ML4DM Assignment 1a - (EEG Object Classification).ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"ML4DM Assignment 1a - (EEG Object Classification).ipynb","provenance":[],"collapsed_sections":[]},"kernelspec":{"name":"python3","display_name":"Python 3"}},"cells":[{"cell_type":"markdown","metadata":{"id":"k6EnzP1mM7xY","colab_type":"text"},"source":["# Import"]},{"cell_type":"code","metadata":{"id":"5b4bn6fnMdM1","colab_type":"code","outputId":"ab0d506f-8f71-4335-fb8e-5fdc82757594","executionInfo":{"status":"ok","timestamp":1574225241829,"user_tz":-330,"elapsed":5062,"user":{"displayName":"201601422 Daiict","photoUrl":"","userId":"06333073350363496926"}},"colab":{"base_uri":"https://localhost:8080/","height":83}},"source":["import warnings\n","warnings.filterwarnings(\"ignore\")\n","\n","import numpy as np \n","import pandas as pd \n","import numpy as np\n","import scipy.io\n","from sklearn.decomposition import PCA\n","import tensorflow as tf\n","from tensorflow import keras\n","from tensorflow.keras import layers\n","from keras.models import Sequential\n","from keras.layers import Dense\n","from sklearn.model_selection import train_test_split\n","from sklearn.preprocessing import OneHotEncoder\n","from sklearn.model_selection import cross_val_score\n","from sklearn.model_selection import StratifiedKFold\n","import matplotlib.pyplot as plt\n","import os\n","from sklearn.preprocessing import StandardScaler\n","from sklearn.preprocessing import OneHotEncoder\n","from sklearn.model_selection import KFold"],"execution_count":1,"outputs":[{"output_type":"display_data","data":{"text/html":["<p style=\"color: red;\">\n","The default version of TensorFlow in Colab will soon switch to TensorFlow 2.x.<br>\n","We recommend you <a href=\"https://www.tensorflow.org/guide/migrate\" target=\"_blank\">upgrade</a> now \n","or ensure your notebook will continue to use TensorFlow 1.x via the <code>%tensorflow_version 1.x</code> magic:\n","<a href=\"https://colab.research.google.com/notebooks/tensorflow_version.ipynb\" target=\"_blank\">more info</a>.</p>\n"],"text/plain":["<IPython.core.display.HTML object>"]},"metadata":{"tags":[]}},{"output_type":"stream","text":["Using TensorFlow backend.\n"],"name":"stderr"}]},{"cell_type":"markdown","metadata":{"id":"wa5WeqLONC_w","colab_type":"text"},"source":["# Import Dataset"]},{"cell_type":"code","metadata":{"id":"21wSqVfJNBh8","colab_type":"code","outputId":"d2754aac-7ad1-4e32-ef13-814b0f38a789","executionInfo":{"status":"ok","timestamp":1574225276390,"user_tz":-330,"elapsed":31613,"user":{"displayName":"201601422 Daiict","photoUrl":"","userId":"06333073350363496926"}},"colab":{"base_uri":"https://localhost:8080/","height":1000}},"source":["from google.colab import drive\n","drive.mount('/content/drive')\n","\n","import scipy.io\n","mat = scipy.io.loadmat('/content/drive/My Drive/Colab Notebooks/Machine Learning For Data Mining/EEG Data/S1.mat')\n","\n","print(len(mat))\n","print(mat)\n","\n","df=pd.DataFrame.from_dict(mat,orient='index')\n","\n","print(df.head())\n","\n","data=df[0].X_3D\n","print(\"------------------\")\n","print(data.shape)\n","print(\"------------------\")\n","\n","data=data.transpose(1,2,0)\n","\n","\n","print(\"------------------\")\n","print(data.shape)\n","print(data[0])\n","print(\"------------------\")\n","\n","label=df[0].categoryLabels\n","\n","y=np.transpose(label)\n","\n","y=y-1\n","print(\"------------------\")\n","print(\"y: \",y.shape)\n","print(y)\n","print(\"------------------\")\n","\n","print(len(data))\n","\n"],"execution_count":2,"outputs":[{"output_type":"stream","text":["Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&response_type=code&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly\n","\n","Enter your authorization code:\n","··········\n","Mounted at /content/drive\n","11\n","{'__header__': b'MATLAB 5.0 MAT-file, Platform: MACI64, Created on: Tue Aug 22 19:12:25 2017', '__version__': '1.0', '__globals__': [], 'sub': array(['S1'], dtype='<U2'), 'Fs': array([[62.5]]), 'N': array([[32]], dtype=uint8), 'T': array([[5188]], dtype=uint16), 'exemplarLabels': array([[40, 64, 29, ..., 59,  3, 12]], dtype=uint8), 'categoryLabels': array([[4, 6, 3, ..., 5, 1, 1]], dtype=uint8), 'X_2D': array([[-0.26424164, -0.47586018, -0.47061757, ...,  0.05198565,\n","         0.08307296, -0.04543913],\n","       [ 0.0028675 , -0.04133774,  0.05108967, ...,  0.09525956,\n","        -0.00872008,  0.12494965],\n","       [-0.22358645,  0.02478356,  0.08543183, ...,  0.0938575 ,\n","         0.06906053,  0.09690079],\n","       ...,\n","       [ 0.17795506, -0.06740315,  0.09080031, ..., -0.11675177,\n","        -0.19592201, -0.16641006],\n","       [ 0.18887348,  0.17863007,  0.32982534, ...,  0.10712342,\n","         0.10700955,  0.0652873 ],\n","       [ 0.02932549, -0.05199817, -0.15791252, ..., -0.09192452,\n","        -0.04275392,  0.08084101]]), 'X_3D': array([[[-0.26424164,  0.0028675 , -0.22358645, ...,  0.17795506,\n","          0.18887348,  0.02932549],\n","        [-0.47586018, -0.04133774,  0.02478356, ..., -0.06740315,\n","          0.17863007, -0.05199817],\n","        [-0.47061757,  0.05108967,  0.08543183, ...,  0.09080031,\n","          0.32982534, -0.15791252],\n","        ...,\n","        [ 0.26171026,  0.14618344,  0.13419194, ..., -0.06100181,\n","         -0.02166311,  0.15729509],\n","        [ 0.01664864,  0.31495295,  0.09732464, ..., -0.22246297,\n","         -0.08866313, -0.03365586],\n","        [ 0.12796299,  0.19186635,  0.00895164, ..., -0.22968245,\n","         -0.05580644,  0.03110747]],\n","\n","       [[ 0.0694119 , -0.13818073, -0.14138363, ...,  0.08345749,\n","          0.17644546, -0.20745765],\n","        [-0.01213287,  0.00301473, -0.05908795, ...,  0.16343303,\n","          0.04568902, -0.12869698],\n","        [ 0.09521431,  0.06106164,  0.16744951, ...,  0.18882636,\n","          0.13860382, -0.18160391],\n","        ...,\n","        [ 0.1149582 ,  0.06243311,  0.12605291, ..., -0.02918271,\n","          0.02641676, -0.04570607],\n","        [ 0.17629073,  0.07099337,  0.17803752, ..., -0.16782443,\n","          0.07102925, -0.04762711],\n","        [ 0.12441933,  0.17428174,  0.17092345, ..., -0.17114856,\n","         -0.02164119, -0.04592316]],\n","\n","       [[-0.03701935, -0.07803175, -0.18127277, ...,  0.11560064,\n","         -0.11411913, -0.02656195],\n","        [ 0.03341169,  0.05085855, -0.06475672, ..., -0.00873957,\n","         -0.05621742,  0.07545269],\n","        [ 0.03736602,  0.07329164,  0.10168618, ..., -0.01803932,\n","          0.05661996,  0.0659789 ],\n","        ...,\n","        [-0.027258  ,  0.07248117,  0.08863233, ..., -0.17443814,\n","         -0.03109104, -0.1193728 ],\n","        [ 0.0365947 ,  0.03080346,  0.04844074, ..., -0.25368436,\n","         -0.06902409, -0.07272953],\n","        [-0.01021573,  0.12006075,  0.07405634, ..., -0.32575964,\n","         -0.0868055 , -0.00740015]],\n","\n","       ...,\n","\n","       [[ 0.0312043 , -0.05747297, -0.00911503, ...,  0.08848574,\n","          0.11227239, -0.07619217],\n","        [-0.04661125,  0.06127449,  0.1098959 , ...,  0.04420064,\n","          0.0718102 , -0.04163834],\n","        [ 0.02757448,  0.06193236,  0.2294212 , ...,  0.09504399,\n","          0.13624097, -0.09169688],\n","        ...,\n","        [ 0.13513299,  0.06478127,  0.04771938, ..., -0.08105062,\n","          0.11105604,  0.03617733],\n","        [ 0.09274329,  0.11914563,  0.03327877, ..., -0.15689004,\n","          0.10180986, -0.07421863],\n","        [ 0.11383087,  0.09816807,  0.06208975, ..., -0.14178735,\n","          0.04711402, -0.08828135]],\n","\n","       [[ 0.00563484, -0.06499   , -0.15210425, ...,  0.04965808,\n","         -0.0042287 , -0.05827923],\n","        [-0.05271585,  0.0147073 , -0.04961749, ...,  0.00808789,\n","          0.02755416,  0.0259631 ],\n","        [-0.05393718, -0.0362699 ,  0.10793566, ...,  0.05622378,\n","          0.07613327, -0.07331583],\n","        ...,\n","        [ 0.03119056,  0.1949324 ,  0.14322966, ..., -0.09833201,\n","          0.16318681, -0.04267162],\n","        [ 0.07318838,  0.1459948 ,  0.11655193, ..., -0.12811501,\n","          0.14241894, -0.03410432],\n","        [ 0.07352434,  0.17163425,  0.1237841 , ..., -0.14529347,\n","          0.04917896, -0.03223843]],\n","\n","       [[-0.02382347, -0.01895111, -0.17685526, ...,  0.06243312,\n","         -0.08120851, -0.08553067],\n","        [-0.01343152,  0.01229406, -0.09517451, ..., -0.00257212,\n","         -0.00555704,  0.08262864],\n","        [-0.01428892,  0.04097209,  0.04658644, ...,  0.05022952,\n","          0.05072851,  0.00500705],\n","        ...,\n","        [ 0.05198565,  0.09525956,  0.0938575 , ..., -0.11675177,\n","          0.10712342, -0.09192452],\n","        [ 0.08307296, -0.00872008,  0.06906053, ..., -0.19592201,\n","          0.10700955, -0.04275392],\n","        [-0.04543913,  0.12494965,  0.09690079, ..., -0.16641006,\n","          0.0652873 ,  0.08084101]]])}\n","                                                             0\n","__header__   b'MATLAB 5.0 MAT-file, Platform: MACI64, Creat...\n","__version__                                                1.0\n","__globals__                                                 []\n","sub                                                       [S1]\n","Fs                                                    [[62.5]]\n","------------------\n","(124, 32, 5188)\n","------------------\n","------------------\n","(32, 5188, 124)\n","[[-0.26424164  0.0694119  -0.03701935 ...  0.0312043   0.00563484\n","  -0.02382347]\n"," [ 0.0028675  -0.13818073 -0.07803175 ... -0.05747297 -0.06499\n","  -0.01895111]\n"," [-0.22358645 -0.14138363 -0.18127277 ... -0.00911503 -0.15210425\n","  -0.17685526]\n"," ...\n"," [ 0.17795506  0.08345749  0.11560064 ...  0.08848574  0.04965808\n","   0.06243312]\n"," [ 0.18887348  0.17644546 -0.11411913 ...  0.11227239 -0.0042287\n","  -0.08120851]\n"," [ 0.02932549 -0.20745765 -0.02656195 ... -0.07619217 -0.05827923\n","  -0.08553067]]\n","------------------\n","------------------\n","y:  (5188, 1)\n","[[3]\n"," [5]\n"," [2]\n"," ...\n"," [4]\n"," [0]\n"," [0]]\n","------------------\n","32\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"26c8B7a0SJL9","colab_type":"text"},"source":["# Dimensionality Reduction and Normalization"]},{"cell_type":"code","metadata":{"id":"ZRbdcD9xTFL4","colab_type":"code","colab":{}},"source":["def do_pca(input_data, variance):\n","  '''\n","  print(\"-----------------------------------------------------------------------------------------------------------------------\")\n","  print()\n","  print(\"initial shape of data: \",input_data.shape)\n","  '''\n","  scaler = StandardScaler()\n","  input_data = scaler.fit_transform(input_data)\n","\n","  pca = PCA(n_components = variance)\n","  principalComponents = pca.fit_transform(input_data)\n","\n","  '''\n","  print(\"Shape of data after PCA: \",principalComponents.shape)\n","\n","  print(\"PCA explained varience ratio: \",pca.explained_variance_ratio_)\n","  plt.plot(pca.explained_variance_ratio_)\n","  plt.title('PCA Explained Variance Ratio')\n","  plt.grid(True,which='major',axis='both', linestyle = '-')\n","  plt.grid(True,which='minor',axis='both', linestyle = '--')\n","  plt.minorticks_on()\n","  plt.show()\n","\n","  print(\"-----------------------------------------------------------------------------------------------------------------------\")\n","  '''\n","  return principalComponents"],"execution_count":0,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"9E84IHOMYSQO","colab_type":"text"},"source":["# Binary Classification"]},{"cell_type":"code","metadata":{"id":"TzGsUTDgSIh7","colab_type":"code","outputId":"28a30914-830d-422b-8525-210e3d2d02bd","executionInfo":{"status":"ok","timestamp":1574192276073,"user_tz":-330,"elapsed":4065594,"user":{"displayName":"201601422 Daiict","photoUrl":"","userId":"06333073350363496926"}},"colab":{"base_uri":"https://localhost:8080/","height":1000}},"source":["mean_accuracy = []\n","for i in range(0,32):\n","\n","  print(\"Time Stamp: \",i)\n","\n","  principal_components = do_pca(data[i],0.95)\n","  bin_data = np.array([])\n","  bin_labels = np.array([])\n","  for j in range(0,5188):\n","    if(y[j]==1):\n","      bin_labels=np.append(bin_labels,1)\n","      bin_data=np.append(bin_data,[principal_components[j]])\n","    elif(y[j]==5):\n","      bin_labels=np.append(bin_labels,0)\n","      bin_data=np.append(bin_data,[principal_components[j]])\n","\n","  bin_data=bin_data.reshape(bin_labels.size,principal_components[0].size)\n","  bin_labels=np.transpose(bin_labels)\n","  bin_labels = bin_labels.reshape((1728, 1))\n","\n","  oh = OneHotEncoder()\n","  bin_labels = oh.fit_transform(bin_labels)\n","\n","  #print(bin_data.shape)\n","  #print(bin_labels.shape)\n","\n","  accuracy = []\n","\n","  cv10 = KFold(n_splits=10, random_state=42, shuffle=True)\n","  k = 0\n","  for train_index, test_index in cv10.split(bin_data):\n","    k =k+1\n","    #print(\"Train Index: \", train_index, \"\\n\")\n","    #print(\"Test Index: \", test_index)\n","      \n","    X_train, X_test, y_train, y_test = bin_data[train_index], bin_data[test_index], bin_labels[train_index], bin_labels[test_index]\n","    model = Sequential([\n","            Dense(32, input_dim = X_train.shape[1], activation = 'relu'),\n","            Dense(32, activation = 'relu'),\n","            Dense(2, activation = 'softmax')\n","        ])\n","    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n","    history = model.fit(X_train, y_train, epochs=20, batch_size=256, validation_split=0.1, shuffle=True, verbose = 0)\n","    test_loss, test_acc = model.evaluate(X_test, y_test, verbose=0)\n","    accuracy.append(test_acc)\n","    print(\"For Fold: \",k, \" Accuracy: \",accuracy[k-1])\n","  mean_accuracy.append(np.mean(accuracy))\n","      \n","\n","print()\n","for