predictML.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "J3OtwrYMh5do"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import keras\n",
    "from keras.datasets import imdb\n",
    "from keras.models import Sequential\n",
    "from keras.layers import LSTM\n",
    "from keras.layers.embeddings import Embedding\n",
    "from keras.preprocessing import sequence\n",
    "from sklearn.metrics import confusion_matrix, precision_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "from keras.layers import Dense,Dropout\n",
    "from keras.regularizers import l2\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 286,
     "status": "ok",
     "timestamp": 1582609074721,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "9Hwc3b1ukDel",
    "outputId": "032c0a82-4990-43c1-d069-3d8f4f3a6288"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((880, 5), (880, 6), (845, 5), (845, 6))"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "datafile = \"splicedDataML/mix6ml\"\n",
    "\n",
    "x_data = pd.read_csv(datafile+\"X.csv\",names=['s1','s2','s3','s4','s5'])\n",
    "y_dataR = np.genfromtxt(datafile+\"Y.csv\",delimiter=',',dtype=int)\n",
    "num_classes = np.max(y_dataR) + 1\n",
    "y_data = keras.utils.to_categorical(y_dataR, num_classes)\n",
    "#x_train, x_test, y_train, y_test = train_test_split(x_data,y_data, test_size=0.20, random_state=0)\n",
    "x_train = x_data[:880]\n",
    "x_test = x_data[880:]\n",
    "y_train = y_data[:880]\n",
    "y_test = y_data[880:]\n",
    "x_train.shape,y_train.shape,x_test.shape,y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 20285,
     "status": "ok",
     "timestamp": 1582609098680,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "E1rwuze5mL10",
    "outputId": "c91ea7eb-f2f2-4dec-f4b7-749e50ad3fa3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/150\n",
      "44/44 [==============================] - 0s 4ms/step - loss: 21.4839 - accuracy: 0.3114 - val_loss: 6.2611 - val_accuracy: 0.3124\n",
      "Epoch 2/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 10.0744 - accuracy: 0.3284 - val_loss: 2.4095 - val_accuracy: 0.4225\n",
      "Epoch 3/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 4.8510 - accuracy: 0.3591 - val_loss: 1.1741 - val_accuracy: 0.5763\n",
      "Epoch 4/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 3.0015 - accuracy: 0.3318 - val_loss: 1.0365 - val_accuracy: 0.3988\n",
      "Epoch 5/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 1.8245 - accuracy: 0.3182 - val_loss: 1.0663 - val_accuracy: 0.3195\n",
      "Epoch 6/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 1.3390 - accuracy: 0.3875 - val_loss: 1.0038 - val_accuracy: 0.3148\n",
      "Epoch 7/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 1.1841 - accuracy: 0.3886 - val_loss: 0.9553 - val_accuracy: 0.3124\n",
      "Epoch 8/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 1.0862 - accuracy: 0.4466 - val_loss: 0.9520 - val_accuracy: 0.3124\n",
      "Epoch 9/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 1.0334 - accuracy: 0.4307 - val_loss: 0.9248 - val_accuracy: 0.3136\n",
      "Epoch 10/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.9853 - accuracy: 0.4511 - val_loss: 0.9019 - val_accuracy: 0.3882\n",
      "Epoch 11/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.9754 - accuracy: 0.4466 - val_loss: 0.9109 - val_accuracy: 0.3420\n",
      "Epoch 12/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.9370 - accuracy: 0.4545 - val_loss: 0.8995 - val_accuracy: 0.3669\n",
      "Epoch 13/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.9166 - accuracy: 0.4773 - val_loss: 0.8761 - val_accuracy: 0.3846\n",
      "Epoch 14/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.9240 - accuracy: 0.4795 - val_loss: 0.8494 - val_accuracy: 0.4059\n",
      "Epoch 15/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.8868 - accuracy: 0.4739 - val_loss: 0.8400 - val_accuracy: 0.4107\n",
      "Epoch 16/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.8719 - accuracy: 0.4989 - val_loss: 0.8221 - val_accuracy: 0.4568\n",
      "Epoch 17/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.8788 - accuracy: 0.4841 - val_loss: 0.8094 - val_accuracy: 0.4568\n",
      "Epoch 18/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.8505 - accuracy: 0.4989 - val_loss: 0.8021 - val_accuracy: 0.4568\n",
      "Epoch 19/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.8512 - accuracy: 0.4886 - val_loss: 0.7943 - val_accuracy: 0.4544\n",
      "Epoch 20/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.8271 - accuracy: 0.4989 - val_loss: 0.7858 - val_accuracy: 0.4568\n",
      "Epoch 21/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.8168 - accuracy: 0.5114 - val_loss: 0.7764 - val_accuracy: 0.4568\n",
      "Epoch 22/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.8005 - accuracy: 0.5136 - val_loss: 0.7664 - val_accuracy: 0.4568\n",
      "Epoch 23/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.7986 - accuracy: 0.5045 - val_loss: 0.7650 - val_accuracy: 0.4391\n",
      "Epoch 24/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.7864 - accuracy: 0.5023 - val_loss: 0.7472 - val_accuracy: 0.4568\n",
      "Epoch 25/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.7785 - accuracy: 0.5148 - val_loss: 0.7367 - val_accuracy: 0.4568\n",
      "Epoch 26/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.7774 - accuracy: 0.5125 - val_loss: 0.7315 - val_accuracy: 0.4568\n",
      "Epoch 27/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.7549 - accuracy: 0.5182 - val_loss: 0.7179 - val_accuracy: 0.4568\n",
      "Epoch 28/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.7343 - accuracy: 0.5136 - val_loss: 0.7099 - val_accuracy: 0.4568\n",
      "Epoch 29/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.7312 - accuracy: 0.5182 - val_loss: 0.7032 - val_accuracy: 0.4568\n",
