376/1376 [==============================] - 1s 938us/step\n",
      "1376/1376 [==============================] - 1s 921us/step\n",
      "Shift is: 17 0.8975290697674418 0.9418604651162791\n",
      "1376/1376 [==============================] - 1s 937us/step\n",
      "1376/1376 [==============================] - 1s 923us/step\n",
      "Shift is: 18 0.9004360465116279 0.9447674418604651\n",
      "1376/1376 [==============================] - 1s 958us/step\n",
      "1376/1376 [==============================] - 1s 925us/step\n",
      "Shift is: 19 0.8880813953488372 0.9425872093023255\n",
      "1376/1376 [==============================] - 1s 931us/step\n",
      "1376/1376 [==============================] - 1s 949us/step\n",
      "Shift is: 20 0.8815406976744186 0.9462209302325582\n",
      "1376/1376 [==============================] - 1s 835us/step\n",
      "1376/1376 [==============================] - 1s 931us/step\n",
      "Shift is: 21 0.8684593023255814 0.9338662790697675\n",
      "1376/1376 [==============================] - 1s 928us/step\n",
      "1376/1376 [==============================] - 1s 962us/step\n",
      "Shift is: 22 0.8713662790697675 0.9324127906976745\n",
      "1376/1376 [==============================] - 1s 918us/step\n",
      "1376/1376 [==============================] - 1s 923us/step\n",
      "Shift is: 23 0.8713662790697675 0.9215116279069767\n",
      "1376/1376 [==============================] - 1s 879us/step\n",
      "1376/1376 [==============================] - 1s 934us/step\n",
      "Shift is: 24 0.8699127906976745 0.9229651162790697\n",
      "1376/1376 [==============================] - 1s 878us/step\n",
      "1376/1376 [==============================] - 1s 1ms/step\n",
      "Shift is: 25 0.8699127906976745 0.9215116279069767\n",
      "1376/1376 [==============================] - 1s 925us/step\n",
      "1376/1376 [==============================] - 1s 856us/step\n",
      "Shift is: 26 0.8662790697674418 0.9229651162790697\n",
      "1376/1376 [==============================] - 1s 936us/step\n",
      "1376/1376 [==============================] - 1s 963us/step\n",
      "Shift is: 27 0.8619186046511628 0.9156976744186046\n",
      "1376/1376 [==============================] - 1s 887us/step\n",
      "1376/1376 [==============================] - 1s 977us/step\n",
      "Shift is: 28 0.8640988372093024 0.9098837209302325\n",
      "1376/1376 [==============================] - 1s 891us/step\n",
      "1376/1376 [==============================] - 1s 967us/step\n",
      "Shift is: 29 0.846656976744186 0.9040697674418605\n",
      "1376/1376 [==============================] - 1s 850us/step\n",
      "1376/1376 [==============================] - 1s 943us/step\n",
      "Shift is: 30 0.8401162790697675 0.8953488372093024\n",
      "1376/1376 [==============================] - 1s 894us/step\n",
      "1376/1376 [==============================] - 1s 939us/step\n",
      "Shift is: 31 0.846656976744186 0.8960755813953488\n",
      "1376/1376 [==============================] - 1s 908us/step\n",
      "1376/1376 [==============================] - 1s 901us/step\n",
      "Shift is: 32 0.8415697674418605 0.8909883720930233\n",
      "1376/1376 [==============================] - 1s 959us/step\n",
      "1376/1376 [==============================] - 1s 869us/step\n",
      "Shift is: 33 0.8343023255813954 0.8880813953488372\n",
      "1376/1376 [==============================] - 1s 898us/step\n",
      "1376/1376 [==============================] - 1s 1ms/step\n",
      "Shift is: 34 0.8343023255813954 0.8866279069767442\n",
      "1376/1376 [==============================] - 1s 913us/step\n",
      "1376/1376 [==============================] - 1s 953us/step\n",
      "Shift is: 35 0.8321220930232558 0.8837209302325582\n",
      "1376/1376 [==============================] - 1s 965us/step\n",
      "1376/1376 [==============================] - 1s 865us/step\n",
      "Shift is: 36 0.8328488372093024 0.8859011627906976\n",
      "1376/1376 [==============================] - 1s 896us/step\n",
      "1376/1376 [==============================] - 1s 895us/step\n",
      "Shift is: 37 0.8343023255813954 0.8808139534883721\n",
      "1376/1376 [==============================] - 1s 959us/step\n",
      "1376/1376 [==============================] - 1s 969us/step\n",
      "Shift is: 38 0.8321220930232558 0.8742732558139535\n",