i in range(0,32):\n","  print(\"For time stamp: \",i,\" Mean Accuracy: \",mean_accuracy[i])\n","\n","plt.figure()\n","plt.plot(np.arange(0,32,1), mean_accuracy)\n","plt.grid()\n","plt.xlabel('Time')\n","plt.ylabel('Accuracy')\n","plt.show()   \n"],"execution_count":0,"outputs":[{"output_type":"stream","text":["Time Stamp:  0\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:66: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:541: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4432: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:793: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3576: The name tf.log is deprecated. Please use tf.math.log instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/math_grad.py:1424: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n","Instructions for updating:\n","Use tf.where in 2.0, which has the same broadcast rule as np.where\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1033: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1020: The name tf.assign is deprecated. Please use tf.compat.v1.assign instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3005: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:190: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:197: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:207: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:216: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:223: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.\n","\n","For Fold:  1  Accuracy:  0.4855491334647802\n","For Fold:  2  Accuracy:  0.5028901735826724\n","For Fold:  3  Accuracy:  0.5260115620717837\n","For Fold:  4  Accuracy:  0.45664740091114375\n","For Fold:  5  Accuracy:  0.5433526032232825\n","For Fold:  6  Accuracy:  0.5028901739272079\n","For Fold:  7  Accuracy:  0.462427747042882\n","For Fold:  8  Accuracy:  0.45086705253992465\n","For Fold:  9  Accuracy:  0.5000000006930773\n","For Fold:  10  Accuracy:  0.4534883707068687\n","Time Stamp:  1\n","For Fold:  1  Accuracy:  0.5144508684301652\n","For Fold:  2  Accuracy:  0.5086705219538915\n","For Fold:  3  Accuracy:  0.5202312140451001\n","For Fold:  4  Accuracy:  0.5086705204034816\n","For Fold:  5  Accuracy:  0.47976878629943537\n","For Fold:  6  Accuracy:  0.5317919080312541\n","For Fold:  7  Accuracy:  0.49132947985492\n","For Fold:  8  Accuracy:  0.5838150290740018\n","For Fold:  9  Accuracy:  0.47093023394429406\n","For Fold:  10  Accuracy:  0.529069768828015\n","Time Stamp:  2\n","For Fold:  1  Accuracy:  0.45664740056660824\n","For Fold:  2  Accuracy:  0.47398843939249224\n","For Fold:  3  Accuracy:  0.4566473997052694\n","For Fold:  4  Accuracy:  0.5144508673965587\n","For Fold:  5  Accuracy:  0.4739884394786261\n","For Fold:  6  Accuracy:  0.4739884413735715\n","For Fold:  7  Accuracy:  0.4450867057191154\n","For Fold:  8  Accuracy:  0.42774566525668767\n","For Fold:  9  Accuracy:  0.4593023241952408\n","For Fold:  10  Accuracy:  0.48255813953488375\n","Time Stamp:  3\n","For Fold:  1  Accuracy:  0.4913294818359992\n","For Fold:  2  Accuracy:  0.5144508687747007\n","For Fold:  3  Accuracy:  0.5144508691192362\n","For Fold:  4  Accuracy:  0.5144508680856297\n","For Fold:  5  Accuracy:  0.41618497127053367\n","For Fold:  6  Accuracy:  0.5491329496995562\n","For Fold:  7  Accuracy:  0.5202312140451001\n","For Fold:  8  Accuracy:  0.4855491331202446\n","For Fold:  9  Accuracy:  0.48837209371633306\n","For Fold:  10  Accuracy:  0.4534883707068687\n","Time Stamp:  4\n","For Fold:  1  Accuracy:  0.5028901744440112\n","For Fold:  2  Accuracy:  0.4739884394786261\n","For Fold:  3  Accuracy:  0.5028901754776177\n","For Fold:  4  Accuracy:  0.5953757239214946\n","For Fold:  5  Accuracy:  0.5028901747885467\n","For Fold:  6  Accuracy:  0.5202312159400455\n","For Fold:  7  Accuracy:  0.5144508675688264\n","For Fold:  8  Accuracy:  0.4682080926578169\n","For Fold:  9  Accuracy:  0.47674418604651164\n","For Fold:  10  Accuracy:  0.48255813953488375\n","Time Stamp:  5\n","For Fold:  1  Accuracy:  0.4913294814914637\n","For Fold:  2  Accuracy:  0.531791909581664\n","For Fold:  3  Accuracy:  0.49710982710639867\n","For Fold:  4  Accuracy:  0.4682080926578169\n","For Fold:  5  Accuracy:  0.4797687871607742\n","For Fold:  6  Accuracy:  0.5549132965203655\n","For Fold:  7  Accuracy:  0.5144508687747007\n","For Fold:  8  Accuracy:  0.4855491338093157\n","For Fold:  9  Accuracy:  0.5348837209302325\n","For Fold:  10  Accuracy:  0.5058139534883721\n","Time Stamp:  6\n","For Fold:  1  Accuracy:  0.5838150294185374\n","For Fold:  2  Accuracy:  0.589595377445221\n","For Fold:  3  Accuracy:  0.49710982831227296\n","For Fold:  4  Accuracy:  0.5317919092371285\n","For Fold:  5  Accuracy:  0.5028901739272079\n","For Fold:  6  Accuracy:  0.49710982762320194\n","For Fold:  7  Accuracy:  0.5606936421353004\n","For Fold:  8  Accuracy:  0.4913294814914637\n","For Fold:  9  Accuracy:  0.5465116279069767\n","For Fold:  10  Accuracy:  0.5116279055905897\n","Time Stamp:  7\n","For Fold:  1  Accuracy:  0.5433526013283371\n","For Fold:  2  Accuracy:  0.49132948063012494\n","For Fold:  3  Accuracy:  0.5260115608659094\n","For Fold:  4  Accuracy:  0.48554913432611896\n","For Fold:  5  Accuracy:  0.5144508675688264\n","For Fold:  6  Accuracy:  0.48554913432611896\n","For Fold:  7  Accuracy:  0.6300578055354212\n","For Fold:  8  Accuracy:  0.543352602878747\n","For Fold:  9  Accuracy:  0.482558140227961\n","For Fold:  10  Accuracy:  0.5523255800091943\n","Time Stamp:  8\n","For Fold:  1  Accuracy:  0.5780346822531927\n","For Fold:  2  Accuracy:  0.508670522298427\n","For Fold:  3  Accuracy:  0.5606936436857102\n","For Fold:  4  Accuracy:  0.5317919076867186\n","For Fold:  5  Accuracy:  0.49132948063012494\n","For Fold:  6  Accuracy:  0.49710982676186316\n","For Fold:  7  Accuracy:  0.5260115624163192\n","For Fold:  8  Accuracy:  0.5317919085480575\n","For Fold:  9  Accuracy:  0.5523255800091943\n","For Fold:  10  Accuracy:  0.5581395348837209\n","Time Stamp:  9\n","For Fold:  1  Accuracy:  0.6127167635225836\n","For Fold:  2  Accuracy:  0.5202312139589662\n","For Fold:  3  Accuracy:  0.5260115612104449\n","For Fold:  4  Accuracy:  0.5664739893006452\n","For Fold:  5  Accuracy:  0.5491329500440917\n","For Fold:  6  Accuracy:  0.6184971112047317\n","For Fold:  7  Accuracy:  0.5722543354323834\n","For Fold:  8  Accuracy:  0.6589595386058609\n","For Fold:  9  Accuracy:  0.6569767441860465\n","For Fold:  10  Accuracy:  0.6162790683812873\n","Time Stamp:  10\n","For Fold:  1  Accuracy:  0.6242774576810054\n","For Fold:  2  Accuracy:  0.589595377445221\n","For Fold:  3  Accuracy:  0.6473988456533134\n","For Fold:  4  Accuracy:  0.5953757246105658\n","For Fold:  5  Accuracy:  0.6069364182521843\n","For Fold:  6  Accuracy:  0.6705202322474794\n","For Fold:  7  Accuracy:  0.6242774580255409\n","For Fold:  8  Accuracy:  0.6994219670405967\n","For Fold:  9  Accuracy:  0.6104651162790697\n","For Fold:  10  Accuracy:  