      "Epoch 30/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.7181 - accuracy: 0.5159 - val_loss: 0.6929 - val_accuracy: 0.4568\n",
      "Epoch 31/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.7013 - accuracy: 0.5330 - val_loss: 0.6847 - val_accuracy: 0.4568\n",
      "Epoch 32/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.6913 - accuracy: 0.5341 - val_loss: 0.6769 - val_accuracy: 0.4568\n",
      "Epoch 33/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.6816 - accuracy: 0.5409 - val_loss: 0.6737 - val_accuracy: 0.4568\n",
      "Epoch 34/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.6689 - accuracy: 0.5534 - val_loss: 0.6591 - val_accuracy: 0.4568\n",
      "Epoch 35/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.6658 - accuracy: 0.5682 - val_loss: 0.6525 - val_accuracy: 0.4568\n",
      "Epoch 36/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.6460 - accuracy: 0.5557 - val_loss: 0.6386 - val_accuracy: 0.4828\n",
      "Epoch 37/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.6334 - accuracy: 0.5693 - val_loss: 0.6181 - val_accuracy: 0.5787\n",
      "Epoch 38/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.6213 - accuracy: 0.6000 - val_loss: 0.6037 - val_accuracy: 0.6000\n",
      "Epoch 39/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.6110 - accuracy: 0.6125 - val_loss: 0.5849 - val_accuracy: 0.6757\n",
      "Epoch 40/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5948 - accuracy: 0.6239 - val_loss: 0.5655 - val_accuracy: 0.6757\n",
      "Epoch 41/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5788 - accuracy: 0.6409 - val_loss: 0.5531 - val_accuracy: 0.6757\n",
      "Epoch 42/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5605 - accuracy: 0.6682 - val_loss: 0.5416 - val_accuracy: 0.6757\n",
      "Epoch 43/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5574 - accuracy: 0.6670 - val_loss: 0.5427 - val_accuracy: 0.6757\n",
      "Epoch 44/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5430 - accuracy: 0.6500 - val_loss: 0.5271 - val_accuracy: 0.6757\n",
      "Epoch 45/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5284 - accuracy: 0.6682 - val_loss: 0.5144 - val_accuracy: 0.6757\n",
      "Epoch 46/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5178 - accuracy: 0.6761 - val_loss: 0.5118 - val_accuracy: 0.6757\n",
      "Epoch 47/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5223 - accuracy: 0.6591 - val_loss: 0.5036 - val_accuracy: 0.6757\n",
      "Epoch 48/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.5063 - accuracy: 0.6841 - val_loss: 0.4886 - val_accuracy: 0.6757\n",
      "Epoch 49/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4918 - accuracy: 0.6864 - val_loss: 0.4816 - val_accuracy: 0.6757\n",
      "Epoch 50/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4842 - accuracy: 0.6977 - val_loss: 0.4726 - val_accuracy: 0.6781\n",
      "Epoch 51/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4752 - accuracy: 0.7045 - val_loss: 0.4664 - val_accuracy: 0.6757\n",
      "Epoch 52/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4644 - accuracy: 0.7045 - val_loss: 0.4687 - val_accuracy: 0.6935\n",
      "Epoch 53/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4510 - accuracy: 0.7193 - val_loss: 0.4495 - val_accuracy: 0.7456\n",
      "Epoch 54/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4513 - accuracy: 0.7250 - val_loss: 0.4428 - val_accuracy: 0.8083\n",
      "Epoch 55/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4354 - accuracy: 0.7398 - val_loss: 0.4315 - val_accuracy: 0.7775\n",
      "Epoch 56/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4345 - accuracy: 0.7330 - val_loss: 0.4651 - val_accuracy: 0.6876\n",
      "Epoch 57/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4294 - accuracy: 0.7511 - val_loss: 0.4179 - val_accuracy: 0.7917\n",
      "Epoch 58/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4172 - accuracy: 0.7489 - val_loss: 0.4085 - val_accuracy: 0.7953\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 59/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4115 - accuracy: 0.7682 - val_loss: 0.4018 - val_accuracy: 0.8201\n",
      "Epoch 60/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.4053 - accuracy: 0.7773 - val_loss: 0.3936 - val_accuracy: 0.7988\n",
      "Epoch 61/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3937 - accuracy: 0.7795 - val_loss: 0.3856 - val_accuracy: 0.8568\n",
      "Epoch 62/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3909 - accuracy: 0.7830 - val_loss: 0.3849 - val_accuracy: 0.8414\n",
      "Epoch 63/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3871 - accuracy: 0.7852 - val_loss: 0.3756 - val_accuracy: 0.8249\n",
      "Epoch 64/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3752 - accuracy: 0.7636 - val_loss: 0.3779 - val_accuracy: 0.8639\n",
      "Epoch 65/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3765 - accuracy: 0.7739 - val_loss: 0.3763 - val_accuracy: 0.8651\n",
      "Epoch 66/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3779 - accuracy: 0.7636 - val_loss: 0.3742 - val_accuracy: 0.8426\n",
      "Epoch 67/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3683 - accuracy: 0.7580 - val_loss: 0.3689 - val_accuracy: 0.7361\n",
      "Epoch 68/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3640 - accuracy: 0.7727 - val_loss: 0.3534 - val_accuracy: 0.8710\n",
      "Epoch 69/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3468 - accuracy: 0.7955 - val_loss: 0.3521 - val_accuracy: 0.8521\n",
      "Epoch 70/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3367 - accuracy: 0.8068 - val_loss: 0.3430 - val_accuracy: 0.8343\n",
      "Epoch 71/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3350 - accuracy: 0.8080 - val_loss: 0.3350 - val_accuracy: 0.8544\n",
      "Epoch 72/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3397 - accuracy: 0.8000 - val_loss: 0.3435 - val_accuracy: 0.7893\n",
