Pima Indians- Basic models.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using pandas "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import StratifiedKFold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "seed=7\n",
    "x=np.random.seed(seed)\n",
    "df=pd.read_csv('pima-indians-diabetes.csv',header=None)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X=df.iloc[:,0:8]\n",
    "Y=df.iloc[:,8]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "nn=Sequential()\n",
    "nn.add(Dense(12,input_dim=8,init='uniform',activation='relu'))\n",
    "nn.add(Dense(8,init='uniform',activation='relu'))\n",
    "nn.add(Dense(1,init='uniform',activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "nn.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy']) #getting error here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1e74656a390>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nn.fit(X,Y, validation_split=0.33,nb_epoch=150,batch_size=10,verbose=0) #figure the error soon!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using the workflow of 'deep learning in python'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "import numpy\n",
    "# fix random seed for reproducibility\n",
    "seed = 7\n",
    "numpy.random.seed(seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# load pima indians dataset\n",
    "dataset = numpy.loadtxt(\"pima-indians-diabetes.csv\", delimiter=\",\")\n",
    "# split into input (X) and output (Y) variables\n",
    "X = dataset[:,0:8]\n",
    "Y = dataset[:,8]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(12, input_dim=8, init='uniform', activation='relu'))\n",
    "model.add(Dense(8, init='uniform', activation='relu'))\n",
    "model.add(Dense(1, init='uniform', activation='sigmoid'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 514 samples, validate on 254 samples\n",
      "Epoch 1/150\n",
      "514/514 [==============================] - 0s - loss: 0.6901 - acc: 0.5681 - val_loss: 0.6814 - val_acc: 0.6732\n",
      "Epoch 2/150\n",
      "514/514 [==============================] - 0s - loss: 0.6773 - acc: 0.6401 - val_loss: 0.6620 - val_acc: 0.6732\n",
      "Epoch 3/150\n",
      "514/514 [==============================] - 0s - loss: 0.6661 - acc: 0.6401 - val_loss: 0.6519 - val_acc: 0.6732\n",
      "Epoch 4/150\n",
      "514/514 [==============================] - 0s - loss: 0.6590 - acc: 0.6401 - val_loss: 0.6513 - val_acc: 0.6732\n",
      "Epoch 5/150\n",
      "514/514 [==============================] - 0s - loss: 0.6568 - acc: 0.6401 - val_loss: 0.6461 - val_acc: 0.6811\n",
      "Epoch 6/150\n",
      "514/514 [==============================] - ETA: 0s - loss: 0.6480 - acc: 0.630 - 0s - loss: 0.6454 - acc: 0.6401 - val_loss: 0.6393 - val_acc: 0.6772\n",
      "Epoch 7/150\n",
      "514/514 [==============================] - 0s - loss: 0.6343 - acc: 0.6556 - val_loss: 0.6513 - val_acc: 0.6417\n",
      "Epoch 8/150\n",
      "514/514 [==============================] - 0s - loss: 0.6380 - acc: 0.6634 - val_loss: 0.6304 - val_acc: 0.6378\n",
      "Epoch 9/150\n",
      "514/514 [==============================] - 0s - loss: 0.6250 - acc: 0.6712 - val_loss: 0.6212 - val_acc: 0.6614\n",
      "Epoch 10/150\n",
      "514/514 [==============================] - 0s - loss: 0.6223 - acc: 0.6634 - val_loss: 0.6153 - val_acc: 0.6693\n",
      "Epoch 11/150\n",
      "514/514 [==============================] - 0s - loss: 0.6097 - acc: 0.6770 - val_loss: 0.6094 - val_acc: 0.6457\n",
      "Epoch 12/150\n",
      "514/514 [==============================] - 0s - loss: 0.6041 - acc: 0.6751 - val_loss: 0.6078 - val_acc: 0.6417\n",
      "Epoch 13/150\n",
      "514/514 [==============================] - 0s - loss: 0.6048 - acc: 0.6770 - val_loss: 0.6113 - val_acc: 0.6417\n",
      "Epoch 14/150\n",
      "514/514 [==============================] - 0s - loss: 0.5963 - acc: 0.6868 - val_loss: 0.6053 - val_acc: 0.7047\n",
      "Epoch 15/150\n",
      "514/514 [==============================] - 0s - loss: 0.5921 - acc: 0.6809 - val_loss: 0.5995 - val_acc: 0.6969\n",
      "Epoch 16/150\n",
      "514/514 [==============================] - 0s - loss: 0.5947 - acc: 0.6809 - val_loss: 0.5950 - val_acc: 0.7126\n",
      "Epoch 17/150\n",
      "514/514 [==============================] - 0s - loss: 0.5900 - acc: 0.6984 - val_loss: 0.5934 - val_acc: 0.6614\n",
      "Epoch 18/150\n",
      "514/514 [==============================] - 0s - loss: 0.5836 - acc: 0.7043 - val_loss: 0.5979 - val_acc: 0.7047\n",
      "Epoch 19/150\n",
      "514/514 [==============================] - 0s - loss: 0.5898 - acc: 0.6887 - val_loss: 0.5924 - val_acc: 0.6417\n",
      "Epoch 20/150\n",
      "514/514 [==============================] - 0s - loss: 0.5886 - acc: 0.6751 - val_loss: 0.5902 - val_acc: 0.6535\n",
      "Epoch 21/150\n",
      "514/514 [==============================] - 0s - loss: 0.5839 - acc: 0.6887 - val_loss: 0.5869 - val_acc: 0.6772\n",
      "Epoch 22/150\n",
      "514/514 [==============================] - 0s - loss: 0.5796 - acc: 0.6907 - val_loss: 0.5971 - val_acc: 0.7126\n",
      "Epoch 23/150\n",
      "514/514 [==============================] - 0s - loss: 0.5801 - acc: 0.7160 - val_loss: 0.5923 - val_acc: 0.7008\n",
      "Epoch 24/150\n",
      "514/514 [==============================] - 0s - loss: 0.5802 - acc: 0.6965 - val_loss: 0.5874 - val_acc: 0.6457\n",
      "Epoch 25/150\n",
      "514/514 [==============================] - 0s - loss: 0.5822 - acc: 0.6926 - val_loss: 0.5862 - val_acc: 0.6575\n",
      "Epoch 26/150\n",
      "514/514 [==============================] - 0s - loss: 0.5793 - acc: 0.7004 - val_loss: 0.5832 - val_acc: 0.6850\n",
      "Epoch 27/150\n",
      "514/514 [==============================] - 0s - loss: 0.5823 - acc: 0.7101 - val_loss: 0.5909 - val_acc: 0.6417\n",
      "Epoch 28/150\n",
      "514/514 [==============================] - 0s - loss: 0.5765 - acc: 0.7043 - val_loss: 0.5850 - val_acc: 0.6614\n",
      "Epoch 29/150\n",
      "514/514 [==============================] - 0s - loss: 0.5749 - acc: 0.7023 - val_loss: 0.5839 - val_acc: 0.6575\n",