0.6686046525489452\n","Time Stamp:  11\n","For Fold:  1  Accuracy:  0.5375722557134022\n","For Fold:  2  Accuracy:  0.5780346831145314\n","For Fold:  3  Accuracy:  0.5260115608659094\n","For Fold:  4  Accuracy:  0.6069364175631132\n","For Fold:  5  Accuracy:  0.6705202332810859\n","For Fold:  6  Accuracy:  0.566473990161984\n","For Fold:  7  Accuracy:  0.5838150309689473\n","For Fold:  8  Accuracy:  0.6473988441029036\n","For Fold:  9  Accuracy:  0.6104651176652243\n","For Fold:  10  Accuracy:  0.5813953474510548\n","Time Stamp:  12\n","For Fold:  1  Accuracy:  0.5722543369827932\n","For Fold:  2  Accuracy:  0.5722543369827932\n","For Fold:  3  Accuracy:  0.5780346825977282\n","For Fold:  4  Accuracy:  0.6589595386058609\n","For Fold:  5  Accuracy:  0.6473988449642424\n","For Fold:  6  Accuracy:  0.5838150309689473\n","For Fold:  7  Accuracy:  0.5895953764116144\n","For Fold:  8  Accuracy:  0.6300578036404758\n","For Fold:  9  Accuracy:  0.6395348837209303\n","For Fold:  10  Accuracy:  0.6220930246419685\n","Time Stamp:  13\n","For Fold:  1  Accuracy:  0.5953757235769591\n","For Fold:  2  Accuracy:  0.5838150299353406\n","For Fold:  3  Accuracy:  0.560693641704631\n","For Fold:  4  Accuracy:  0.6184971101711251\n","For Fold:  5  Accuracy:  0.6011560697086974\n","For Fold:  6  Accuracy:  0.5953757242660302\n","For Fold:  7  Accuracy:  0.6647398854266702\n","For Fold:  8  Accuracy:  0.6300578048463502\n","For Fold:  9  Accuracy:  0.6279069781303406\n","For Fold:  10  Accuracy:  0.5988372106884801\n","Time Stamp:  14\n","For Fold:  1  Accuracy:  0.5260115617272482\n","For Fold:  2  Accuracy:  0.5838150306244116\n","For Fold:  3  Accuracy:  0.6127167647284579\n","For Fold:  4  Accuracy:  0.6358381513226239\n","For Fold:  5  Accuracy:  0.5838150302798761\n","For Fold:  6  Accuracy:  0.6011560703977684\n","For Fold:  7  Accuracy:  0.5895953771006854\n","For Fold:  8  Accuracy:  0.6589595396394674\n","For Fold:  9  Accuracy:  0.5813953481441321\n","For Fold:  10  Accuracy:  0.6104651162790697\n","Time Stamp:  15\n","For Fold:  1  Accuracy:  0.5144508684301652\n","For Fold:  2  Accuracy:  0.5549132949699557\n","For Fold:  3  Accuracy:  0.6531791921295872\n","For Fold:  4  Accuracy:  0.5606936423075681\n","For Fold:  5  Accuracy:  0.5722543354323834\n","For Fold:  6  Accuracy:  0.5606936421353004\n","For Fold:  7  Accuracy:  0.566473990161984\n","For Fold:  8  Accuracy:  0.6011560703977684\n","For Fold:  9  Accuracy:  0.5697674418604651\n","For Fold:  10  Accuracy:  0.5755813953488372\n","Time Stamp:  16\n","For Fold:  1  Accuracy:  0.566473990161984\n","For Fold:  2  Accuracy:  0.6531791928186582\n","For Fold:  3  Accuracy:  0.6473988449642424\n","For Fold:  4  Accuracy:  0.5953757242660302\n","For Fold:  5  Accuracy:  0.5549132953144912\n","For Fold:  6  Accuracy:  0.554913296864901\n","For Fold:  7  Accuracy:  0.566473990161984\n","For Fold:  8  Accuracy:  0.6184971112047317\n","For Fold:  9  Accuracy:  0.5988372106884801\n","For Fold:  10  Accuracy:  0.6162790711535964\n","Time Stamp:  17\n","For Fold:  1  Accuracy:  0.5433526032232825\n","For Fold:  2  Accuracy:  0.6358381516671594\n","For Fold:  3  Accuracy:  0.6300578051908857\n","For Fold:  4  Accuracy:  0.6300578055354212\n","For Fold:  5  Accuracy:  0.6069364175631132\n","For Fold:  6  Accuracy:  0.6127167640393869\n","For Fold:  7  Accuracy:  0.6011560707423039\n","For Fold:  8  Accuracy:  0.6184971099988573\n","For Fold:  9  Accuracy:  0.622093023255814\n","For Fold:  10  Accuracy:  0.5930232558139535\n","Time Stamp:  18\n","For Fold:  1  Accuracy:  0.589595377445221\n","For Fold:  2  Accuracy:  0.601156071431375\n","For Fold:  3  Accuracy:  0.6647398849098669\n","For Fold:  4  Accuracy:  0.6358381520116949\n","For Fold:  5  Accuracy:  0.6011560710868394\n","For Fold:  6  Accuracy:  0.5549132965203655\n","For Fold:  7  Accuracy:  0.5664739905065195\n","For Fold:  8  Accuracy:  0.6184971112047317\n","For Fold:  9  Accuracy:  0.6337209316187127\n","For Fold:  10  Accuracy:  0.6046511627906976\n","Time Stamp:  19\n","For Fold:  1  Accuracy:  0.6358381513226239\n","For Fold:  2  Accuracy:  0.6473988453087779\n","For Fold:  3  Accuracy:  0.5780346838036025\n","For Fold:  4  Accuracy:  0.5895953762393467\n","For Fold:  5  Accuracy:  0.6242774576810054\n","For Fold:  6  Accuracy:  0.5722543373273287\n","For Fold:  7  Accuracy:  0.5606936417907649\n","For Fold:  8  Accuracy:  0.5838150294185374\n","For Fold:  9  Accuracy:  0.598837207916171\n","For Fold:  10  Accuracy:  0.6104651148929152\n","Time Stamp:  20\n","For Fold:  1  Accuracy:  0.5549132949699557\n","For Fold:  2  Accuracy:  0.5260115627608547\n","For Fold:  3  Accuracy:  0.589595377445221\n","For Fold:  4  Accuracy:  0.554913296864901\n","For Fold:  5  Accuracy:  0.6184971105156606\n","For Fold:  6  Accuracy:  0.5144508672242909\n","For Fold:  7  Accuracy:  0.5664739905065195\n","For Fold:  8  Accuracy:  0.5664739886115742\n","For Fold:  9  Accuracy:  0.5813953502233639\n","For Fold:  10  Accuracy:  0.5523255813953488\n","Time Stamp:  21\n","For Fold:  1  Accuracy:  0.5375722545075279\n","For Fold:  2  Accuracy:  0.5260115612104449\n","For Fold:  3  Accuracy:  0.5664739905065195\n","For Fold:  4  Accuracy:  0.5375722550243311\n","For Fold:  5  Accuracy:  0.5780346838036025\n","For Fold:  6  Accuracy:  0.601156071431375\n","For Fold:  7  Accuracy:  0.5838150309689473\n","For Fold:  8  Accuracy:  0.4913294802855894\n","For Fold:  9  Accuracy:  0.5581395362698754\n","For Fold:  10  Accuracy:  0.5116279069767442\n","Time Stamp:  22\n","For Fold:  1  Accuracy:  0.5953757242660302\n","For Fold:  2  Accuracy:  0.601156071431375\n","For Fold:  3  Accuracy:  0.531791909581664\n","For Fold:  4  Accuracy:  0.5780346827699959\n","For Fold:  5  Accuracy:  0.5433526032232825\n","For Fold:  6  Accuracy:  0.5491329488382174\n","For Fold:  7  Accuracy:  0.5838150309689473\n","For Fold:  8  Accuracy:  0.5606936417907649\n","For Fold:  9  Accuracy:  0.6337209295394809\n","For Fold:  10  Accuracy:  0.4883720930232558\n","Time Stamp:  23\n","For Fold:  1  Accuracy:  0.5433526012422032\n","For Fold:  2  Accuracy:  0.5780346838036025\n","For Fold:  3  Accuracy:  0.55491329617583\n","For Fold:  4  Accuracy:  0.5895953764116144\n","For Fold:  5  Accuracy:  0.5549132949699557\n","For Fold:  6  Accuracy:  0.6184971099988573\n","For Fold:  7  Accuracy:  0.5895953777897565\n","For Fold:  8  Accuracy:  0.5953757242660302\n","For Fold:  9  Accuracy:  0.5813953502233639\n","For Fold:  10  Accuracy:  0.5523255813953488\n","Time Stamp:  24\n","For Fold:  1  Accuracy:  0.5549132949699557\n","For Fold:  2  Accuracy:  0.5895953758948111\n","For Fold:  3  Accuracy:  0.578034684148138\n","For Fold:  4  Accuracy:  0.5433526020174082\n","For Fold:  5  Accuracy:  0.5722543373273287\n","For Fold:  6  Accuracy:  0.4682080926578169\n","For Fold:  7  Accuracy:  0.5722543366382576\n","For Fold:  8  Accuracy:  0.5260115608659094\n","For Fold:  9  Accuracy:  0.5930232558139535\n","For