      "Epoch 73/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3249 - accuracy: 0.8364 - val_loss: 0.3255 - val_accuracy: 0.8473\n",
      "Epoch 74/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3132 - accuracy: 0.8375 - val_loss: 0.3159 - val_accuracy: 0.8651\n",
      "Epoch 75/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3029 - accuracy: 0.8568 - val_loss: 0.3287 - val_accuracy: 0.8533\n",
      "Epoch 76/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.3101 - accuracy: 0.8364 - val_loss: 0.3208 - val_accuracy: 0.8533\n",
      "Epoch 77/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2941 - accuracy: 0.8489 - val_loss: 0.2928 - val_accuracy: 0.8663\n",
      "Epoch 78/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2929 - accuracy: 0.8557 - val_loss: 0.3024 - val_accuracy: 0.8675\n",
      "Epoch 79/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2790 - accuracy: 0.8705 - val_loss: 0.2871 - val_accuracy: 0.8651\n",
      "Epoch 80/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2747 - accuracy: 0.8750 - val_loss: 0.3007 - val_accuracy: 0.8639\n",
      "Epoch 81/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2806 - accuracy: 0.8557 - val_loss: 0.2889 - val_accuracy: 0.8533\n",
      "Epoch 82/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2678 - accuracy: 0.8648 - val_loss: 0.2808 - val_accuracy: 0.8438\n",
      "Epoch 83/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2732 - accuracy: 0.8648 - val_loss: 0.2812 - val_accuracy: 0.8675\n",
      "Epoch 84/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2646 - accuracy: 0.8580 - val_loss: 0.2722 - val_accuracy: 0.8627\n",
      "Epoch 85/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2483 - accuracy: 0.8739 - val_loss: 0.2711 - val_accuracy: 0.8675\n",
      "Epoch 86/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2454 - accuracy: 0.8818 - val_loss: 0.2686 - val_accuracy: 0.8580\n",
      "Epoch 87/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2470 - accuracy: 0.8636 - val_loss: 0.2616 - val_accuracy: 0.8710\n",
      "Epoch 88/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2407 - accuracy: 0.8716 - val_loss: 0.2659 - val_accuracy: 0.8639\n",
      "Epoch 89/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2339 - accuracy: 0.8761 - val_loss: 0.2570 - val_accuracy: 0.8675\n",
      "Epoch 90/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2397 - accuracy: 0.8807 - val_loss: 0.2672 - val_accuracy: 0.8686\n",
      "Epoch 91/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2352 - accuracy: 0.8818 - val_loss: 0.2541 - val_accuracy: 0.8473\n",
      "Epoch 92/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2295 - accuracy: 0.8795 - val_loss: 0.2525 - val_accuracy: 0.8604\n",
      "Epoch 93/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2234 - accuracy: 0.8795 - val_loss: 0.2573 - val_accuracy: 0.8639\n",
      "Epoch 94/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2256 - accuracy: 0.8693 - val_loss: 0.2553 - val_accuracy: 0.8568\n",
      "Epoch 95/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2274 - accuracy: 0.8773 - val_loss: 0.2445 - val_accuracy: 0.8651\n",
      "Epoch 96/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2231 - accuracy: 0.8761 - val_loss: 0.2361 - val_accuracy: 0.8639\n",
      "Epoch 97/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2123 - accuracy: 0.8852 - val_loss: 0.2562 - val_accuracy: 0.8107\n",
      "Epoch 98/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2123 - accuracy: 0.8807 - val_loss: 0.2435 - val_accuracy: 0.8686\n",
      "Epoch 99/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2202 - accuracy: 0.8693 - val_loss: 0.2303 - val_accuracy: 0.8663\n",
      "Epoch 100/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1985 - accuracy: 0.8955 - val_loss: 0.2349 - val_accuracy: 0.8639\n",
      "Epoch 101/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2126 - accuracy: 0.8716 - val_loss: 0.2615 - val_accuracy: 0.8580\n",
      "Epoch 102/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2112 - accuracy: 0.8693 - val_loss: 0.2256 - val_accuracy: 0.8675\n",
      "Epoch 103/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1941 - accuracy: 0.8955 - val_loss: 0.2209 - val_accuracy: 0.8734\n",
      "Epoch 104/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2077 - accuracy: 0.8659 - val_loss: 0.2174 - val_accuracy: 0.8615\n",
      "Epoch 105/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.2036 - accuracy: 0.8784 - val_loss: 0.2197 - val_accuracy: 0.8722\n",
      "Epoch 106/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1959 - accuracy: 0.8909 - val_loss: 0.2221 - val_accuracy: 0.8734\n",
      "Epoch 107/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1998 - accuracy: 0.8705 - val_loss: 0.2143 - val_accuracy: 0.8675\n",
      "Epoch 108/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1960 - accuracy: 0.8784 - val_loss: 0.2258 - val_accuracy: 0.8722\n",
      "Epoch 109/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1879 - accuracy: 0.8875 - val_loss: 0.2112 - val_accuracy: 0.8675\n",
      "Epoch 110/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1904 - accuracy: 0.8841 - val_loss: 0.2129 - val_accuracy: 0.8663\n",
      "Epoch 111/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1862 - accuracy: 0.8898 - val_loss: 0.2026 - val_accuracy: 0.8615\n",
      "Epoch 112/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1778 - accuracy: 0.8898 - val_loss: 0.2202 - val_accuracy: 0.8580\n",
      "Epoch 113/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1846 - accuracy: 0.8795 - val_loss: 0.2117 - val_accuracy: 0.8734\n",
      "Epoch 114/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1940 - accuracy: 0.8773 - val_loss: 0.2197 - val_accuracy: 0.8533\n",
      "Epoch 115/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.1885 - accuracy: 0.8614 - val_loss: 0.2082 - val_accuracy: 0.8663\n",
