Zurich_data_time_series.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "# LSTM for international airline passengers problem with memory\n",
    "import numpy\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas\n",
    "import math\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import LSTM\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# convert an array of values into a dataset matrix\n",
    "def create_dataset(dataset, look_back=1):\n",
    "    dataX, dataY = [], []\n",
    "    for i in range(len(dataset)-look_back-1):\n",
    "        a = dataset[i:(i+look_back), 0]\n",
    "        dataX.append(a)\n",
    "        dataY.append(dataset[i + look_back, 0])\n",
    "    return numpy.array(dataX), numpy.array(dataY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# fix random seed for reproducibility\n",
    "numpy.random.seed(7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# load the dataset\n",
    "dataframe = pandas.read_csv('train.csv', usecols=[1], engine='python', skipfooter=3)\n",
    "dataset = dataframe.values\n",
    "dataset = dataset.astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Zuerich</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>62.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>70.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>55.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>85.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Zuerich\n",
       "0     58.0\n",
       "1     62.6\n",
       "2     70.0\n",
       "3     55.7\n",
       "4     85.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataframe.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# normalize the dataset\n",
    "scaler = MinMaxScaler(feature_range=(0, 1))\n",
    "dataset = scaler.fit_transform(dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# split into train and test sets\n",
    "train_size = int(len(dataset) * 0.67)\n",
    "test_size = len(dataset) - train_size\n",
    "train, test = dataset[0:train_size,:], dataset[train_size:len(dataset),:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# reshape into X=t and Y=t+1\n",
    "look_back = 3\n",
    "trainX, trainY = create_dataset(train, look_back)\n",
    "testX, testY = create_dataset(test, look_back)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# reshape input to be [samples, time steps, features]\n",
    "trainX = numpy.reshape(trainX, (trainX.shape[0], trainX.shape[1], 1))\n",
    "testX = numpy.reshape(testX, (testX.shape[0], testX.shape[1], 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# create and fit the LSTM network\n",
    "batch_size = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\keras\\models.py:826: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`.\n",
      "  warnings.warn('The `nb_epoch` argument in `fit` '\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(LSTM(4, batch_input_shape=(batch_size, look_back, 1), stateful=True))\n",
    "model.add(Dense(1))\n",
    "model.compile(loss='mean_squared_error', optimizer='adam')\n",
    "for i in range(100):\n",
    "    model.fit(trainX, trainY, nb_epoch=1, batch_size=batch_size, verbose=0, shuffle=False)\n",
    "    model.reset_states()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# make predictions\n",
    "trainPredict = model.predict(trainX, batch_size=batch_size)\n",
    "model.reset_states()\n",
    "testPredict = model.predict(testX, batch_size=batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# invert predictions\n",
    "trainPredict = scaler.inverse_transform(trainPredict)\n",
    "trainY = scaler.inverse_transform([trainY])\n",
    "testPredict = scaler.inverse_transform(testPredict)\n",
    "testY = scaler.inverse_transform([testY])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Score: 15.65 RMSE\n",
      "Test Score: 14.28 RMSE\n"
     ]
    }
   ],
   "source": [
    "# calculate root mean squared error\n",
    "trainScore = math.sqrt(mean_squared_error(trainY[0], trainPredict[:,0]))\n",
    "print('Train Score: %.2f RMSE' % (trainScore))\n",
    "testScore = math.sqrt(mean_squared_error(testY[0], testPredict[:,0]))\n",
    "print('Test Score: %.2f RMSE' % (testScore))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# shift train predictions for plotting\n",
    "trainPredictPlot = numpy.empty_like(dataset)\n",
    "trainPredictPlot[:, :] = numpy.nan\n",
    "trainPredictPlot[look_back:len(trainPredict)+look_back, :] = trainPredict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# shift test predictions for plotting\n",
    "testPredictPlot = numpy.empty_like(dataset)\n",
    "testPredictPlot[:, :] = numpy.nan\n",
    "testPredictPlot[len(trainPredict)+(look_back*2)+1:len(dataset)-1, :] = testPredict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\njlL0CoVCkeX8P5Bex+Xxox26AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x26001614198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot baseline and predictions\n",
    "plt.plot(scaler.inverse_transform(dataset))\n",
    "plt.plot(trainPredictPlot)\n",
    "plt.plot(testPredictPlot)\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}