Fold:  10  Accuracy:  0.5058139541814494\n","Time Stamp:  25\n","For Fold:  1  Accuracy:  0.5722543361214544\n","For Fold:  2  Accuracy:  0.6011560707423039\n","For Fold:  3  Accuracy:  0.5780346825977282\n","For Fold:  4  Accuracy:  0.5780346822531927\n","For Fold:  5  Accuracy:  0.5433526013283371\n","For Fold:  6  Accuracy:  0.5606936436857102\n","For Fold:  7  Accuracy:  0.5895953777897565\n","For Fold:  8  Accuracy:  0.5895953771006854\n","For Fold:  9  Accuracy:  0.6569767427998919\n","For Fold:  10  Accuracy:  0.5930232572001081\n","Time Stamp:  26\n","For Fold:  1  Accuracy:  0.6011560702255007\n","For Fold:  2  Accuracy:  0.6589595392949319\n","For Fold:  3  Accuracy:  0.5953757227156203\n","For Fold:  4  Accuracy:  0.6011560710868394\n","For Fold:  5  Accuracy:  0.6531791928186582\n","For Fold:  6  Accuracy:  0.5953757230601558\n","For Fold:  7  Accuracy:  0.6300578048463502\n","For Fold:  8  Accuracy:  0.6069364175631132\n","For Fold:  9  Accuracy:  0.6511627906976745\n","For Fold:  10  Accuracy:  0.5755813967349918\n","Time Stamp:  27\n","For Fold:  1  Accuracy:  0.6300578048463502\n","For Fold:  2  Accuracy:  0.6127167650729934\n","For Fold:  3  Accuracy:  0.6069364175631132\n","For Fold:  4  Accuracy:  0.5722543356046511\n","For Fold:  5  Accuracy:  0.6184971108601961\n","For Fold:  6  Accuracy:  0.6647398864602767\n","For Fold:  7  Accuracy:  0.6069364172185776\n","For Fold:  8  Accuracy:  0.5664739905065195\n","For Fold:  9  Accuracy:  0.627906976744186\n","For Fold:  10  Accuracy:  0.6453488358231478\n","Time Stamp:  28\n","For Fold:  1  Accuracy:  0.6127167643839224\n","For Fold:  2  Accuracy:  0.6011560695364296\n","For Fold:  3  Accuracy:  0.578034684148138\n","For Fold:  4  Accuracy:  0.6069364175631132\n","For Fold:  5  Accuracy:  0.543352602878747\n","For Fold:  6  Accuracy:  0.5953757227156203\n","For Fold:  7  Accuracy:  0.5722543366382576\n","For Fold:  8  Accuracy:  0.5780346838036025\n","For Fold:  9  Accuracy:  0.5755813953488372\n","For Fold:  10  Accuracy:  0.6046511627906976\n","Time Stamp:  29\n","For Fold:  1  Accuracy:  0.6127167636948514\n","For Fold:  2  Accuracy:  0.5895953777897565\n","For Fold:  3  Accuracy:  0.6300578055354212\n","For Fold:  4  Accuracy:  0.5606936421353004\n","For Fold:  5  Accuracy:  0.6184971115492672\n","For Fold:  6  Accuracy:  0.5953757232324236\n","For Fold:  7  Accuracy:  0.6531791910959807\n","For Fold:  8  Accuracy:  0.55491329617583\n","For Fold:  9  Accuracy:  0.6104651155859925\n","For Fold:  10  Accuracy:  0.5290697667487833\n","Time Stamp:  30\n","For Fold:  1  Accuracy:  0.5722543354323834\n","For Fold:  2  Accuracy:  0.5838150290740018\n","For Fold:  3  Accuracy:  0.5780346825977282\n","For Fold:  4  Accuracy:  0.473988441029036\n","For Fold:  5  Accuracy:  0.6184971112047317\n","For Fold:  6  Accuracy:  0.5375722560579377\n","For Fold:  7  Accuracy:  0.6011560707423039\n","For Fold:  8  Accuracy:  0.578034683459067\n","For Fold:  9  Accuracy:  0.5755813953488372\n","For Fold:  10  Accuracy:  0.5465116265208222\n","Time Stamp:  31\n","For Fold:  1  Accuracy:  0.5317919092371285\n","For Fold:  2  Accuracy:  0.6416184991770397\n","For Fold:  3  Accuracy:  0.5664739889561097\n","For Fold:  4  Accuracy:  0.543352602189676\n","For Fold:  5  Accuracy:  0.5838150299353406\n","For Fold:  6  Accuracy:  0.5838150306244116\n","For Fold:  7  Accuracy:  0.6184971112047317\n","For Fold:  8  Accuracy:  0.6184971112047317\n","For Fold:  9  Accuracy:  0.6279069760511088\n","For Fold:  10  Accuracy:  0.5348837209302325\n","\n","For time stamp:  0  Mean Accuracy:  0.4884124218163624\n","For time stamp:  1  Mean Accuracy:  0.5138728330864559\n","For time stamp:  2  Mean Accuracy:  0.4664403822619054\n","For time stamp:  3  Mean Accuracy:  0.4947640820374202\n","For time stamp:  4  Mean Accuracy:  0.5040227189858382\n","For time stamp:  5  Mean Accuracy:  0.5063819071521104\n","For time stamp:  6  Mean Accuracy:  0.53124748030879\n","For time stamp:  7  Mean Accuracy:  0.5254536907696759\n","For time stamp:  8  Mean Accuracy:  0.5335898649173327\n","For time stamp:  9  Mean Accuracy:  0.589753327584704\n","For time stamp:  10  Mean Accuracy:  0.6336873249783921\n","For time stamp:  11  Mean Accuracy:  0.5908623480888157\n","For time stamp:  12  Mean Accuracy:  0.6093997859517355\n","For time stamp:  13  Mean Accuracy:  0.6076455178454625\n","For time stamp:  14  Mean Accuracy:  0.598376799024374\n","For time stamp:  15  Mean Accuracy:  0.5729163873174014\n","For time stamp:  16  Mean Accuracy:  0.5972341727599099\n","For time stamp:  17  Mean Accuracy:  0.6093729007030178\n","For time stamp:  18  Mean Accuracy:  0.6070742049526024\n","For time stamp:  19  Mean Accuracy:  0.6001209845701074\n","For time stamp:  20  Mean Accuracy:  0.5625050420517689\n","For time stamp:  21  Mean Accuracy:  0.5491732770984957\n","For time stamp:  22  Mean Accuracy:  0.5665445635433015\n","For time stamp:  23  Mean Accuracy:  0.5757998396276561\n","For time stamp:  24  Mean Accuracy:  0.5503461494515028\n","For time stamp:  25  Mean Accuracy:  0.5862716771619167\n","For time stamp:  26  Mean Accuracy:  0.6168940729043837\n","For time stamp:  27  Mean Accuracy:  0.6151868540700012\n","For time stamp:  28  Mean Accuracy:  0.5868093839807366\n","For time stamp:  29  Mean Accuracy:  0.5954563793543607\n","For time stamp:  30  Mean Accuracy:  0.5665445631466849\n","For time stamp:  31  Mean Accuracy:  0.585065197951051\n"],"name":"stdout"},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAY4AAAEGCAYAAABy53LJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]},{"cell_type":"markdown","metadata":{"id":"K23Z8ByOPTxe","colab_type":"text"},"source":["# Multi Classification"]},{"cell_type":"code","metadata":{"id":"lO1WSKjeOW4v","colab_type":"code","outputId":"4e60328a-9d78-4616-acd5-9f99b3cf05ad","executionInfo":{"status":"ok","timestamp":1574230385684,"user_tz":-330,"elapsed":709021,"user":{"displayName":"201601422 Daiict","photoUrl":"","userId":"06333073350363496926"}},"colab":{"base_uri":"https://localhost:8080/","height":1000}},"source":["mean_accuracy = []\n","for i in range(0,32):\n","\n","  print(\"Time Stamp: \",i)\n","\n","  principal_components = do_pca(data[i],0.95)\n","  #print(principal_components.shape)\n","  #print(y.shape)\n","\n","  multi_data = principal_components\n","  multi_labels = y\n","\n","  oh = OneHotEncoder()\n","  multi_labels = oh.fit_transform(multi_labels)\n","\n","  #print(multi_data.shape)\n","  #print(multi_labels.shape)\n","\n","  accuracy = []\n","\n","  cv10 = KFold(n_splits=10, random_state=42, shuffle=True)\n","  k = 0\n","  for train_index, test_index in cv10.split(multi_data):\n","    k =k+1\n","    #print(\"Train Index: \", train_index, \"\\n\")\n","    #print(\"Test Index: \", test_index)\n","      \n","    X_train, X_test, y_train, y_test = multi_data[train_index], multi_data[test_index], multi_labels[train_index], multi_labels[test_index]\n","    model = Sequential([\n","            