      "Epoch 116/150\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1731 - accuracy: 0.8943 - val_loss: 0.2094 - val_accuracy: 0.8663\n",
      "Epoch 117/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1738 - accuracy: 0.8898 - val_loss: 0.2145 - val_accuracy: 0.8781\n",
      "Epoch 118/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1731 - accuracy: 0.8852 - val_loss: 0.2056 - val_accuracy: 0.8592\n",
      "Epoch 119/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1817 - accuracy: 0.8795 - val_loss: 0.2055 - val_accuracy: 0.8651\n",
      "Epoch 120/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1871 - accuracy: 0.8716 - val_loss: 0.2085 - val_accuracy: 0.8675\n",
      "Epoch 121/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1844 - accuracy: 0.8716 - val_loss: 0.2024 - val_accuracy: 0.8651\n",
      "Epoch 122/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1719 - accuracy: 0.8807 - val_loss: 0.2055 - val_accuracy: 0.8485\n",
      "Epoch 123/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1707 - accuracy: 0.8784 - val_loss: 0.1931 - val_accuracy: 0.8734\n",
      "Epoch 124/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1661 - accuracy: 0.8841 - val_loss: 0.2230 - val_accuracy: 0.8130\n",
      "Epoch 125/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1851 - accuracy: 0.8648 - val_loss: 0.2001 - val_accuracy: 0.8627\n",
      "Epoch 126/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1715 - accuracy: 0.8864 - val_loss: 0.2165 - val_accuracy: 0.8284\n",
      "Epoch 127/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1641 - accuracy: 0.8875 - val_loss: 0.1987 - val_accuracy: 0.8627\n",
      "Epoch 128/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1609 - accuracy: 0.8875 - val_loss: 0.2042 - val_accuracy: 0.8521\n",
      "Epoch 129/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1762 - accuracy: 0.8648 - val_loss: 0.2095 - val_accuracy: 0.8686\n",
      "Epoch 130/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1655 - accuracy: 0.8705 - val_loss: 0.1881 - val_accuracy: 0.8675\n",
      "Epoch 131/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1611 - accuracy: 0.8932 - val_loss: 0.1835 - val_accuracy: 0.8663\n",
      "Epoch 132/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1586 - accuracy: 0.8864 - val_loss: 0.1807 - val_accuracy: 0.8710\n",
      "Epoch 133/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.1574 - accuracy: 0.8909 - val_loss: 0.1822 - val_accuracy: 0.8746\n",
      "Epoch 134/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.1632 - accuracy: 0.8761 - val_loss: 0.1793 - val_accuracy: 0.8639\n",
      "Epoch 135/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1588 - accuracy: 0.8898 - val_loss: 0.2620 - val_accuracy: 0.7574\n",
      "Epoch 136/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1714 - accuracy: 0.8773 - val_loss: 0.1881 - val_accuracy: 0.8580\n",
      "Epoch 137/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1582 - accuracy: 0.8818 - val_loss: 0.1810 - val_accuracy: 0.8604\n",
      "Epoch 138/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.1604 - accuracy: 0.8739 - val_loss: 0.2090 - val_accuracy: 0.8710\n",
      "Epoch 139/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.1633 - accuracy: 0.8693 - val_loss: 0.1954 - val_accuracy: 0.8675\n",
      "Epoch 140/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.1538 - accuracy: 0.8773 - val_loss: 0.1824 - val_accuracy: 0.8497\n",
      "Epoch 141/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1559 - accuracy: 0.8852 - val_loss: 0.1778 - val_accuracy: 0.8580\n",
      "Epoch 142/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1543 - accuracy: 0.8852 - val_loss: 0.1900 - val_accuracy: 0.8686\n",
      "Epoch 143/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1509 - accuracy: 0.8977 - val_loss: 0.1806 - val_accuracy: 0.8698\n",
      "Epoch 144/150\n",
      "44/44 [==============================] - 0s 2ms/step - loss: 0.1563 - accuracy: 0.8830 - val_loss: 0.1921 - val_accuracy: 0.8663\n",
      "Epoch 145/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1613 - accuracy: 0.8670 - val_loss: 0.1957 - val_accuracy: 0.8544\n",
      "Epoch 146/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1604 - accuracy: 0.8761 - val_loss: 0.1834 - val_accuracy: 0.8663\n",
      "Epoch 147/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1584 - accuracy: 0.8739 - val_loss: 0.1757 - val_accuracy: 0.8556\n",
      "Epoch 148/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1484 - accuracy: 0.8864 - val_loss: 0.1919 - val_accuracy: 0.8710\n",
      "Epoch 149/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1502 - accuracy: 0.8909 - val_loss: 0.1791 - val_accuracy: 0.8651\n",
      "Epoch 150/150\n",
      "44/44 [==============================] - 0s 1ms/step - loss: 0.1492 - accuracy: 0.8898 - val_loss: 0.1931 - val_accuracy: 0.8284\n",
      "WARNING:tensorflow:From /home/sanjay/.local/lib/python3.7/site-packages/tensorflow/python/training/tracking/tracking.py:111: Model.state_updates (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n",
      "WARNING:tensorflow:From /home/sanjay/.local/lib/python3.7/site-packages/tensorflow/python/training/tracking/tracking.py:111: Layer.updates (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "This property should not be used in TensorFlow 2.0, as updates are applied automatically.\n",
      "INFO:tensorflow:Assets written to: modelSave/assets\n"
     ]
    }
   ],
   "source": [
    "#defifne a sequentail Model\n",
    "model = Sequential()\n",
    "#Hidden Layer-1\n",
    "model.add(Dense(100,activation='relu',input_dim=5,kernel_regularizer=l2(0.01)))\n",
    "model.add(Dropout(0.3, noise_shape=None, seed=None))\n",
    "\n",
    "#Hidden Layer-2\n",
    "model.add(Dense(100,activation = 'relu',kernel_regularizer=l2(0.01)))\n",
    "model.add(Dropout(0.3, noise_shape=None, seed=None))\n",
    "\n",
    "#Output layer\n",