      "1376/1376 [==============================] - 1s 882us/step\n",
      "1376/1376 [==============================] - 1s 926us/step\n",
      "Shift is: 39 0.8321220930232558 0.8662790697674418\n",
      "1376/1376 [==============================] - 1s 856us/step\n",
      "1376/1376 [==============================] - 1s 966us/step\n",
      "Shift is: 40 0.8277616279069767 0.8670058139534884\n"
     ]
    }
   ],
   "source": [
    "for shift in range(41):\n",
    "    Features_sound, IMU ,Labels =  get_data_sample_shift(shift_samples = shift)\n",
    "    evaluation1 = model1.evaluate([Features_sound,IMU],Labels)\n",
    "    Accuracy_vanilla.append(evaluation1[1])\n",
    "    \n",
    "    evaluation2 = model2.evaluate([Features_sound,IMU],Labels)\n",
    "    Accuracy_1000ms.append(evaluation2[1])\n",
    "    \n",
    "    print(\"Shift is:\",shift,evaluation1[1],evaluation2[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.9607843137254902,\n",
       " 0.9593023255813954,\n",
       " 0.9476744186046512,\n",
       " 0.9454941860465116,\n",
       " 0.9367732558139535,\n",
       " 0.934593023255814,\n",
       " 0.9338662790697675,\n",
       " 0.9265988372093024,\n",
       " 0.9280523255813954,\n",
       " 0.9200581395348837,\n",
       " 0.9164244186046512,\n",
       " 0.9127906976744186,\n",
       " 0.9055232558139535,\n",
       " 0.9084302325581395,\n",
       " 0.8968023255813954,\n",
       " 0.8938953488372093,\n",
       " 0.8938953488372093,\n",
       " 0.8975290697674418,\n",
       " 0.9004360465116279,\n",
       " 0.8880813953488372,\n",
       " 0.8815406976744186,\n",
       " 0.8684593023255814,\n",
       " 0.8713662790697675,\n",
       " 0.8713662790697675,\n",
       " 0.8699127906976745,\n",
       " 0.8699127906976745,\n",
       " 0.8662790697674418,\n",
       " 0.8619186046511628,\n",
       " 0.8640988372093024,\n",
       " 0.846656976744186,\n",
       " 0.8401162790697675,\n",
       " 0.846656976744186,\n",
       " 0.8415697674418605,\n",
       " 0.8343023255813954,\n",
       " 0.8343023255813954,\n",
       " 0.8321220930232558,\n",
       " 0.8328488372093024,\n",
       " 0.8343023255813954,\n",
       " 0.8321220930232558,\n",
       " 0.8321220930232558,\n",
       " 0.8277616279069767]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Accuracy_vanilla"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.9724037763253449,\n",
       " 0.970203488372093,\n",
       " 0.970203488372093,\n",
       " 0.967296511627907,\n",
       " 0.970203488372093,\n",
       " 0.9643895348837209,\n",
       " 0.9622093023255814,\n",
       " 0.9607558139534884,\n",
       " 0.9607558139534884,\n",
       " 0.9578488372093024,\n",
       " 0.9593023255813954,\n",
       " 0.9593023255813954,\n",
       " 0.9578488372093024,\n",
       " 0.9571220930232558,\n",
       " 0.9549418604651163,\n",
       " 0.9491279069767442,\n",
       " 0.9469476744186046,\n",
       " 0.9418604651162791,\n",
       " 0.9447674418604651,\n",
       " 0.9425872093023255,\n",
       " 0.9462209302325582,\n",
       " 0.9338662790697675,\n",
       " 0.9324127906976745,\n",
       " 0.9215116279069767,\n",
       " 0.9229651162790697,\n",
       " 0.9215116279069767,\n",
       " 0.9229651162790697,\n",
       " 0.9156976744186046,\n",
       " 0.9098837209302325,\n",
       " 0.9040697674418605,\n",
       " 0.8953488372093024,\n",
       " 0.8960755813953488,\n",
       " 0.8909883720930233,\n",
       " 0.8880813953488372,\n",
       " 0.8866279069767442,\n",
       " 0.8837209302325582,\n",
       " 0.8859011627906976,\n",
       " 0.8808139534883721,\n",
       " 0.8742732558139535,\n",
       " 0.8662790697674418,\n",
       " 0.8670058139534884]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Accuracy_1000ms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "plt.figure(figsize=(10,5))\n",
    "plt.rcParams.update({'font.size': 18})\n",
    "\n",
    "y = np.arange(len(Accuracy_vanilla[:41]))\n",
    "\n",
    "y = y * 50\n",
    "\n",
    "plt.plot(y,(np.array(Accuracy_1000ms[:41])*100))\n",
    "plt.plot(y,(np.array(Accuracy_vanilla[:41])*100))\n",
    "\n",
    "#plt.title('Fused Model Accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Time Synchronization Error (ms)')\n",
    "plt.legend(['Augmentation 1000ms','No Augmentation'], loc='lower left')\n",
    "plt.show()"
   ]
  },
  {
   "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}