Dense(32, input_dim = X_train.shape[1], activation = 'relu'),\n","            Dense(32, activation = 'relu'),\n","            Dense(6, activation = 'softmax')\n","        ])\n","    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n","    history = model.fit(X_train, y_train, epochs=20, batch_size=256, validation_split=0.1, shuffle=True, verbose = 0)\n","    test_loss, test_acc = model.evaluate(X_test, y_test, verbose=0)\n","    accuracy.append(test_acc)\n","    print(\"For Fold: \",k, \" Accuracy: \",accuracy[k-1])\n","  mean_accuracy.append(np.mean(accuracy))\n","      \n","\n","print()\n","for i in range(0,32):\n","  print(\"For time stamp: \",i,\" Mean Accuracy: \",mean_accuracy[i])\n","\n","plt.figure()\n","plt.plot(np.arange(0,32,1), mean_accuracy)\n","plt.grid()\n","plt.xlabel('Time')\n","plt.ylabel('Accuracy')\n","plt.show()"],"execution_count":4,"outputs":[{"output_type":"stream","text":["Time Stamp:  0\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:66: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:541: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:4432: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/optimizers.py:793: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3576: The name tf.log is deprecated. Please use tf.math.log instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/math_grad.py:1424: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n","Instructions for updating:\n","Use tf.where in 2.0, which has the same broadcast rule as np.where\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1033: The name tf.assign_add is deprecated. Please use tf.compat.v1.assign_add instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:1020: The name tf.assign is deprecated. Please use tf.compat.v1.assign instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3005: The name tf.Session is deprecated. Please use tf.compat.v1.Session instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:190: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:197: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:207: The name tf.global_variables is deprecated. Please use tf.compat.v1.global_variables instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:216: The name tf.is_variable_initialized is deprecated. Please use tf.compat.v1.is_variable_initialized instead.\n","\n","WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:223: The name tf.variables_initializer is deprecated. Please use tf.compat.v1.variables_initializer instead.\n","\n","For Fold:  1  Accuracy:  0.15799614652158667\n","For Fold:  2  Accuracy:  0.15028901751330823\n","For Fold:  3  Accuracy:  0.14836223523970513\n","For Fold:  4  Accuracy:  0.17726396934375138\n","For Fold:  5  Accuracy:  0.18689788053949905\n","For Fold:  6  Accuracy:  0.15992292879518977\n","For Fold:  7  Accuracy:  0.1445086706063651\n","For Fold:  8  Accuracy:  0.1368015417703544\n","For Fold:  9  Accuracy:  0.1621621622196956\n","For Fold:  10  Accuracy:  0.18339768345521684\n","Time Stamp:  1\n","For Fold:  1  Accuracy:  0.1579961464354528\n","For Fold:  2  Accuracy:  0.18689788053949905\n","For Fold:  3  Accuracy:  0.17919075161735445\n","For Fold:  4  Accuracy:  0.1522157997869113\n","For Fold:  5  Accuracy:  0.1541425820605144\n","For Fold:  6  Accuracy:  0.18111753389095755\n","For Fold:  7  Accuracy:  0.18111753380482368\n","For Fold:  8  Accuracy:  0.17533718698401443\n","For Fold:  9  Accuracy:  0.1621621622196956\n","For Fold:  10  Accuracy:  0.14671814671814673\n","Time Stamp:  2\n","For Fold:  1  Accuracy:  0.1734104047965452\n","For Fold:  2  Accuracy:  0.1541425819456692\n","For Fold:  3  Accuracy:  0.1560693642479836\n","For Fold:  4  Accuracy:  0.18111753380482368\n","For Fold:  5  Accuracy:  0.142581888332762\n","For Fold:  6  Accuracy:  0.18304431607842675\n","For Fold:  7  Accuracy:  0.15799614660772054\n","For Fold:  8  Accuracy:  0.13680154142581888\n","For Fold:  9  Accuracy:  0.13513513513513514\n","For Fold:  10  Accuracy:  0.13706563706563707\n","Time Stamp:  3\n","For Fold:  1  Accuracy:  0.1657032755872877\n","For Fold:  2  Accuracy:  0.15414258188824662\n","For Fold:  3  Accuracy:  0.1849710984381637\n","For Fold:  4  Accuracy:  0.1984585743533853\n","For Fold:  5  Accuracy:  0.16184971098265896\n","For Fold:  6  Accuracy:  0.15799614660772054\n","For Fold:  7  Accuracy:  0.19653179190751446\n","For Fold:  8  Accuracy:  0.14643545296610205\n","For Fold:  9  Accuracy:  0.15057915080928433\n","For Fold:  10  Accuracy:  0.14864864870618208\n","Time Stamp:  4\n","For Fold:  1  Accuracy:  0.17341040471041133\n","For Fold:  2  Accuracy:  0.1406551061452928\n","For Fold:  3  Accuracy:  0.17341040471041133\n","For Fold:  4  Accuracy:  0.1657032757021329\n","For Fold:  5  Accuracy:  0.2003853565408545\n","For Fold:  6  Accuracy:  0.17148362235067438\n","For Fold:  7  Accuracy:  0.17919075153122058\n","For Fold:  8  Accuracy:  0.17919075161735445\n","For Fold:  9  Accuracy:  0.15250965256718596\n","For Fold:  10  Accuracy:  0.1467181468332136\n","Time Stamp:  5\n","For Fold:  1  Accuracy:  0.1791907515025093\n","For Fold:  2  Accuracy:  0.14836223523970513\n","For Fold:  3  Accuracy:  0.15799614660772054\n","For Fold:  4  Accuracy:  0.12138728323699421\n","For Fold:  5  Accuracy:  0.13680154159808663\n","For Fold:  6  Accuracy:  0.19075144543124073\n","For Fold:  7  Accuracy:  0.15028901751330823\n","For Fold:  8  Accuracy:  0.13294797696474658\n","For Fold:  9  Accuracy:  0.18146718152471492\n","For Fold:  10  Accuracy:  0.15057915057915058\n","Time Stamp:  6\n","For Fold:  1  Accuracy:  0.18111753380482368\n","For Fold:  2  Accuracy:  0.1637764934285298\n","For Fold:  3  Accuracy:  0.19653179190751446\n","For Fold:  4  Accuracy:  0.18304431607842675\n","For Fold:  5  Accuracy:  0.17919075153122058\n","For Fold:  6  Accuracy:  0.16955684007707128\n","For Fold:  7  Accuracy:  0.19075144525897297\n","For Fold:  8  Accuracy:  0.1637764933423959\n","For Fold:  9  Accuracy:  0.18146718152471492\n","For Fold:  10  Accuracy:  0.13899613905367245\n","Time Stamp:  7\n","For Fold:  1  Accuracy:  0.16955684016320516\n","For Fold:  2  Accuracy:  0.211946050182473\n","For Fold:  3  Accuracy:  0.1599229287090559\n","For Fold:  4  Accuracy:  0.1637764934285298\n","For Fold:  5  Accuracy:  0.17533718698401443\n","For Fold:  6  Accuracy:  0.16763005788960209\n","For Fold:  7  Accuracy:  0.17533718698401443\n","For Fold:  8  Accuracy:  0.14836223523970513\n","For Fold:  9  Accuracy:  0.18918918924672262\n","For Fold:  10  Accuracy:  0.16023166028919367\n","Time Stamp:  8\n","For Fold:  1  Accuracy:  0.2080924857214007\n","For Fold:  2  Accuracy:  0.23506743746571\n","For Fold:  3  Accuracy:  0.16955684016320516\n","For Fold:  4  Accuracy:  0.16955684016320516\n","For Fold:  5  Accuracy:  0.16570327552986513\n","For Fold:  6  Accuracy:  0.21965317927688532\n","For Fold:  7  Accuracy:  0.17726396934375138\n","For Fold:  8  Accuracy:  0.19267822744644217\n","For