      "Epoch 30/150\n",
      "514/514 [==============================] - 0s - loss: 0.5737 - acc: 0.6946 - val_loss: 0.5825 - val_acc: 0.6457\n",
      "Epoch 31/150\n",
      "514/514 [==============================] - 0s - loss: 0.5697 - acc: 0.7198 - val_loss: 0.5854 - val_acc: 0.6693\n",
      "Epoch 32/150\n",
      "514/514 [==============================] - 0s - loss: 0.5691 - acc: 0.7082 - val_loss: 0.5820 - val_acc: 0.68500\n",
      "Epoch 33/150\n",
      "514/514 [==============================] - 0s - loss: 0.5708 - acc: 0.7237 - val_loss: 0.5797 - val_acc: 0.6535\n",
      "Epoch 34/150\n",
      "514/514 [==============================] - 0s - loss: 0.5741 - acc: 0.6868 - val_loss: 0.5856 - val_acc: 0.6535\n",
      "Epoch 35/150\n",
      "514/514 [==============================] - 0s - loss: 0.5728 - acc: 0.7140 - val_loss: 0.5816 - val_acc: 0.6496\n",
      "Epoch 36/150\n",
      "514/514 [==============================] - 0s - loss: 0.5771 - acc: 0.7160 - val_loss: 0.5882 - val_acc: 0.6654\n",
      "Epoch 37/150\n",
      "514/514 [==============================] - 0s - loss: 0.5671 - acc: 0.7218 - val_loss: 0.5912 - val_acc: 0.6535\n",
      "Epoch 38/150\n",
      "514/514 [==============================] - 0s - loss: 0.5675 - acc: 0.7257 - val_loss: 0.5788 - val_acc: 0.6575\n",
      "Epoch 39/150\n",
      "514/514 [==============================] - 0s - loss: 0.5710 - acc: 0.7179 - val_loss: 0.5772 - val_acc: 0.6614\n",
      "Epoch 40/150\n",
      "514/514 [==============================] - 0s - loss: 0.5683 - acc: 0.7043 - val_loss: 0.5776 - val_acc: 0.6732\n",
      "Epoch 41/150\n",
      "514/514 [==============================] - 0s - loss: 0.5649 - acc: 0.7121 - val_loss: 0.5767 - val_acc: 0.6693\n",
      "Epoch 42/150\n",
      "514/514 [==============================] - 0s - loss: 0.5639 - acc: 0.7237 - val_loss: 0.5780 - val_acc: 0.6772\n",
      "Epoch 43/150\n",
      "514/514 [==============================] - 0s - loss: 0.5599 - acc: 0.7101 - val_loss: 0.5897 - val_acc: 0.6575\n",
      "Epoch 44/150\n",
      "514/514 [==============================] - 0s - loss: 0.5645 - acc: 0.6984 - val_loss: 0.6003 - val_acc: 0.6811\n",
      "Epoch 45/150\n",
      "514/514 [==============================] - 0s - loss: 0.5744 - acc: 0.7140 - val_loss: 0.5753 - val_acc: 0.6772\n",
      "Epoch 46/150\n",
      "514/514 [==============================] - 0s - loss: 0.5599 - acc: 0.7043 - val_loss: 0.5741 - val_acc: 0.6732\n",
      "Epoch 47/150\n",
      "514/514 [==============================] - 0s - loss: 0.5595 - acc: 0.7218 - val_loss: 0.5738 - val_acc: 0.6654\n",
      "Epoch 48/150\n",
      "514/514 [==============================] - 0s - loss: 0.5600 - acc: 0.7374 - val_loss: 0.5910 - val_acc: 0.6693\n",
      "Epoch 49/150\n",
      "514/514 [==============================] - 0s - loss: 0.5695 - acc: 0.7140 - val_loss: 0.5838 - val_acc: 0.6575\n",
      "Epoch 50/150\n",
      "514/514 [==============================] - 0s - loss: 0.5560 - acc: 0.7198 - val_loss: 0.5786 - val_acc: 0.7008\n",
      "Epoch 51/150\n",
      "514/514 [==============================] - 0s - loss: 0.5650 - acc: 0.7101 - val_loss: 0.5710 - val_acc: 0.6732\n",
      "Epoch 52/150\n",
      "514/514 [==============================] - 0s - loss: 0.5548 - acc: 0.7257 - val_loss: 0.5755 - val_acc: 0.6575\n",
      "Epoch 53/150\n",
      "514/514 [==============================] - 0s - loss: 0.5539 - acc: 0.7354 - val_loss: 0.5786 - val_acc: 0.7008\n",
      "Epoch 54/150\n",
      "514/514 [==============================] - 0s - loss: 0.5617 - acc: 0.7101 - val_loss: 0.5690 - val_acc: 0.6969\n",
      "Epoch 55/150\n",
      "514/514 [==============================] - 0s - loss: 0.5607 - acc: 0.7140 - val_loss: 0.5738 - val_acc: 0.6850\n",
      "Epoch 56/150\n",
      "514/514 [==============================] - 0s - loss: 0.5570 - acc: 0.7276 - val_loss: 0.5887 - val_acc: 0.6654\n",
      "Epoch 57/150\n",
      "514/514 [==============================] - 0s - loss: 0.5624 - acc: 0.7315 - val_loss: 0.5688 - val_acc: 0.6693\n",
      "Epoch 58/150\n",
      "514/514 [==============================] - 0s - loss: 0.5533 - acc: 0.7296 - val_loss: 0.5684 - val_acc: 0.6929\n",
      "Epoch 59/150\n",
      "514/514 [==============================] - 0s - loss: 0.5572 - acc: 0.7160 - val_loss: 0.5650 - val_acc: 0.7047\n",
      "Epoch 60/150\n",
      "514/514 [==============================] - 0s - loss: 0.5585 - acc: 0.7257 - val_loss: 0.5745 - val_acc: 0.66930.\n",
      "Epoch 61/150\n",
      "514/514 [==============================] - 0s - loss: 0.5516 - acc: 0.7315 - val_loss: 0.5674 - val_acc: 0.6890\n",
      "Epoch 62/150\n",
      "514/514 [==============================] - 0s - loss: 0.5521 - acc: 0.7335 - val_loss: 0.5724 - val_acc: 0.6772\n",
      "Epoch 63/150\n",
      "514/514 [==============================] - 0s - loss: 0.5507 - acc: 0.7296 - val_loss: 0.5659 - val_acc: 0.7126\n",
      "Epoch 64/150\n",
      "514/514 [==============================] - 0s - loss: 0.5524 - acc: 0.7237 - val_loss: 0.5650 - val_acc: 0.6850\n",
      "Epoch 65/150\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "514/514 [==============================] - 0s - loss: 0.5469 - acc: 0.7471 - val_loss: 0.5656 - val_acc: 0.6850\n",
      "Epoch 66/150\n",
      "514/514 [==============================] - 0s - loss: 0.5528 - acc: 0.7276 - val_loss: 0.5796 - val_acc: 0.6850\n",
      "Epoch 67/150\n",
      "514/514 [==============================] - 0s - loss: 0.5524 - acc: 0.7335 - val_loss: 0.5630 - val_acc: 0.6929\n",
      "Epoch 68/150\n",
      "514/514 [==============================] - 0s - loss: 0.5520 - acc: 0.6907 - val_loss: 0.5615 - val_acc: 0.7165\n",
      "Epoch 69/150\n",
      "514/514 [==============================] - 0s - loss: 0.5567 - acc: 0.7140 - val_loss: 0.5631 - val_acc: 0.70470.715\n",
      "Epoch 70/150\n",
      "514/514 [==============================] - 0s - loss: 0.5502 - acc: 0.7198 - val_loss: 0.5572 - val_acc: 0.7244\n",
      "Epoch 71/150\n",
      "514/514 [==============================] - 0s - loss: 0.5463 - acc: 0.7432 - val_loss: 0.5723 - val_acc: 0.6969\n",