    "model.add(Dense(6,activation='sigmoid'))\n",
    "model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])\n",
    "model_output = model.fit(x_train,y_train,epochs=150,batch_size=20,verbose=1,validation_data=(x_test,y_test),)\n",
    "model.save('modelSave')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 329
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 288,
     "status": "ok",
     "timestamp": 1582605199619,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "TkbYdVzzClzz",
    "outputId": "f93d86c4-348b-45e4-e356-982599e76a31"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense (Dense)                (None, 100)               600       \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 100)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 100)               10100     \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 100)               0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 6)                 606       \n",
      "=================================================================\n",
      "Total params: 11,306\n",
      "Trainable params: 11,306\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = keras.models.load_model('modelSave')\n",
    "model.summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 139
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 480,
     "status": "ok",
     "timestamp": 1582609102373,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "c3yjoJHWnDmN",
    "outputId": "86c43d5a-8bf1-489e-9649-b57c0eb1fae6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Accuracy :  0.7411287860075633\n",
      "Validation Accuracy :  0.727873765428861\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[262,   0,   0,   2,   0],\n",
       "       [  0, 122,   0,   0,   0],\n",
       "       [  0,   0, 185,   0,   0],\n",
       "       [  0,   0,  34, 130,   0],\n",
       "       [ 49,  21,  20,  19,   1]])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('Training Accuracy : ' , np.mean(model_output.history[\"accuracy\"]))\n",
    "print('Validation Accuracy : ' , np.mean(model_output.history[\"val_accuracy\"]))\n",
    "y_pred = model.predict(x_test)\n",
    "rounded = [np.argmax(x) for x in y_pred]\n",
    "with open('predict.txt', 'w') as f:\n",
    "    i=0\n",
    "#     print(x_data)\n",
    "#     print(y_test)\n",
    "#     print(rounded)\n",
    "    for item in rounded:\n",
    "        f.write(str(item)+\"\\n\")\n",
    "        i=i+1\n",
    "y_testComp = [np.argmax(x) for x in y_test]\n",
    "confusion_matrix(y_testComp,rounded)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 573
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 661,
     "status": "ok",
     "timestamp": 1582611027779,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "tmU9xIVEnHPq",
    "outputId": "f0f908fb-a98b-4d2d-902d-da480677aa6c"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot training & validation accuracy values\n",
    "plt.plot(model_output.history['accuracy'])\n",
    "plt.plot(model_output.history['val_accuracy'])\n",
    "plt.title('Model Accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Test'], loc='upper left')\n",
    "plt.show()\n",
    "\n",
    "# Plot training & validation loss values\n",
    "plt.plot(model_output.history['loss'])\n",
    "plt.plot(model_output.history['val_loss'])\n",
    "plt.title('Model Loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 89
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 347,
     "status": "ok",
     "timestamp": 1582511495326,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "Ndrk72zLnQnH",
    "outputId": "fe874f09-50ab-4d21-df66-c19a76a11cce"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.84244373, 0.85314685, 0.77405858, 0.86092715, 1.        ])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision_score(y_testComp,rounded,average=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 104
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 382,
     "status": "ok",
     "timestamp": 1582612311267,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "WuEZmqdhoIUd",
    "outputId": "c18986b1-a020-4576-e364-ff18504ed115"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[655,   0,   0,   2,   0],\n",
       "       [  0, 214,   0,   0,   0],\n",
       "       [  0,   0, 349,   0,   0],\n",
       "       [  0,   0,  34, 274,   0],\n",
       "       [ 73,  29,  52,  37,   6]])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testfile = \"splicedDataML/mix6ml\"\n",
    "new_dataX = pd.read_csv(testfile+\"X.csv\",names=['s1','s2','s3','s4','s5'])\n",
    "new_dataYR = np.genfromtxt(testfile+\"Y.csv\",delimiter=',',dtype=int)\n",
    "new_dataY = keras.utils.to_categorical(new_dataYR, num_classes)\n",
    "new_dataX.shape,new_dataY.shape\n",
    "new_predYT = model.predict(new_dataX)\n",
    "new_predY = [np.argmax(x) for x in new_predYT]\n",
    "new_predYComp = [np.argmax(x) for x in new_dataY]\n",
    "confusion_matrix(new_predYComp,new_predY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 367,
     "status": "ok",
     "timestamp": 1582519272293,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "aZxPQnds0FSL",
    "outputId": "9bd60e32-3c10-4a86-dff0-44f98ecd6511"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1000, 5), (1000, 6), (725, 5), (725, 6), (1725, 5), (1725, 6))"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "datafile = \"splicedDataML/mix6ml\"\n",
    "\n",
    "test = np.genfromtxt(datafile+\"X.csv\",delimiter=',')\n",
    "y_dataR = np.genfromtxt(datafile+\"Y.csv\",delimiter=',',dtype=int)\n",
    "num_classes = np.max(y_dataR) + 1\n",
    "y_data = keras.utils.to_categorical(y_dataR, num_classes)\n",
    "x_trainT = test[:1000]\n",
    "x_testT = test[1000:]\n",
    "y_trainT = y_data[:1000]\n",
    "y_testT = y_data[1000:]\n",
    "x_trainT.shape,y_trainT.shape,x_testT.shape,y_testT.shape, test.shape, y_data.shape\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 362,