Fold:  9  Accuracy:  0.2297297299598635\n","For Fold:  10  Accuracy:  0.20656370662124002\n","Time Stamp:  9\n","For Fold:  1  Accuracy:  0.22350674373795762\n","For Fold:  2  Accuracy:  0.21194605026860688\n","For Fold:  3  Accuracy:  0.22157996146435452\n","For Fold:  4  Accuracy:  0.22350674379538019\n","For Fold:  5  Accuracy:  0.2543352602878747\n","For Fold:  6  Accuracy:  0.22543352609769457\n","For Fold:  7  Accuracy:  0.22350674373795762\n","For Fold:  8  Accuracy:  0.19460500972004524\n","For Fold:  9  Accuracy:  0.19111969117722455\n","For Fold:  10  Accuracy:  0.24710424716178053\n","Time Stamp:  10\n","For Fold:  1  Accuracy:  0.21579961472967915\n","For Fold:  2  Accuracy:  0.25048169574066853\n","For Fold:  3  Accuracy:  0.2312138730046377\n","For Fold:  4  Accuracy:  0.2562620425614778\n","For Fold:  5  Accuracy:  0.24084778428651923\n","For Fold:  6  Accuracy:  0.23121387288979253\n","For Fold:  7  Accuracy:  0.23506743755184387\n","For Fold:  8  Accuracy:  0.23699421973931306\n","For Fold:  9  Accuracy:  0.24903474903474904\n","For Fold:  10  Accuracy:  0.22972972984479662\n","Time Stamp:  11\n","For Fold:  1  Accuracy:  0.22928709064490074\n","For Fold:  2  Accuracy:  0.2408477843726531\n","For Fold:  3  Accuracy:  0.2350674373795761\n","For Fold:  4  Accuracy:  0.21579961481581303\n","For Fold:  5  Accuracy:  0.23892100201291616\n","For Fold:  6  Accuracy:  0.2427745665601223\n","For Fold:  7  Accuracy:  0.28323699439192096\n","For Fold:  8  Accuracy:  0.20616570327552985\n","For Fold:  9  Accuracy:  0.21621621621621623\n","For Fold:  10  Accuracy:  0.21428571434324772\n","Time Stamp:  12\n","For Fold:  1  Accuracy:  0.25626204238921\n","For Fold:  2  Accuracy:  0.2427745665601223\n","For Fold:  3  Accuracy:  0.23699421973931306\n","For Fold:  4  Accuracy:  0.20423892108806066\n","For Fold:  5  Accuracy:  0.23699421973931306\n","For Fold:  6  Accuracy:  0.2697495183905655\n","For Fold:  7  Accuracy:  0.27167630080772504\n","For Fold:  8  Accuracy:  0.23892100209905004\n","For Fold:  9  Accuracy:  0.2355212356363024\n","For Fold:  10  Accuracy:  0.2490347491498159\n","Time Stamp:  13\n","For Fold:  1  Accuracy:  0.23699421973931306\n","For Fold:  2  Accuracy:  0.2658959538433593\n","For Fold:  3  Accuracy:  0.23699421971060178\n","For Fold:  4  Accuracy:  0.22928709061618946\n","For Fold:  5  Accuracy:  0.2601156071086839\n","For Fold:  6  Accuracy:  0.23506743755184387\n","For Fold:  7  Accuracy:  0.22928709061618946\n","For Fold:  8  Accuracy:  0.23506743746571\n","For Fold:  9  Accuracy:  0.23359073364826702\n","For Fold:  10  Accuracy:  0.19884169884169883\n","Time Stamp:  14\n","For Fold:  1  Accuracy:  0.22350674391022538\n","For Fold:  2  Accuracy:  0.23506743746571\n","For Fold:  3  Accuracy:  0.2177263971468387\n","For Fold:  4  Accuracy:  0.21001926782273603\n","For Fold:  5  Accuracy:  0.22928709073103462\n","For Fold:  6  Accuracy:  0.21387283254220996\n","For Fold:  7  Accuracy:  0.20616570344779764\n","For Fold:  8  Accuracy:  0.20231213878574628\n","For Fold:  9  Accuracy:  0.23166023177529854\n","For Fold:  10  Accuracy:  0.24131274154287508\n","Time Stamp:  15\n","For Fold:  1  Accuracy:  0.21772639708941613\n","For Fold:  2  Accuracy:  0.26204238926744183\n","For Fold:  3  Accuracy:  0.20423892108806066\n","For Fold:  4  Accuracy:  0.22350674373795762\n","For Fold:  5  Accuracy:  0.20231213890059146\n","For Fold:  6  Accuracy:  0.22350674391022538\n","For Fold:  7  Accuracy:  0.21194605044087464\n","For Fold:  8  Accuracy:  0.21579961470096787\n","For Fold:  9  Accuracy:  0.21814671826178503\n","For Fold:  10  Accuracy:  0.19111969111969113\n","Time Stamp:  16\n","For Fold:  1  Accuracy:  0.2100192679088699\n","For Fold:  2  Accuracy:  0.21579961481581303\n","For Fold:  3  Accuracy:  0.262042389382287\n","For Fold:  4  Accuracy:  0.2254335260115607\n","For Fold:  5  Accuracy:  0.23699421999771472\n","For Fold:  6  Accuracy:  0.24084778425780795\n","For Fold:  7  Accuracy:  0.21772639726168388\n","For Fold:  8  Accuracy:  0.2447013488050141\n","For Fold:  9  Accuracy:  0.2046332047482715\n","For Fold:  10  Accuracy:  0.21042471048224387\n","Time Stamp:  17\n","For Fold:  1  Accuracy:  0.2215799615217771\n","For Fold:  2  Accuracy:  0.2408477843726531\n","For Fold:  3  Accuracy:  0.23699421982544694\n","For Fold:  4  Accuracy:  0.2350674374369987\n","For Fold:  5  Accuracy:  0.19267822753257605\n","For Fold:  6  Accuracy:  0.24855491346706546\n","For Fold:  7  Accuracy:  0.211946050182473\n","For Fold:  8  Accuracy:  0.22928709073103462\n","For Fold:  9  Accuracy:  0.19498069509576185\n","For Fold:  10  Accuracy:  0.22007722013475353\n","Time Stamp:  18\n","For Fold:  1  Accuracy:  0.2003853564547206\n","For Fold:  2  Accuracy:  0.2658959538433593\n","For Fold:  3  Accuracy:  0.23121387283236994\n","For Fold:  4  Accuracy:  0.21965317927688532\n","For Fold:  5  Accuracy:  0.2331406551633956\n","For Fold:  6  Accuracy:  0.20423892117419454\n","For Fold:  7  Accuracy:  0.25048169562582334\n","For Fold:  8  Accuracy:  0.23121387291850382\n","For Fold:  9  Accuracy:  0.2683397683973018\n","For Fold:  10  Accuracy:  0.2162162163312831\n","Time Stamp:  19\n","For Fold:  1  Accuracy:  0.23314065519210692\n","For Fold:  2  Accuracy:  0.22736030845743155\n","For Fold:  3  Accuracy:  0.24470134891985929\n","For Fold:  4  Accuracy:  0.23506743772411162\n","For Fold:  5  Accuracy:  0.23699421973931306\n","For Fold:  6  Accuracy:  0.21579961481581303\n","For Fold:  7  Accuracy:  0.24662813119346236\n","For Fold:  8  Accuracy:  0.21772639697457094\n","For Fold:  9  Accuracy:  0.20077220088726766\n","For Fold:  10  Accuracy:  0.2239382240532908\n","Time Stamp:  20\n","For Fold:  1  Accuracy:  0.23506743746571\n","For Fold:  2  Accuracy:  0.25048169565453465\n","For Fold:  3  Accuracy:  0.23699421982544694\n","For Fold:  4  Accuracy:  0.22928709061618946\n","For Fold:  5  Accuracy:  0.20809248563526683\n","For Fold:  6  Accuracy:  0.1926782273603083\n","For Fold:  7  Accuracy:  0.21387283245607608\n","For Fold:  8  Accuracy:  0.25240847801427163\n","For Fold:  9  Accuracy:  0.20656370656370657\n","For Fold:  10  Accuracy:  0.22779922785676124\n","Time Stamp:  21\n","For Fold:  1  Accuracy:  0.21965317927688532\n","For Fold:  2  Accuracy:  0.20423892108806066\n","For Fold:  3  Accuracy:  0.21579961481581303\n","For Fold:  4  Accuracy:  0.19075144525897297\n","For Fold:  5  Accuracy:  0.19460500963391136\n","For Fold:  6  Accuracy:  0.17919075153122058\n","For Fold:  7  Accuracy:  0.1849710984381637\n","For Fold:  8  Accuracy:  0.2061657036200654\n","For Fold:  9  Accuracy:  0.21814671826178503\n","For Fold:  10  Accuracy:  0.1969111969111969\n","Time Stamp:  22\n","For Fold:  1  Accuracy:  0.21001926799500378\n","For Fold:  2  Accuracy:  0.17919075161735445\n","For Fold:  3  Accuracy:  0.20809248560655552\n","For Fold:  4  Accuracy:  0.1772639692576175\n","For Fold:  5  Accuracy:  0.2003853564547206\n","For Fold:  6  