      "Epoch 72/150\n",
      "514/514 [==============================] - 0s - loss: 0.5507 - acc: 0.7393 - val_loss: 0.5599 - val_acc: 0.7087\n",
      "Epoch 73/150\n",
      "514/514 [==============================] - 0s - loss: 0.5457 - acc: 0.7412 - val_loss: 0.5684 - val_acc: 0.7008\n",
      "Epoch 74/150\n",
      "514/514 [==============================] - 0s - loss: 0.5477 - acc: 0.7237 - val_loss: 0.5643 - val_acc: 0.7008\n",
      "Epoch 75/150\n",
      "514/514 [==============================] - 0s - loss: 0.5417 - acc: 0.7432 - val_loss: 0.5581 - val_acc: 0.7244\n",
      "Epoch 76/150\n",
      "514/514 [==============================] - 0s - loss: 0.5450 - acc: 0.7393 - val_loss: 0.5613 - val_acc: 0.7205\n",
      "Epoch 77/150\n",
      "514/514 [==============================] - 0s - loss: 0.5478 - acc: 0.7471 - val_loss: 0.5636 - val_acc: 0.7480\n",
      "Epoch 78/150\n",
      "514/514 [==============================] - 0s - loss: 0.5406 - acc: 0.7451 - val_loss: 0.5589 - val_acc: 0.7165\n",
      "Epoch 79/150\n",
      "514/514 [==============================] - 0s - loss: 0.5399 - acc: 0.7374 - val_loss: 0.5657 - val_acc: 0.6890\n",
      "Epoch 80/150\n",
      "514/514 [==============================] - 0s - loss: 0.5372 - acc: 0.7529 - val_loss: 0.5559 - val_acc: 0.7165\n",
      "Epoch 81/150\n",
      "514/514 [==============================] - 0s - loss: 0.5382 - acc: 0.7471 - val_loss: 0.5568 - val_acc: 0.7205\n",
      "Epoch 82/150\n",
      "514/514 [==============================] - 0s - loss: 0.5479 - acc: 0.7218 - val_loss: 0.5624 - val_acc: 0.7008\n",
      "Epoch 83/150\n",
      "514/514 [==============================] - 0s - loss: 0.5402 - acc: 0.7432 - val_loss: 0.5568 - val_acc: 0.7205\n",
      "Epoch 84/150\n",
      "514/514 [==============================] - 0s - loss: 0.5322 - acc: 0.7568 - val_loss: 0.5537 - val_acc: 0.7126\n",
      "Epoch 85/150\n",
      "514/514 [==============================] - 0s - loss: 0.5400 - acc: 0.7529 - val_loss: 0.5576 - val_acc: 0.7283\n",
      "Epoch 86/150\n",
      "514/514 [==============================] - 0s - loss: 0.5310 - acc: 0.7588 - val_loss: 0.5563 - val_acc: 0.7126\n",
      "Epoch 87/150\n",
      "514/514 [==============================] - 0s - loss: 0.5350 - acc: 0.7471 - val_loss: 0.5564 - val_acc: 0.7283\n",
      "Epoch 88/150\n",
      "514/514 [==============================] - 0s - loss: 0.5280 - acc: 0.7568 - val_loss: 0.5568 - val_acc: 0.7008\n",
      "Epoch 89/150\n",
      "514/514 [==============================] - 0s - loss: 0.5348 - acc: 0.7451 - val_loss: 0.5500 - val_acc: 0.7244\n",
      "Epoch 90/150\n",
      "514/514 [==============================] - 0s - loss: 0.5358 - acc: 0.7549 - val_loss: 0.5481 - val_acc: 0.7362\n",
      "Epoch 91/150\n",
      "514/514 [==============================] - 0s - loss: 0.5326 - acc: 0.7451 - val_loss: 0.5583 - val_acc: 0.7283\n",
      "Epoch 92/150\n",
      "514/514 [==============================] - 0s - loss: 0.5303 - acc: 0.7588 - val_loss: 0.5558 - val_acc: 0.7323\n",
      "Epoch 93/150\n",
      "514/514 [==============================] - 0s - loss: 0.5341 - acc: 0.7412 - val_loss: 0.5503 - val_acc: 0.7205\n",
      "Epoch 94/150\n",
      "514/514 [==============================] - 0s - loss: 0.5287 - acc: 0.7471 - val_loss: 0.5583 - val_acc: 0.7283\n",
      "Epoch 95/150\n",
      "514/514 [==============================] - 0s - loss: 0.5250 - acc: 0.7588 - val_loss: 0.5500 - val_acc: 0.7165\n",
      "Epoch 96/150\n",
      "514/514 [==============================] - 0s - loss: 0.5425 - acc: 0.7237 - val_loss: 0.5480 - val_acc: 0.7402\n",
      "Epoch 97/150\n",
      "514/514 [==============================] - 0s - loss: 0.5245 - acc: 0.7568 - val_loss: 0.5452 - val_acc: 0.7362\n",
      "Epoch 98/150\n",
      "514/514 [==============================] - 0s - loss: 0.5268 - acc: 0.7471 - val_loss: 0.5500 - val_acc: 0.7441\n",
      "Epoch 99/150\n",
      "514/514 [==============================] - 0s - loss: 0.5254 - acc: 0.7471 - val_loss: 0.5476 - val_acc: 0.7205\n",
      "Epoch 100/150\n",
      "514/514 [==============================] - 0s - loss: 0.5223 - acc: 0.7510 - val_loss: 0.5433 - val_acc: 0.7283\n",
      "Epoch 101/150\n",
      "514/514 [==============================] - 0s - loss: 0.5227 - acc: 0.7335 - val_loss: 0.5438 - val_acc: 0.7402\n",
      "Epoch 102/150\n",
      "514/514 [==============================] - 0s - loss: 0.5245 - acc: 0.7354 - val_loss: 0.5430 - val_acc: 0.7480\n",
      "Epoch 103/150\n",
      "514/514 [==============================] - 0s - loss: 0.5287 - acc: 0.7354 - val_loss: 0.5655 - val_acc: 0.72440.7\n",
      "Epoch 104/150\n",
      "514/514 [==============================] - 0s - loss: 0.5206 - acc: 0.7412 - val_loss: 0.5516 - val_acc: 0.7362\n",
      "Epoch 105/150\n",
      "514/514 [==============================] - 0s - loss: 0.5226 - acc: 0.7471 - val_loss: 0.5662 - val_acc: 0.7165\n",
      "Epoch 106/150\n",
      "514/514 [==============================] - 0s - loss: 0.5336 - acc: 0.7529 - val_loss: 0.5458 - val_acc: 0.7323\n",
      "Epoch 107/150\n",
      "514/514 [==============================] - 0s - loss: 0.5177 - acc: 0.7490 - val_loss: 0.5474 - val_acc: 0.7244\n",
      "Epoch 108/150\n",
      "514/514 [==============================] - 0s - loss: 0.5175 - acc: 0.7490 - val_loss: 0.5444 - val_acc: 0.7559\n",
      "Epoch 109/150\n",
      "514/514 [==============================] - 0s - loss: 0.5169 - acc: 0.7549 - val_loss: 0.5388 - val_acc: 0.7520\n",
      "Epoch 110/150\n",
      "514/514 [==============================] - 0s - loss: 0.5233 - acc: 0.7432 - val_loss: 0.5487 - val_acc: 0.7362\n",
      "Epoch 111/150\n",
      "514/514 [==============================] - 0s - loss: 0.5135 - acc: 0.7510 - val_loss: 0.5401 - val_acc: 0.7283\n",
      "Epoch 112/150\n",
      "514/514 [==============================] - 0s - loss: 0.5177 - acc: 0.7626 - val_loss: 0.5522 - val_acc: 0.7283\n",
      "Epoch 113/150\n",