     "status": "ok",
     "timestamp": 1582519012191,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "yW9dj2wbXw1W",
    "outputId": "bb1625e9-2ec3-400a-991c-131e4336b63a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1725, 5)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 330,
     "status": "ok",
     "timestamp": 1582519294514,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "Ao3NANmYV116",
    "outputId": "eb62765f-0296-423a-d73e-ff4a7e90688c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1000, 1, 5), (1000, 1, 6), (725, 1, 5), (725, 1, 6), (1725, 5), (1725, 6))"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_data = test.reshape(test.shape[0],1,5)\n",
    "x_train = x_trainT.reshape(x_trainT.shape[0],1,5)\n",
    "x_test = x_testT.reshape(x_testT.shape[0],1,5)\n",
    "y_train = y_trainT.reshape(y_trainT.shape[0],1,6)\n",
    "y_test = y_testT.reshape(y_testT.shape[0],1,6)\n",
    "x_train.shape,y_train.shape,x_test.shape,y_test.shape, test.shape, y_data.shape\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 354,
     "status": "ok",
     "timestamp": 1582518945726,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "uAXxVZNiXFzN",
    "outputId": "93fe49ab-279c-40ac-9f18-30e07e70bcfa"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1725, 1, 5)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 360,
     "status": "ok",
     "timestamp": 1582517538337,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "trVHvxr4YNK4",
    "outputId": "3f461ebf-a805-41de-d5a2-b500df6fbd63"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15\n"
     ]
    }
   ],
   "source": [
    "samples = list()\n",
    "length = 115\n",
    "for i in range(0,x_data.shape[0],length):\n",
    "\t# grab from i to i + 200\n",
    "\tsample = x_data[i:i+length]\n",
    "\tsamples.append(sample)\n",
    "print(len(samples))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 8845,
     "status": "ok",
     "timestamp": 1582519336534,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "czYaTT7WVeSF",
    "outputId": "c181643e-188b-493f-b803-87f8c9c91aa2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "14/14 - 0s - loss: 0.5421 - accuracy: 0.4420 - val_loss: 0.5161 - val_accuracy: 0.3641\n",
      "Epoch 2/50\n",
      "14/14 - 0s - loss: 0.4857 - accuracy: 0.3930 - val_loss: 0.4856 - val_accuracy: 0.3641\n",
      "Epoch 3/50\n",
      "14/14 - 0s - loss: 0.4540 - accuracy: 0.3930 - val_loss: 0.4652 - val_accuracy: 0.3641\n",
      "Epoch 4/50\n",
      "14/14 - 0s - loss: 0.4341 - accuracy: 0.3930 - val_loss: 0.4502 - val_accuracy: 0.3641\n",
      "Epoch 5/50\n",
      "14/14 - 0s - loss: 0.4187 - accuracy: 0.3930 - val_loss: 0.4333 - val_accuracy: 0.3641\n",
      "Epoch 6/50\n",
      "14/14 - 0s - loss: 0.4035 - accuracy: 0.3930 - val_loss: 0.4296 - val_accuracy: 0.3641\n",
      "Epoch 7/50\n",
      "14/14 - 0s - loss: 0.3978 - accuracy: 0.3930 - val_loss: 0.4255 - val_accuracy: 0.3641\n",
      "Epoch 8/50\n",
      "14/14 - 0s - loss: 0.3926 - accuracy: 0.3930 - val_loss: 0.4214 - val_accuracy: 0.3641\n",
      "Epoch 9/50\n",
      "14/14 - 0s - loss: 0.3880 - accuracy: 0.3930 - val_loss: 0.4177 - val_accuracy: 0.3641\n",
      "Epoch 10/50\n",
      "14/14 - 0s - loss: 0.3860 - accuracy: 0.3930 - val_loss: 0.4186 - val_accuracy: 0.3641\n",
      "Epoch 11/50\n",
      "14/14 - 0s - loss: 0.3851 - accuracy: 0.3930 - val_loss: 0.4160 - val_accuracy: 0.3641\n",
      "Epoch 12/50\n",
      "14/14 - 0s - loss: 0.3816 - accuracy: 0.3930 - val_loss: 0.4142 - val_accuracy: 0.3641\n",
      "Epoch 13/50\n",
      "14/14 - 0s - loss: 0.3794 - accuracy: 0.3930 - val_loss: 0.4129 - val_accuracy: 0.3641\n",
      "Epoch 14/50\n",
      "14/14 - 0s - loss: 0.3771 - accuracy: 0.3930 - val_loss: 0.4117 - val_accuracy: 0.3641\n",
      "Epoch 15/50\n",
      "14/14 - 0s - loss: 0.3757 - accuracy: 0.3930 - val_loss: 0.4106 - val_accuracy: 0.3641\n",
      "Epoch 16/50\n",
      "14/14 - 0s - loss: 0.3735 - accuracy: 0.3930 - val_loss: 0.4101 - val_accuracy: 0.3641\n",
      "Epoch 17/50\n",
      "14/14 - 0s - loss: 0.3717 - accuracy: 0.4500 - val_loss: 0.4090 - val_accuracy: 0.3641\n",
      "Epoch 18/50\n",
      "14/14 - 0s - loss: 0.3704 - accuracy: 0.5670 - val_loss: 0.4082 - val_accuracy: 0.3779\n",
      "Epoch 19/50\n",
      "14/14 - 0s - loss: 0.3844 - accuracy: 0.4850 - val_loss: 0.4062 - val_accuracy: 0.3641\n",
      "Epoch 20/50\n",
      "14/14 - 0s - loss: 0.3831 - accuracy: 0.3930 - val_loss: 0.4077 - val_accuracy: 0.3641\n",
      "Epoch 21/50\n",
      "14/14 - 0s - loss: 0.3803 - accuracy: 0.3930 - val_loss: 0.4053 - val_accuracy: 0.3641\n",
      "Epoch 22/50\n",
      "14/14 - 0s - loss: 0.3786 - accuracy: 0.3930 - val_loss: 0.4065 - val_accuracy: 0.3641\n",
      "Epoch 23/50\n",
      "14/14 - 0s - loss: 0.3792 - accuracy: 0.3930 - val_loss: 0.4083 - val_accuracy: 0.3641\n",
      "Epoch 24/50\n",
      "14/14 - 0s - loss: 0.3774 - accuracy: 0.3930 - val_loss: 0.4074 - val_accuracy: 0.3641\n",
      "Epoch 25/50\n",
      "14/14 - 0s - loss: 0.3749 - accuracy: 0.3930 - val_loss: 0.4062 - val_accuracy: 0.3641\n",
      "Epoch 26/50\n",
      "14/14 - 0s - loss: 0.3757 - accuracy: 0.3930 - val_loss: 0.4081 - val_accuracy: 0.3641\n",
      "Epoch 27/50\n",
      "14/14 - 0s - loss: 0.3742 - accuracy: 0.3930 - val_loss: 0.4073 - val_accuracy: 0.3641\n",
      "Epoch 28/50\n",
      "14/14 - 0s - loss: 0.3737 - accuracy: 0.3930 - val_loss: 0.4066 - val_accuracy: 0.3641\n",
      "Epoch 29/50\n",
      "14/14 - 0s - loss: 0.3723 - accuracy: 0.3930 - val_loss: 0.4069 - val_accuracy: 0.3641\n",
      "Epoch 30/50\n",
      "14/14 - 0s - loss: 0.3712 - accuracy: 0.3930 - val_loss: 0.4054 - val_accuracy: 0.3641\n",
      "Epoch 31/50\n",
      "14/14 - 0s - loss: 0.3690 - accuracy: 0.3930 - val_loss: 0.4037 - val_accuracy: 0.3641\n",
      "Epoch 32/50\n",
      "14/14 - 0s - loss: 0.3694 - accuracy: 0.3930 - val_loss: 0.4064 - val_accuracy: 0.3641\n",