Accuracy:  0.19460500980617912\n","For Fold:  7  Accuracy:  0.20038535662698836\n","For Fold:  8  Accuracy:  0.19845857452565305\n","For Fold:  9  Accuracy:  0.17953667953667954\n","For Fold:  10  Accuracy:  0.21621621621621623\n","Time Stamp:  23\n","For Fold:  1  Accuracy:  0.22543352618382845\n","For Fold:  2  Accuracy:  0.2215799615217771\n","For Fold:  3  Accuracy:  0.21001926782273603\n","For Fold:  4  Accuracy:  0.20423892117419454\n","For Fold:  5  Accuracy:  0.2023121388144576\n","For Fold:  6  Accuracy:  0.16763005797573596\n","For Fold:  7  Accuracy:  0.19460500972004524\n","For Fold:  8  Accuracy:  0.18882466298536987\n","For Fold:  9  Accuracy:  0.1969111969111969\n","For Fold:  10  Accuracy:  0.21814671814671815\n","Time Stamp:  24\n","For Fold:  1  Accuracy:  0.18497109835202982\n","For Fold:  2  Accuracy:  0.18497109835202982\n","For Fold:  3  Accuracy:  0.20809248554913296\n","For Fold:  4  Accuracy:  0.16570327552986513\n","For Fold:  5  Accuracy:  0.1849710984381637\n","For Fold:  6  Accuracy:  0.19460500963391136\n","For Fold:  7  Accuracy:  0.21194605026860688\n","For Fold:  8  Accuracy:  0.22736030828516376\n","For Fold:  9  Accuracy:  0.18532818532818532\n","For Fold:  10  Accuracy:  0.22393822399575739\n","Time Stamp:  25\n","For Fold:  1  Accuracy:  0.20423892108806066\n","For Fold:  2  Accuracy:  0.211946050182473\n","For Fold:  3  Accuracy:  0.2023121388144576\n","For Fold:  4  Accuracy:  0.211946050182473\n","For Fold:  5  Accuracy:  0.19267822744644217\n","For Fold:  6  Accuracy:  0.22928709055876687\n","For Fold:  7  Accuracy:  0.18111753380482368\n","For Fold:  8  Accuracy:  0.1868978805969216\n","For Fold:  9  Accuracy:  0.1969111970262638\n","For Fold:  10  Accuracy:  0.20270270276023614\n","Time Stamp:  26\n","For Fold:  1  Accuracy:  0.18882466281310212\n","For Fold:  2  Accuracy:  0.22157996163662227\n","For Fold:  3  Accuracy:  0.21965317927688532\n","For Fold:  4  Accuracy:  0.2331406551633956\n","For Fold:  5  Accuracy:  0.211946050182473\n","For Fold:  6  Accuracy:  0.1907514450867052\n","For Fold:  7  Accuracy:  0.18304431616456063\n","For Fold:  8  Accuracy:  0.19653179207978222\n","For Fold:  9  Accuracy:  0.19498069503822843\n","For Fold:  10  Accuracy:  0.19111969111969113\n","Time Stamp:  27\n","For Fold:  1  Accuracy:  0.18304431607842675\n","For Fold:  2  Accuracy:  0.22543352618382845\n","For Fold:  3  Accuracy:  0.17533718695530312\n","For Fold:  4  Accuracy:  0.2350674373795761\n","For Fold:  5  Accuracy:  0.19460500980617912\n","For Fold:  6  Accuracy:  0.1907514451728391\n","For Fold:  7  Accuracy:  0.20809248563526683\n","For Fold:  8  Accuracy:  0.2080924857214007\n","For Fold:  9  Accuracy:  0.22007722019228695\n","For Fold:  10  Accuracy:  0.19691119696873036\n","Time Stamp:  28\n","For Fold:  1  Accuracy:  0.20423892108806066\n","For Fold:  2  Accuracy:  0.20423892100192678\n","For Fold:  3  Accuracy:  0.20616570336166373\n","For Fold:  4  Accuracy:  0.18689788053949905\n","For Fold:  5  Accuracy:  0.1984585743533853\n","For Fold:  6  Accuracy:  0.17341040471041133\n","For Fold:  7  Accuracy:  0.20809248563526683\n","For Fold:  8  Accuracy:  0.20616570336166373\n","For Fold:  9  Accuracy:  0.20270270270270271\n","For Fold:  10  Accuracy:  0.20463320469073806\n","Time Stamp:  29\n","For Fold:  1  Accuracy:  0.19460500963391136\n","For Fold:  2  Accuracy:  0.211946050182473\n","For Fold:  3  Accuracy:  0.2138728323699422\n","For Fold:  4  Accuracy:  0.1907514451728391\n","For Fold:  5  Accuracy:  0.18689788062563292\n","For Fold:  6  Accuracy:  0.19653179196493703\n","For Fold:  7  Accuracy:  0.19460500969133396\n","For Fold:  8  Accuracy:  0.2100192679088699\n","For Fold:  9  Accuracy:  0.1872586873162207\n","For Fold:  10  Accuracy:  0.17567567590580943\n","Time Stamp:  30\n","For Fold:  1  Accuracy:  0.1772639692576175\n","For Fold:  2  Accuracy:  0.20616570333295245\n","For Fold:  3  Accuracy:  0.19267822753257605\n","For Fold:  4  Accuracy:  0.1984585743533853\n","For Fold:  5  Accuracy:  0.1907514450867052\n","For Fold:  6  Accuracy:  0.19267822744644217\n","For Fold:  7  Accuracy:  0.19653179199364834\n","For Fold:  8  Accuracy:  0.17726396917148363\n","For Fold:  9  Accuracy:  0.20077220088726766\n","For Fold:  10  Accuracy:  0.17760617766371103\n","Time Stamp:  31\n","For Fold:  1  Accuracy:  0.1868978805969216\n","For Fold:  2  Accuracy:  0.188824662899236\n","For Fold:  3  Accuracy:  0.2100192679088699\n","For Fold:  4  Accuracy:  0.20423892100192678\n","For Fold:  5  Accuracy:  0.19653179207978222\n","For Fold:  6  Accuracy:  0.17533718698401443\n","For Fold:  7  Accuracy:  0.20423892117419454\n","For Fold:  8  Accuracy:  0.1984585743533853\n","For Fold:  9  Accuracy:  0.22007722013475353\n","For Fold:  10  Accuracy:  0.20077220077220076\n","\n","For time stamp:  0  Mean Accuracy:  0.1607602236004672\n","For time stamp:  1  Mean Accuracy:  0.167689572405737\n","For time stamp:  2  Mean Accuracy:  0.1557364549440522\n","For time stamp:  3  Mean Accuracy:  0.16653164322465458\n","For time stamp:  4  Mean Accuracy:  0.16826574727087515\n","For time stamp:  5  Mean Accuracy:  0.15497727301981767\n","For time stamp:  6  Mean Accuracy:  0.17482089860073427\n","For time stamp:  7  Mean Accuracy:  0.1721289829116516\n","For time stamp:  8  Mean Accuracy:  0.19738656916915687\n","For time stamp:  9  Mean Accuracy:  0.22166439774488764\n","For time stamp:  10  Mean Accuracy:  0.23766450193834773\n","For time stamp:  11  Mean Accuracy:  0.23226021240128966\n","For time stamp:  12  Mean Accuracy:  0.24421667755994783\n","For time stamp:  13  Mean Accuracy:  0.23611414891418567\n","For time stamp:  14  Mean Accuracy:  0.22109305851704725\n","For time stamp:  15  Mean Accuracy:  0.2170345408517012\n","For time stamp:  16  Mean Accuracy:  0.22686224636712668\n","For time stamp:  17  Mean Accuracy:  0.22320136003005403\n","For time stamp:  18  Mean Accuracy:  0.23207794920178376\n","For time stamp:  19  Mean Accuracy:  0.22821285379572273\n","For time stamp:  20  Mean Accuracy:  0.22532454014482717\n","For time stamp:  21  Mean Accuracy:  0.2010433638836075\n","For time stamp:  22  Mean Accuracy:  0.1964153667642968\n","For time stamp:  23  Mean Accuracy:  0.202970146125606\n","For time stamp:  24  Mean Accuracy:  0.19718868337328463\n","For time stamp:  25  Mean Accuracy:  0.20200377924609186\n","For time stamp:  26  Mean Accuracy:  0.2031572448561446\n","For time stamp:  27  Mean Accuracy:  0.20374123100938374\n","For time stamp:  28  Mean Accuracy:  0.1995004501445318\n","For time stamp:  29  Mean Accuracy:  0.19621636507719695\n","For time stamp:  30  Mean Accuracy:  0.19101702867257891\n","For time stamp:  31  Mean Accuracy:  0.19853966279052854\n"],"name":"stdout"},{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAYgAAAEGCAYAAAB/+QKOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npQlCKaWUU/8PEWhC0B+sMHwAAAAASUVORK5CYII=\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]}]}