      "514/514 [==============================] - 0s - loss: 0.5282 - acc: 0.7529 - val_loss: 0.5392 - val_acc: 0.7362\n",
      "Epoch 114/150\n",
      "514/514 [==============================] - 0s - loss: 0.5118 - acc: 0.7626 - val_loss: 0.5666 - val_acc: 0.7205\n",
      "Epoch 115/150\n",
      "514/514 [==============================] - 0s - loss: 0.5064 - acc: 0.7743 - val_loss: 0.5506 - val_acc: 0.7323\n",
      "Epoch 116/150\n",
      "514/514 [==============================] - 0s - loss: 0.5164 - acc: 0.7549 - val_loss: 0.5330 - val_acc: 0.7402\n",
      "Epoch 117/150\n",
      "514/514 [==============================] - 0s - loss: 0.5101 - acc: 0.7432 - val_loss: 0.5363 - val_acc: 0.7480\n",
      "Epoch 118/150\n",
      "514/514 [==============================] - 0s - loss: 0.5153 - acc: 0.7529 - val_loss: 0.5482 - val_acc: 0.7205\n",
      "Epoch 119/150\n",
      "514/514 [==============================] - 0s - loss: 0.5264 - acc: 0.7237 - val_loss: 0.5832 - val_acc: 0.6772\n",
      "Epoch 120/150\n",
      "514/514 [==============================] - ETA: 0s - loss: 0.5273 - acc: 0.756 - 0s - loss: 0.5268 - acc: 0.7549 - val_loss: 0.5481 - val_acc: 0.7559\n",
      "Epoch 121/150\n",
      "514/514 [==============================] - 0s - loss: 0.5140 - acc: 0.7529 - val_loss: 0.5379 - val_acc: 0.7402\n",
      "Epoch 122/150\n",
      "514/514 [==============================] - 0s - loss: 0.5139 - acc: 0.7607 - val_loss: 0.5357 - val_acc: 0.7402\n",
      "Epoch 123/150\n",
      "514/514 [==============================] - 0s - loss: 0.5051 - acc: 0.7568 - val_loss: 0.5415 - val_acc: 0.7520\n",
      "Epoch 124/150\n",
      "514/514 [==============================] - 0s - loss: 0.5002 - acc: 0.7626 - val_loss: 0.5334 - val_acc: 0.7441\n",
      "Epoch 125/150\n",
      "514/514 [==============================] - 0s - loss: 0.5111 - acc: 0.7549 - val_loss: 0.5345 - val_acc: 0.7402\n",
      "Epoch 126/150\n",
      "514/514 [==============================] - 0s - loss: 0.5146 - acc: 0.7471 - val_loss: 0.5387 - val_acc: 0.7323\n",
      "Epoch 127/150\n",
      "514/514 [==============================] - 0s - loss: 0.5005 - acc: 0.7588 - val_loss: 0.5286 - val_acc: 0.7402\n",
      "Epoch 128/150\n",
      "514/514 [==============================] - 0s - loss: 0.5105 - acc: 0.7471 - val_loss: 0.5339 - val_acc: 0.7402\n",
      "Epoch 129/150\n",
      "514/514 [==============================] - 0s - loss: 0.5070 - acc: 0.7510 - val_loss: 0.5285 - val_acc: 0.7362\n",
      "Epoch 130/150\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "514/514 [==============================] - 0s - loss: 0.5068 - acc: 0.7529 - val_loss: 0.5337 - val_acc: 0.7402\n",
      "Epoch 131/150\n",
      "514/514 [==============================] - 0s - loss: 0.5083 - acc: 0.7588 - val_loss: 0.5289 - val_acc: 0.7362\n",
      "Epoch 132/150\n",
      "514/514 [==============================] - 0s - loss: 0.5003 - acc: 0.7588 - val_loss: 0.5271 - val_acc: 0.7402\n",
      "Epoch 133/150\n",
      "514/514 [==============================] - 0s - loss: 0.4994 - acc: 0.7568 - val_loss: 0.5314 - val_acc: 0.7283\n",
      "Epoch 134/150\n",
      "514/514 [==============================] - 0s - loss: 0.5007 - acc: 0.7763 - val_loss: 0.5317 - val_acc: 0.7520\n",
      "Epoch 135/150\n",
      "514/514 [==============================] - 0s - loss: 0.4994 - acc: 0.7665 - val_loss: 0.5323 - val_acc: 0.7480\n",
      "Epoch 136/150\n",
      "514/514 [==============================] - 0s - loss: 0.5089 - acc: 0.7568 - val_loss: 0.5325 - val_acc: 0.7362\n",
      "Epoch 137/150\n",
      "514/514 [==============================] - 0s - loss: 0.5035 - acc: 0.7549 - val_loss: 0.5342 - val_acc: 0.7362\n",
      "Epoch 138/150\n",
      "514/514 [==============================] - 0s - loss: 0.5006 - acc: 0.7549 - val_loss: 0.5363 - val_acc: 0.7402\n",
      "Epoch 139/150\n",
      "514/514 [==============================] - 0s - loss: 0.5020 - acc: 0.7529 - val_loss: 0.5466 - val_acc: 0.7441\n",
      "Epoch 140/150\n",
      "514/514 [==============================] - 0s - loss: 0.5015 - acc: 0.7646 - val_loss: 0.5334 - val_acc: 0.7480\n",
      "Epoch 141/150\n",
      "514/514 [==============================] - 0s - loss: 0.4987 - acc: 0.7607 - val_loss: 0.5269 - val_acc: 0.7559\n",
      "Epoch 142/150\n",
      "514/514 [==============================] - 0s - loss: 0.5013 - acc: 0.7510 - val_loss: 0.5276 - val_acc: 0.7362\n",
      "Epoch 143/150\n",
      "514/514 [==============================] - 0s - loss: 0.4922 - acc: 0.7626 - val_loss: 0.5392 - val_acc: 0.7402\n",
      "Epoch 144/150\n",
      "514/514 [==============================] - 0s - loss: 0.4920 - acc: 0.7704 - val_loss: 0.5294 - val_acc: 0.7480\n",
      "Epoch 145/150\n",
      "514/514 [==============================] - 0s - loss: 0.4984 - acc: 0.7529 - val_loss: 0.5215 - val_acc: 0.7402\n",
      "Epoch 146/150\n",
      "514/514 [==============================] - 0s - loss: 0.4923 - acc: 0.7646 - val_loss: 0.5328 - val_acc: 0.7520\n",
      "Epoch 147/150\n",
      "514/514 [==============================] - 0s - loss: 0.4965 - acc: 0.7626 - val_loss: 0.5246 - val_acc: 0.7441\n",
      "Epoch 148/150\n",
      "514/514 [==============================] - 0s - loss: 0.4965 - acc: 0.7802 - val_loss: 0.5198 - val_acc: 0.7520\n",
      "Epoch 149/150\n",
      "514/514 [==============================] - 0s - loss: 0.4988 - acc: 0.7510 - val_loss: 0.5298 - val_acc: 0.7402\n",
      "Epoch 150/150\n",
      "514/514 [==============================] - 0s - loss: 0.4999 - acc: 0.7490 - val_loss: 0.5186 - val_acc: 0.7480\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1e7451688d0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X,Y,validation_split=0.33, nb_epoch=150, batch_size=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train,X_test,y_train,y_test=train_test_split(X,Y,test_size=0.33,random_state=seed,)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 768 samples, validate on 254 samples\n",
      "Epoch 1/150\n",
      "768/768 [==============================] - 0s - loss: 0.4003 - acc: 0.8060 - val_loss: 0.4120 - val_acc: 0.7717\n",