      "Epoch 33/50\n",
      "14/14 - 0s - loss: 0.3687 - accuracy: 0.3930 - val_loss: 0.4061 - val_accuracy: 0.3683\n",
      "Epoch 34/50\n",
      "14/14 - 0s - loss: 0.3655 - accuracy: 0.5970 - val_loss: 0.4056 - val_accuracy: 0.3779\n",
      "Epoch 35/50\n",
      "14/14 - 0s - loss: 0.3652 - accuracy: 0.5970 - val_loss: 0.4069 - val_accuracy: 0.3779\n",
      "Epoch 36/50\n",
      "14/14 - 0s - loss: 0.3669 - accuracy: 0.5970 - val_loss: 0.4062 - val_accuracy: 0.3779\n",
      "Epoch 37/50\n",
      "14/14 - 0s - loss: 0.3659 - accuracy: 0.5970 - val_loss: 0.4049 - val_accuracy: 0.3779\n",
      "Epoch 38/50\n",
      "14/14 - 0s - loss: 0.3645 - accuracy: 0.5970 - val_loss: 0.4029 - val_accuracy: 0.3779\n",
      "Epoch 39/50\n",
      "14/14 - 0s - loss: 0.3643 - accuracy: 0.5970 - val_loss: 0.4047 - val_accuracy: 0.3779\n",
      "Epoch 40/50\n",
      "14/14 - 0s - loss: 0.3634 - accuracy: 0.5970 - val_loss: 0.4042 - val_accuracy: 0.3779\n",
      "Epoch 41/50\n",
      "14/14 - 0s - loss: 0.3626 - accuracy: 0.5970 - val_loss: 0.4039 - val_accuracy: 0.3779\n",
      "Epoch 42/50\n",
      "14/14 - 0s - loss: 0.3618 - accuracy: 0.5970 - val_loss: 0.4035 - val_accuracy: 0.3779\n",
      "Epoch 43/50\n",
      "14/14 - 0s - loss: 0.3610 - accuracy: 0.5970 - val_loss: 0.4029 - val_accuracy: 0.3779\n",
      "Epoch 44/50\n",
      "14/14 - 0s - loss: 0.3597 - accuracy: 0.5970 - val_loss: 0.4028 - val_accuracy: 0.3779\n",
      "Epoch 45/50\n",
      "14/14 - 0s - loss: 0.3593 - accuracy: 0.5970 - val_loss: 0.4015 - val_accuracy: 0.3779\n",
      "Epoch 46/50\n",
      "14/14 - 0s - loss: 0.3581 - accuracy: 0.5970 - val_loss: 0.3988 - val_accuracy: 0.3779\n",
      "Epoch 47/50\n",
      "14/14 - 0s - loss: 0.3578 - accuracy: 0.5970 - val_loss: 0.4015 - val_accuracy: 0.3779\n",
      "Epoch 48/50\n",
      "14/14 - 0s - loss: 0.3571 - accuracy: 0.5970 - val_loss: 0.3971 - val_accuracy: 0.3779\n",
      "Epoch 49/50\n",
      "14/14 - 0s - loss: 0.3548 - accuracy: 0.5970 - val_loss: 0.3972 - val_accuracy: 0.3779\n",
      "Epoch 50/50\n",
      "14/14 - 0s - loss: 0.3545 - accuracy: 0.5970 - val_loss: 0.3965 - val_accuracy: 0.3779\n"
     ]
    }
   ],
   "source": [
    "#Hidden Layer-1\n",
    "model = Sequential()\n",
    "model.add(LSTM(128, input_shape=(x_train.shape[1], x_train.shape[2])))\n",
    "model.add(Dense(6, activation='sigmoid'))\n",
    "model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])\n",
    "model_output = model.fit(x_train, y_trainT, epochs=50, batch_size=72, validation_data=(x_test, y_testT), verbose=2, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "9uagCKKrbdJ9"
   },
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Input 0 of layer lstm_1 is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: [None, 6]",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-15-0daa300e1909>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mLSTM\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m32\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput_shape\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mDense\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m6\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mactivation\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'sigmoid'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'binary_crossentropy'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0moptimizer\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'adam'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mmetrics\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'accuracy'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mmodel_output\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_data\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m50\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m72\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mshuffle\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/tensorflow/python/training/tracking/base.py\u001b[0m in \u001b[0;36m_method_wrapper\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    457\u001b[0m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_self_setattr_tracking\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m  \u001b[0;31m# pylint: disable=protected-access\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    458\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 459\u001b[0;31m       \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    460\u001b[0m     \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    461\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_self_setattr_tracking\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mprevious_value\u001b[0m  \u001b[0;31m# pylint: disable=protected-access\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/tensorflow/python/keras/engine/sequential.py\u001b[0m in \u001b[0;36madd\u001b[0;34m(self, layer)\u001b[0m\n\u001b[1;32m    219\u001b[0m       \u001b[0;31m# If the model is being built continuously on top of an input layer:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    220\u001b[0m       \u001b[0;31m# refresh its output.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 221\u001b[0;31m       \u001b[0moutput_tensor\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlayer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    222\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnest\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflatten\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput_tensor\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    223\u001b[0m         \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mSINGLE_LAYER_OUTPUT_ERROR_MSG\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/tensorflow/python/keras/layers/recurrent.