      "Epoch 2/150\n",
      "768/768 [==============================] - 0s - loss: 0.4024 - acc: 0.8138 - val_loss: 0.4135 - val_acc: 0.7874\n",
      "Epoch 3/150\n",
      "768/768 [==============================] - 0s - loss: 0.4068 - acc: 0.8021 - val_loss: 0.4167 - val_acc: 0.7795\n",
      "Epoch 4/150\n",
      "768/768 [==============================] - 0s - loss: 0.4035 - acc: 0.7943 - val_loss: 0.4097 - val_acc: 0.7874\n",
      "Epoch 5/150\n",
      "768/768 [==============================] - 0s - loss: 0.4019 - acc: 0.7930 - val_loss: 0.4189 - val_acc: 0.7835\n",
      "Epoch 6/150\n",
      "768/768 [==============================] - 0s - loss: 0.4144 - acc: 0.7943 - val_loss: 0.4361 - val_acc: 0.7835\n",
      "Epoch 7/150\n",
      "768/768 [==============================] - 0s - loss: 0.4118 - acc: 0.7852 - val_loss: 0.4212 - val_acc: 0.7953\n",
      "Epoch 8/150\n",
      "768/768 [==============================] - 0s - loss: 0.4105 - acc: 0.8034 - val_loss: 0.4247 - val_acc: 0.7953\n",
      "Epoch 9/150\n",
      "768/768 [==============================] - 0s - loss: 0.4051 - acc: 0.8047 - val_loss: 0.4071 - val_acc: 0.79530.80\n",
      "Epoch 10/150\n",
      "768/768 [==============================] - 0s - loss: 0.4016 - acc: 0.8086 - val_loss: 0.4229 - val_acc: 0.7913\n",
      "Epoch 11/150\n",
      "768/768 [==============================] - 0s - loss: 0.3996 - acc: 0.8060 - val_loss: 0.4418 - val_acc: 0.7953\n",
      "Epoch 12/150\n",
      "768/768 [==============================] - 0s - loss: 0.3981 - acc: 0.8086 - val_loss: 0.4262 - val_acc: 0.7913\n",
      "Epoch 13/150\n",
      "768/768 [==============================] - 0s - loss: 0.4046 - acc: 0.7930 - val_loss: 0.4234 - val_acc: 0.7913\n",
      "Epoch 14/150\n",
      "768/768 [==============================] - 0s - loss: 0.3978 - acc: 0.8008 - val_loss: 0.4108 - val_acc: 0.7913\n",
      "Epoch 15/150\n",
      "768/768 [==============================] - 0s - loss: 0.4036 - acc: 0.7982 - val_loss: 0.4139 - val_acc: 0.7756\n",
      "Epoch 16/150\n",
      "768/768 [==============================] - 0s - loss: 0.3959 - acc: 0.8034 - val_loss: 0.4182 - val_acc: 0.7874\n",
      "Epoch 17/150\n",
      "768/768 [==============================] - 0s - loss: 0.4024 - acc: 0.7995 - val_loss: 0.4103 - val_acc: 0.7953\n",
      "Epoch 18/150\n",
      "768/768 [==============================] - ETA: 0s - loss: 0.4039 - acc: 0.813 - 0s - loss: 0.4055 - acc: 0.8112 - val_loss: 0.4301 - val_acc: 0.7874\n",
      "Epoch 19/150\n",
      "768/768 [==============================] - 0s - loss: 0.3926 - acc: 0.8034 - val_loss: 0.4120 - val_acc: 0.7953\n",
      "Epoch 20/150\n",
      "768/768 [==============================] - 0s - loss: 0.4022 - acc: 0.7995 - val_loss: 0.4245 - val_acc: 0.7874\n",
      "Epoch 21/150\n",
      "768/768 [==============================] - 0s - loss: 0.4018 - acc: 0.8112 - val_loss: 0.4224 - val_acc: 0.7992\n",
      "Epoch 22/150\n",
      "768/768 [==============================] - 0s - loss: 0.3936 - acc: 0.8047 - val_loss: 0.4365 - val_acc: 0.7953\n",
      "Epoch 23/150\n",
      "768/768 [==============================] - 0s - loss: 0.4115 - acc: 0.8034 - val_loss: 0.4322 - val_acc: 0.7874\n",
      "Epoch 24/150\n",
      "768/768 [==============================] - 0s - loss: 0.4080 - acc: 0.8008 - val_loss: 0.4310 - val_acc: 0.7795\n",
      "Epoch 25/150\n",
      "768/768 [==============================] - 0s - loss: 0.4049 - acc: 0.8008 - val_loss: 0.4403 - val_acc: 0.7795\n",
      "Epoch 26/150\n",
      "768/768 [==============================] - 0s - loss: 0.4099 - acc: 0.8021 - val_loss: 0.4167 - val_acc: 0.7992\n",
      "Epoch 27/150\n",
      "768/768 [==============================] - 0s - loss: 0.4003 - acc: 0.8099 - val_loss: 0.4072 - val_acc: 0.7953\n",
      "Epoch 28/150\n",
      "768/768 [==============================] - 0s - loss: 0.4002 - acc: 0.8125 - val_loss: 0.4330 - val_acc: 0.7953\n",
      "Epoch 29/150\n",
      "768/768 [==============================] - 0s - loss: 0.4035 - acc: 0.8021 - val_loss: 0.4732 - val_acc: 0.7717\n",
      "Epoch 30/150\n",
      "768/768 [==============================] - 0s - loss: 0.4002 - acc: 0.7956 - val_loss: 0.4118 - val_acc: 0.7992\n",
      "Epoch 31/150\n",
      "768/768 [==============================] - 0s - loss: 0.4012 - acc: 0.8073 - val_loss: 0.4293 - val_acc: 0.7756\n",
      "Epoch 32/150\n",
      "768/768 [==============================] - 0s - loss: 0.4049 - acc: 0.7904 - val_loss: 0.4222 - val_acc: 0.7874\n",
      "Epoch 33/150\n",
      "768/768 [==============================] - 0s - loss: 0.3999 - acc: 0.8060 - val_loss: 0.4261 - val_acc: 0.7795\n",
      "Epoch 34/150\n",
      "768/768 [==============================] - 0s - loss: 0.4043 - acc: 0.8112 - val_loss: 0.4193 - val_acc: 0.7756\n",
      "Epoch 35/150\n",
      "768/768 [==============================] - 0s - loss: 0.4036 - acc: 0.8060 - val_loss: 0.4480 - val_acc: 0.7835\n",
      "Epoch 36/150\n",
      "768/768 [==============================] - 0s - loss: 0.4097 - acc: 0.8073 - val_loss: 0.4579 - val_acc: 0.7992\n",
      "Epoch 37/150\n",
      "768/768 [==============================] - 0s - loss: 0.4043 - acc: 0.7930 - val_loss: 0.4373 - val_acc: 0.7795\n",
      "Epoch 38/150\n",
      "768/768 [==============================] - 0s - loss: 0.4066 - acc: 0.7904 - val_loss: 0.4077 - val_acc: 0.7953\n",
      "Epoch 39/150\n",
      "768/768 [==============================] - 0s - loss: 0.3950 - acc: 0.8125 - val_loss: 0.4144 - val_acc: 0.7913\n",
      "Epoch 40/150\n",
      "768/768 [==============================] - 0s - loss: 0.4142 - acc: 0.7878 - val_loss: 0.4382 - val_acc: 0.7874\n",
      "Epoch 41/150\n",
      "768/768 [==============================] - 0s - loss: 0.4046 - acc: 0.7956 - val_loss: 0.4518 - val_acc: 0.8031\n",
      "Epoch 42/150\n",
      "768/768 [==============================] - 0s - loss: 0.4123 - acc: 0.8008 - val_loss: 0.4182 - val_acc: 0.7756\n",
      "Epoch 43/150\n",
      "768/768 [==============================] - 0s - loss: 0.4141 - acc: 0.8086 - val_loss: 0.4206 - val_acc: 0.7992\n",
      "Epoch 44/150\n",
      "768/768 [==============================] - 0s - loss: 0.4122 - acc: 0.7995 - val_loss: 0.4155 - val_acc: 0.7835\n",
      "Epoch 45/150\n",
      "768/768 [==============================] - 0s - loss: 0.3931 - acc: 0.8008 - val_loss: 0.4123 - val_acc: 0.7874\n",
      "Epoch 46/150\n",
      "768/768 [==============================] - 0s - loss: 0.3962 - acc: 0.8112 - val_loss: 0.4100 - val_acc: 0.7835\n",
      "Epoch 47/150\n",
      "768/768 [==============================] - 0s - loss: 0.3966 - acc: 0.8034 - val_loss: 0.4277 - val_acc: 0.7795\n",
      "Epoch 48/150\n",
      "768/768 [==============================] - 0s - loss: 0.4058 - acc: 0.7956 - val_loss: 0.4147 - val_acc: 0.7795\n",
      "Epoch 49/150\n",
      "768/768 [==============================] - 0s - loss: 0.4012 - acc: 0.7943 - val_loss: 0.4332 - val_acc: 0.8031\n",
      "Epoch 50/150\n",
      "768/768 [==============================] - 0s - loss: 0.4094 - acc: 0.7982 - val_loss: 0.4819 - val_acc: 0.7756\n",
      "Epoch 51/150\n",
      "768/768 [==============================] - 0s - loss: 0.4103 - acc: 0.7995 - val_loss: 0.4186 - val_acc: 0.7717\n",
      "Epoch 52/150\n",
      "768/768 [==============================] - 0s - loss: 0.4023 - acc: 0.7995 - val_loss: 0.4242 - val_acc: 0.7874\n",
      "Epoch 53/150\n",
      "768/768 [==============================] - 0s - loss: 0.4047 - acc: 0.8047 - val_loss: 0.4154 - val_acc: 0.7953\n",
      "Epoch 54/150\n",
      "768/768 [==============================] - 0s - loss: 0.4122 - acc: 0.7969 - val_loss: 0.4272 - val_acc: 0.8031\n",
      "Epoch 55/150\n",
      "768/768 [==============================] - 0s - loss: 0.4040 - acc: 0.8047 - val_loss: 0.4117 - val_acc: 0.7756\n",
      "Epoch 56/150\n",
      "768/768 [==============================] - 0s - loss: 0.4134 - acc: 0.7930 - val_loss: 0.4174 - val_acc: 0.8031\n",
      "Epoch 57/150\n",
      "768/768 [==============================] - 0s - loss: 0.3951 - acc: 0.8047 - val_loss: 0.4109 - val_acc: 0.7953\n",
      "Epoch 58/150\n",
      "768/768 [==============================] - 0s - loss: 0.4053 - acc: 0.8021 - val_loss: 0.4257 - val_acc: 0.7874\n",
      "Epoch 59/150\n",
      "768/768 [==============================] - 0s - loss: 0.3948 - acc: 0.8073 - val_loss: 0.4149 - val_acc: 0.8031\n",
      "Epoch 60/150\n",
      "768/768 [==============================] - 0s - loss: 0.4061 - acc: 0.7930 - val_loss: 0.4274 - val_acc: 0.7795\n",
      "Epoch 61/150\n",
      "768/768 [==============================] - 0s - loss: 0.4138 - acc: 0.8047 - val_loss: 0.4306 - val_acc: 0.7953\n",
      "Epoch 62/150\n",
      "768/768 [==============================] - 0s - loss: 0.4061 - acc: 0.8008 - val_loss: 0.4103 - val_acc: 0.7874\n",
      "Epoch 63/150\n",
      "768/768 [==============================] - 0s - loss: 0.3922 - acc: 0.8021 - val_loss: 0.4242 - val_acc: 0.8110\n",
      "Epoch 64/150\n",
      "768/768 [==============================] - 0s - loss: 0.4086 - acc: 0.7982 - val_loss: 0.4230 - val_acc: 0.7913\n",
      "Epoch 65/150\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "768/768 [==============================] - 0s - loss: 0.4000 - acc: 0.8112 - val_loss: 0.4595 - val_acc: 0.7874\n",
      "Epoch 66/150\n",
      "768/768 [==============================] - 0s - loss: 0.4011 - acc: 0.8034 - val_loss: 0.4190 - val_acc: 0.7992\n",
      "Epoch 67/150\n",
      "768/768 [==============================] - 0s - loss: 0.3944 - acc: 0.8099 - val_loss: 0.4478 - val_acc: 0.7953\n",
      "Epoch 68/150\n",
      "768/768 [==============================] - 0s - loss: 0.4046 - acc: 0.8008 - val_loss: 0.4254 - val_acc: 0.7992\n",
      "Epoch 69/150\n",
      "768/768 [==============================] - 0s - loss: 0.4008 - acc: 0.8034 - val_loss: 0.4215 - val_acc: 0.7913\n",
      "Epoch 70/150\n",
      "768/768 [==============================] - 0s - loss: 0.4040 - acc: 0.8008 - val_loss: 0.4345 - val_acc: 0.7835\n",
      "Epoch 71/150\n",
      "768/768 [==============================] - 0s - loss: 0.4027 - acc: 0.7969 - val_loss: 0.4072 - val_acc: 0.7992\n",
      "Epoch 72/150\n",
      "768/768 [==============================] - 0s - loss: 0.4050 - acc: 0.8125 - val_loss: 0.4588 - val_acc: 0.8031\n",
      "Epoch 73/150\n",
      "768/768 [==============================] - 0s - loss: 0.3980 - acc: 0.8112 - val_loss: 0.4067 - val_acc: 0.7913\n",
      "Epoch 74/150\n",
      "768/768 [==============================] - 0s - loss: 0.4036 - acc: 0.7969 - val_loss: 0.4155 - val_acc: 0.7913\n",
      "Epoch 75/150\n",
      "768/768 [==============================] - 0s - loss: 0.4079 - acc: 0.8047 - val_loss: 0.4522 - val_acc: 0.7874\n",
      "Epoch 76/150\n",
      "768/768 [==============================] - 0s - loss: 0.4010 - acc: 0.8021 - val_loss: 0.4133 - val_acc: 0.8031\n",
      "Epoch 77/150\n",
      "768/768 [==============================] - 0s - loss: 0.3968 - acc: 0.8086 - val_loss: 0.4179 - val_acc: 0.7953\n",
      "Epoch 78/150\n",
      "768/768 [==============================] - 0s - loss: 0.3996 - acc: 0.8112 - val_loss: 0.4300 - val_acc: 0.7874\n",
      "Epoch 79/150\n",
      "768/768 [==============================] - 0s - loss: 0.3970 - acc: 0.8073 - val_loss: 0.4257 - val_acc: 0.7953\n",
      "Epoch 80/150\n",
      "768/768 [==============================] - 0s - loss: 0.4069 - acc: 0.8047 - val_loss: 0.4367 - val_acc: 0.7953\n",
      "Epoch 81/150\n",
      "768/768 [==============================] - 0s - loss: 0.4078 - acc: 0.7995 - val_loss: 0.4092 - val_acc: 0.8031\n",
      "Epoch 82/150\n",
      "768/768 [==============================] - 0s - loss: 0.3928 - acc: 0.8138 - val_loss: 0.4730 - val_acc: 0.7874\n",
      "Epoch 83/150\n",
      "768/768 [==============================] - 0s - loss: 0.4044 - acc: 0.8099 - val_loss: 0.4136 - val_acc: 0.7874\n",
      "Epoch 84/150\n",
      "768/768 [==============================] - 0s - loss: 0.4087 - acc: 0.8047 - val_loss: 0.4473 - val_acc: 0.7913\n",
      "Epoch 85/150\n",