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs, initial_state, constants, **kwargs)\u001b[0m\n\u001b[1;32m    662\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    663\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0minitial_state\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mconstants\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 664\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0msuper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mRNN\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    665\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    666\u001b[0m     \u001b[0;31m# If any of `initial_state` or `constants` are specified and are Keras\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    925\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0m_in_functional_construction_mode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    926\u001b[0m       return self._functional_construction_call(inputs, args, kwargs,\n\u001b[0;32m--> 927\u001b[0;31m                                                 input_list)\n\u001b[0m\u001b[1;32m    928\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    929\u001b[0m     \u001b[0;31m# Maintains info about the `Layer.call` stack.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer.py\u001b[0m in \u001b[0;36m_functional_construction_call\u001b[0;34m(self, inputs, args, kwargs, input_list)\u001b[0m\n\u001b[1;32m   1091\u001b[0m       \u001b[0;31m# TODO(reedwm): We should assert input compatibility after the inputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1092\u001b[0m       \u001b[0;31m# are casted, not before.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1093\u001b[0;31m       \u001b[0minput_spec\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0massert_input_compatibility\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minput_spec\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1094\u001b[0m       \u001b[0mgraph\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mbackend\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_graph\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1095\u001b[0m       \u001b[0;31m# Use `self._name_scope()` to avoid auto-incrementing the name.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.local/lib/python3.7/site-packages/tensorflow/python/keras/engine/input_spec.py\u001b[0m in \u001b[0;36massert_input_compatibility\u001b[0;34m(input_spec, inputs, layer_name)\u001b[0m\n\u001b[1;32m    185\u001b[0m                          \u001b[0;34m'expected ndim='\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mspec\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndim\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m', found ndim='\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    186\u001b[0m                          \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mndim\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m'. Full shape received: '\u001b[0m \u001b[0;34m+\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 187\u001b[0;31m                          str(x.shape.as_list()))\n\u001b[0m\u001b[1;32m    188\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mspec\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmax_ndim\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    189\u001b[0m       \u001b[0mndim\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndims\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: Input 0 of layer lstm_1 is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: [None, 6]"
     ]
    }
   ],
   "source": [
    "model.add(LSTM(32, input_shape=(x_data.shape[1], x_data.shape[2])))\n",
    "model.add(Dense(6, activation='sigmoid'))\n",
    "model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])\n",
    "model_output = model.fit(x_data, y_data, epochs=50, batch_size=72, verbose=2, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 573
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 763,
     "status": "ok",
     "timestamp": 1582519517797,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "Mwdk0D6jd6mY",
    "outputId": "baea7c90-a9b0-46cb-b1a9-0bb1573f2c14"
   },
   "outputs": [],
   "source": [
    "plt.plot(model_output.history['acc'])\n",
    "plt.plot(model_output.history['val_acc'])\n",
    "plt.title('Model accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Test'], loc='upper left')\n",
    "plt.show()\n",
    "# Plot training & validation loss values\n",
    "plt.plot(model_output.history['loss'])\n",
    "plt.plot(model_output.history['val_loss'])\n",
    "plt.title('model_output loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "nxbaK1iEhMnp"
   },
   "outputs": [],
   "source": [
    "new_predYT= model.predict(x_train)\n",
    "new_predY = [argmax(x) for x in new_predYT]\n",
    "new_predYComp = [argmax(x) for x in y_train]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 955,
     "status": "ok",
     "timestamp": 1582519869232,
     "user": {
      "displayName": "Srinivasan Seshan",
      "photoUrl": "",
      "userId": "06632033567786526417"
     },
     "user_tz": 300
    },
    "id": "IlT8aCjlhrTX",
    "outputId": "4fc8c3ed-99dc-47f0-f2a0-c75828b34dd3"
   },
   "outputs": [],
   "source": [
    "new_predY"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "6XUYBerthJlZ"
   },
   "outputs": [],
   "source": [
    "testfile = \"splicedDataML/mix6ml\"\n",
    "new_dataX = pd.read_csv(testfile+\"X.csv\",names=['s1','s2','s3','s4','s5'])\n",
    "new_dataYR = np.genfromtxt(testfile+\"Y.csv\",delimiter=',',dtype=int)\n",
    "new_dataY = keras.utils.to_categorical(new_dataYR, num_classes)\n",
    "new_dataX.shape,new_dataY.shape\n",
    "new_predYT = model.predict(new_dataX)\n",
    "new_predY = [argmax(x) for x in new_predYT]\n",
    "new_predYComp = [argmax(x) for x in new_dataY]\n",
    "confusion_matrix(new_predYComp,new_predY)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyMkzJA381+M5MURbjWhO/5N",
   "mount_file_id": "1EvNBzEgtFC1-c6IvCxWJOt5f2WmYxCdp",
   "name": "sanjay 2020.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}