      "768/768 [==============================] - 0s - loss: 0.4123 - acc: 0.7917 - val_loss: 0.4191 - val_acc: 0.7874\n",
      "Epoch 86/150\n",
      "768/768 [==============================] - 0s - loss: 0.3970 - acc: 0.8216 - val_loss: 0.4202 - val_acc: 0.8071\n",
      "Epoch 87/150\n",
      "768/768 [==============================] - 0s - loss: 0.3926 - acc: 0.8099 - val_loss: 0.4496 - val_acc: 0.7835\n",
      "Epoch 88/150\n",
      "768/768 [==============================] - ETA: 0s - loss: 0.4019 - acc: 0.801 - 0s - loss: 0.4026 - acc: 0.7995 - val_loss: 0.4175 - val_acc: 0.8031\n",
      "Epoch 89/150\n",
      "768/768 [==============================] - 0s - loss: 0.4029 - acc: 0.8021 - val_loss: 0.4325 - val_acc: 0.7795\n",
      "Epoch 90/150\n",
      "768/768 [==============================] - 0s - loss: 0.4067 - acc: 0.7904 - val_loss: 0.4331 - val_acc: 0.7795\n",
      "Epoch 91/150\n",
      "768/768 [==============================] - 0s - loss: 0.3952 - acc: 0.8112 - val_loss: 0.4143 - val_acc: 0.7953\n",
      "Epoch 92/150\n",
      "768/768 [==============================] - ETA: 0s - loss: 0.4112 - acc: 0.807 - 0s - loss: 0.4100 - acc: 0.8086 - val_loss: 0.4727 - val_acc: 0.7717\n",
      "Epoch 93/150\n",
      "768/768 [==============================] - 0s - loss: 0.4116 - acc: 0.7982 - val_loss: 0.4748 - val_acc: 0.7795\n",
      "Epoch 94/150\n",
      "768/768 [==============================] - 0s - loss: 0.4150 - acc: 0.7917 - val_loss: 0.4124 - val_acc: 0.7913\n",
      "Epoch 95/150\n",
      "768/768 [==============================] - 0s - loss: 0.4088 - acc: 0.8021 - val_loss: 0.4079 - val_acc: 0.8071\n",
      "Epoch 96/150\n",
      "768/768 [==============================] - 0s - loss: 0.4014 - acc: 0.8073 - val_loss: 0.4436 - val_acc: 0.7913\n",
      "Epoch 97/150\n",
      "768/768 [==============================] - 0s - loss: 0.3882 - acc: 0.8034 - val_loss: 0.4567 - val_acc: 0.7835\n",
      "Epoch 98/150\n",
      "768/768 [==============================] - 0s - loss: 0.4088 - acc: 0.8008 - val_loss: 0.4316 - val_acc: 0.77950\n",
      "Epoch 99/150\n",
      "768/768 [==============================] - 0s - loss: 0.4061 - acc: 0.8112 - val_loss: 0.4093 - val_acc: 0.7992\n",
      "Epoch 100/150\n",
      "768/768 [==============================] - 0s - loss: 0.3956 - acc: 0.8138 - val_loss: 0.4311 - val_acc: 0.8071\n",
      "Epoch 101/150\n",
      "768/768 [==============================] - 0s - loss: 0.3967 - acc: 0.8047 - val_loss: 0.4163 - val_acc: 0.7874\n",
      "Epoch 102/150\n",
      "768/768 [==============================] - 0s - loss: 0.4039 - acc: 0.8034 - val_loss: 0.4273 - val_acc: 0.7953\n",
      "Epoch 103/150\n",
      "768/768 [==============================] - 0s - loss: 0.4079 - acc: 0.7839 - val_loss: 0.4260 - val_acc: 0.7913\n",
      "Epoch 104/150\n",
      "768/768 [==============================] - 0s - loss: 0.4005 - acc: 0.7943 - val_loss: 0.4371 - val_acc: 0.7913\n",
      "Epoch 105/150\n",
      "768/768 [==============================] - 0s - loss: 0.3968 - acc: 0.8021 - val_loss: 0.4331 - val_acc: 0.7913- ETA: 0s - loss: 0.4\n",
      "Epoch 106/150\n",
      "768/768 [==============================] - 0s - loss: 0.4096 - acc: 0.7930 - val_loss: 0.4261 - val_acc: 0.7913 ETA: 0s - loss: 0.4057 - acc: 0.79\n",
      "Epoch 107/150\n",
      "768/768 [==============================] - 0s - loss: 0.4028 - acc: 0.8073 - val_loss: 0.4111 - val_acc: 0.7953\n",
      "Epoch 108/150\n",
      "768/768 [==============================] - 0s - loss: 0.4049 - acc: 0.8060 - val_loss: 0.4438 - val_acc: 0.7913\n",
      "Epoch 109/150\n",
      "768/768 [==============================] - 0s - loss: 0.4204 - acc: 0.7891 - val_loss: 0.4242 - val_acc: 0.7953\n",
      "Epoch 110/150\n",
      "768/768 [==============================] - 0s - loss: 0.3961 - acc: 0.8021 - val_loss: 0.4305 - val_acc: 0.7913\n",
      "Epoch 111/150\n",
      "768/768 [==============================] - 0s - loss: 0.4030 - acc: 0.8177 - val_loss: 0.4668 - val_acc: 0.7795\n",
      "Epoch 112/150\n",
      "768/768 [==============================] - 0s - loss: 0.4043 - acc: 0.8034 - val_loss: 0.4259 - val_acc: 0.80310.8\n",
      "Epoch 113/150\n",
      "768/768 [==============================] - 0s - loss: 0.3995 - acc: 0.7995 - val_loss: 0.4180 - val_acc: 0.7992\n",
      "Epoch 114/150\n",
      "768/768 [==============================] - 0s - loss: 0.3970 - acc: 0.8099 - val_loss: 0.4251 - val_acc: 0.7874\n",
      "Epoch 115/150\n",
      "768/768 [==============================] - 0s - loss: 0.3903 - acc: 0.8060 - val_loss: 0.4289 - val_acc: 0.7874\n",
      "Epoch 116/150\n",
      "768/768 [==============================] - 0s - loss: 0.3914 - acc: 0.8125 - val_loss: 0.4247 - val_acc: 0.7992\n",
      "Epoch 117/150\n",
      "768/768 [==============================] - 0s - loss: 0.3945 - acc: 0.8060 - val_loss: 0.4288 - val_acc: 0.7953\n",
      "Epoch 118/150\n",
      "768/768 [==============================] - 0s - loss: 0.4023 - acc: 0.7969 - val_loss: 0.4087 - val_acc: 0.7913\n",
      "Epoch 119/150\n",
      "768/768 [==============================] - 0s - loss: 0.3941 - acc: 0.8021 - val_loss: 0.4078 - val_acc: 0.7874\n",
      "Epoch 120/150\n",
      "768/768 [==============================] - 0s - loss: 0.3912 - acc: 0.8073 - val_loss: 0.4113 - val_acc: 0.7913\n",
      "Epoch 121/150\n",
      "768/768 [==============================] - 0s - loss: 0.4223 - acc: 0.8034 - val_loss: 0.4534 - val_acc: 0.7953\n",
      "Epoch 122/150\n",
      "768/768 [==============================] - 0s - loss: 0.4111 - acc: 0.8021 - val_loss: 0.4376 - val_acc: 0.7795\n",
      "Epoch 123/150\n",
      "768/768 [==============================] - 0s - loss: 0.4241 - acc: 0.8086 - val_loss: 0.4257 - val_acc: 0.7953\n",
      "Epoch 124/150\n",
      "768/768 [==============================] - 0s - loss: 0.4071 - acc: 0.8047 - val_loss: 0.4453 - val_acc: 0.7953\n",
      "Epoch 125/150\n",
      "768/768 [==============================] - 0s - loss: 0.4089 - acc: 0.8047 - val_loss: 0.4533 - val_acc: 0.7953\n",
      "Epoch 126/150\n",
      "768/768 [==============================] - 0s - loss: 0.4081 - acc: 0.8073 - val_loss: 0.4247 - val_acc: 0.8031\n",
      "Epoch 127/150\n",
      "768/768 [==============================] - 0s - loss: 0.4029 - acc: 0.8073 - val_loss: 0.4218 - val_acc: 0.7874\n",
      "Epoch 128/150\n",
      "768/768 [==============================] - 0s - loss: 0.3962 - acc: 0.8073 - val_loss: 0.4261 - val_acc: 0.8110\n",
      "Epoch 129/150\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "768/768 [==============================] - 0s - loss: 0.4032 - acc: 0.7956 - val_loss: 0.4275 - val_acc: 0.7677\n",
      "Epoch 130/150\n",
      "768/768 [==============================] - 0s - loss: 0.4066 - acc: 0.8164 - val_loss: 0.4329 - val_acc: 0.8150\n",
      "Epoch 131/150\n",
      "768/768 [==============================] - 0s - loss: 0.4028 - acc: 0.8047 - val_loss: 0.4196 - val_acc: 0.7913\n",
      "Epoch 132/150\n",
      "768/768 [==============================] - 0s - loss: 0.4063 - acc: 0.8125 - val_loss: 0.4735 - val_acc: 0.7835\n",
      "Epoch 133/150\n",
      "768/768 [==============================] - 0s - loss: 0.4097 - acc: 0.8021 - val_loss: 0.4789 - val_acc: 0.7362\n",
      "Epoch 134/150\n",
      "768/768 [==============================] - 0s - loss: 0.4084 - acc: 0.8125 - val_loss: 0.4530 - val_acc: 0.7913\n",
      "Epoch 135/150\n",
      "768/768 [==============================] - 0s - loss: 0.4136 - acc: 0.8034 - val_loss: 0.4287 - val_acc: 0.7913\n",
      "Epoch 136/150\n",
      "768/768 [==============================] - 0s - loss: 0.4112 - acc: 0.8164 - val_loss: 0.4286 - val_acc: 0.7835\n",
      "Epoch 137/150\n",
      "768/768 [==============================] - 0s - loss: 0.3950 - acc: 0.8190 - val_loss: 0.4164 - val_acc: 0.7913\n",
      "Epoch 138/150\n",
      "768/768 [==============================] - 0s - loss: 0.4014 - acc: 0.8073 - val_loss: 0.4218 - val_acc: 0.8031\n",
      "Epoch 139/150\n",
      "768/768 [==============================] - 0s - loss: 0.4036 - acc: 0.8008 - val_loss: 0.4328 - val_acc: 0.7874\n",
      "Epoch 140/150\n",
      "768/768 [==============================] - 0s - loss: 0.4053 - acc: 0.8086 - val_loss: 0.4103 - val_acc: 0.7913\n",
      "Epoch 141/150\n",
      "768/768 [==============================] - 0s - loss: 0.3989 - acc: 0.8047 - val_loss: 0.4274 - val_acc: 0.7953\n",
      "Epoch 142/150\n",
      "768/768 [==============================] - 0s - loss: 0.4137 - acc: 0.8021 - val_loss: 0.4116 - val_acc: 0.7992\n",
      "Epoch 143/150\n",
      "768/768 [==============================] - 0s - loss: 0.4017 - acc: 0.8190 - val_loss: 0.4195 - val_acc: 0.77950.\n",
      "Epoch 144/150\n",
      "768/768 [==============================] - 0s - loss: 0.3924 - acc: 0.8047 - val_loss: 0.4168 - val_acc: 0.7874\n",
      "Epoch 145/150\n",
      "768/768 [==============================] - 0s - loss: 0.3998 - acc: 0.8112 - val_loss: 0.4119 - val_acc: 0.8189\n",
      "Epoch 146/150\n",
      "768/768 [==============================] - 0s - loss: 0.3902 - acc: 0.8164 - val_loss: 0.4403 - val_acc: 0.8031\n",
      "Epoch 147/150\n",
      "768/768 [==============================] - 0s - loss: 0.3961 - acc: 0.8151 - val_loss: 0.4555 - val_acc: 0.7835\n",
      "Epoch 148/150\n",
      "768/768 [==============================] - 0s - loss: 0.3998 - acc: 0.7969 - val_loss: 0.4227 - val_acc: 0.8031\n",
      "Epoch 149/150\n",
      "768/768 [==============================] - 0s - loss: 0.4119 - acc: 0.8125 - val_loss: 0.4194 - val_acc: 0.7992\n",
      "Epoch 150/150\n",
      "768/768 [==============================] - 0s - loss: 0.4036 - acc: 0.8125 - val_loss: 0.4267 - val_acc: 0.8031\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1e74656a550>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X,Y,validation_data=[X_test,y_test], nb_epoch=150, batch_size=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import StratifiedKFold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "StratifiedKFold(n_splits=10, random_state=7, shuffle=True)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kfold=StratifiedKFold(n_splits=10,shuffle=True,random_state=seed)\n",
    "kfold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "acc: 71.43%\n",
      "acc: 72.73%\n",
      "acc: 74.03%\n",
      "acc: 75.32%\n",
      "acc: 80.52%\n",
      "acc: 74.03%\n",
      "acc: 71.43%\n",
      "acc: 70.13%\n",
      "acc: 73.68%\n",
      "acc: 76.32%\n",
      "73.9610391855 2.82009561385\n"
     ]
    }
   ],
   "source": [
    "cvscores=[]\n",
    "for train , test in kfold.split(X,Y):\n",
    "    model=Sequential()\n",
    "    model.add(Dense(12,input_dim=8,init='uniform',activation='relu'))\n",
    "    model.add(Dense(8,init='uniform',activation='relu'))\n",
    "    model.add(Dense(1,init='uniform',activation='sigmoid'))\n",
    "    model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])\n",
    "    model.fit(X[train],Y[train],nb_epoch=150,batch_size=10,verbose=0)\n",
    "    scores=model.evaluate(X[test],Y[test],verbose=0)\n",
    "    print(\"%s: %.2f%%\" % (model.metrics_names[1],scores[1]*100))\n",
    "    cvscores.append(scores[1]*100)\n",
    "print(np.mean(cvscores),np.std(cvscores))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.wrappers.scikit_learn import KerasClassifier\n",
    "from sklearn.model_selection import cross_val_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def create_model():\n",
    "    model=Sequential()\n",
    "    model.add(Dense(12,input_dim=8,init='uniform',activation='relu'))\n",
    "    model.add(Dense(8,init='uniform',activation='relu'))\n",
    "    model.add(Dense(1,init='uniform',activation='sigmoid'))\n",
    "    model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.746172248583\n"
     ]
    }
   ],
   "source": [
    "model=KerasClassifier(build_fn=create_model,nb_epoch=150,batch_size=10,verbose=0)\n",
    "kfold=StratifiedKFold(n_splits=10,shuffle=True,random_state=seed)\n",
    "results=cross_val_score(model,X,Y,cv=kfold)\n",
    "print(results.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}