daily pastebin goal
47%
SHARE
TWEET

Untitled

a guest Nov 19th, 2017 74 Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
  1. {
  2.  "cells": [
  3.   {
  4.    "cell_type": "code",
  5.    "execution_count": 22,
  6.    "metadata": {},
  7.    "outputs": [],
  8.    "source": [
  9.     "import numpy\n",
  10.     "import matplotlib.pyplot as plt\n",
  11.     "from pandas import read_csv\n",
  12.     "import math\n",
  13.     "from keras.models import Sequential\n",
  14.     "from keras.layers import Dense\n",
  15.     "from keras.layers import LSTM\n",
  16.     "from sklearn.preprocessing import MinMaxScaler\n",
  17.     "from sklearn.metrics import mean_squared_error"
  18.    ]
  19.   },
  20.   {
  21.    "cell_type": "code",
  22.    "execution_count": 23,
  23.    "metadata": {
  24.     "scrolled": true
  25.    },
  26.    "outputs": [
  27.     {
  28.      "data": {
  29.       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
  30.       "text/plain": [
  31.        "<matplotlib.figure.Figure at 0x11faf588>"
  32.       ]
  33.      },
  34.      "metadata": {},
  35.      "output_type": "display_data"
  36.     }
  37.    ],
  38.    "source": [
  39.     "# convert an array of values into a dataset matrix\n",
  40.     "def create_dataset(dataset, look_back=1):\n",
  41.     "\tdataX, dataY = [], []\n",
  42.     "\tfor i in range(len(dataset)-look_back-1):\n",
  43.     "\t\ta = dataset[i:(i+look_back), 0]\n",
  44.     "\t\tdataX.append(a)\n",
  45.     "\t\tdataY.append(dataset[i + look_back, 0])\n",
  46.     "\treturn numpy.array(dataX), numpy.array(dataY)\n",
  47.     "\n",
  48.     "# fix random seed for reproducibility\n",
  49.     "numpy.random.seed(7)\n",
  50.     "\n",
  51.     "# load the dataset\n",
  52.     "dataframe = read_csv('excel.csv', usecols=[1], engine='python', skipfooter=3)\n",
  53.     "dataset = dataframe.values\n",
  54.     "dataset = dataset.astype('float32')\n",
  55.     "\n",
  56.     "plt.plot(dataset)\n",
  57.     "plt.show()"
  58.    ]
  59.   },
  60.   {
  61.    "cell_type": "code",
  62.    "execution_count": 24,
  63.    "metadata": {},
  64.    "outputs": [],
  65.    "source": [
  66.     "# normalize the dataset\n",
  67.     "scaler = MinMaxScaler(feature_range=(0, 1))\n",
  68.     "dataset = scaler.fit_transform(dataset)\n",
  69.     "\n",
  70.     "# split into train and test sets\n",
  71.     "train_size = int(len(dataset) * 0.67)\n",
  72.     "test_size = len(dataset) - train_size\n",
  73.     "train, test = dataset[0:train_size,:], dataset[train_size:len(dataset),:]\n",
  74.     "\n",
  75.     "# reshape into X=t and Y=t+1\n",
  76.     "look_back = 3\n",
  77.     "trainX, trainY = create_dataset(train, look_back)\n",
  78.     "testX, testY = create_dataset(test, look_back)\n",
  79.     "\n",
  80.     "# reshape input to be [samples, time steps, features]\n",
  81.     "trainX = numpy.reshape(trainX, (trainX.shape[0], trainX.shape[1], 1))\n",
  82.     "testX = numpy.reshape(testX, (testX.shape[0], testX.shape[1], 1))"
  83.    ]
  84.   },
  85.   {
  86.    "cell_type": "code",
  87.    "execution_count": 25,
  88.    "metadata": {
  89.     "scrolled": true
  90.    },
  91.    "outputs": [
  92.     {
  93.      "name": "stdout",
  94.      "output_type": "stream",
  95.      "text": [
  96.       "Epoch 1/1\n",
  97.       " - 3s - loss: 0.0428\n",
  98.       "Epoch 1/1\n",
  99.       " - 2s - loss: 0.0264\n",
  100.       "Epoch 1/1\n",
  101.       " - 2s - loss: 0.0262\n",
  102.       "Epoch 1/1\n",
  103.       " - 2s - loss: 0.0261\n",
  104.       "Epoch 1/1\n",
  105.       " - 2s - loss: 0.0259\n",
  106.       "Epoch 1/1\n",
  107.       " - 2s - loss: 0.0259\n",
  108.       "Epoch 1/1\n",
  109.       " - 2s - loss: 0.0258\n",
  110.       "Epoch 1/1\n",
  111.       " - 2s - loss: 0.0257\n",
  112.       "Epoch 1/1\n",
  113.       " - 2s - loss: 0.0257\n",
  114.       "Epoch 1/1\n",
  115.       " - 3s - loss: 0.0256\n",
  116.       "Epoch 1/1\n",
  117.       " - 2s - loss: 0.0256\n",
  118.       "Epoch 1/1\n",
  119.       " - 2s - loss: 0.0255\n",
  120.       "Epoch 1/1\n",
  121.       " - 2s - loss: 0.0255\n",
  122.       "Epoch 1/1\n",
  123.       " - 2s - loss: 0.0255\n",
  124.       "Epoch 1/1\n",
  125.       " - 2s - loss: 0.0255\n",
  126.       "Epoch 1/1\n",
  127.       " - 2s - loss: 0.0254\n",
  128.       "Epoch 1/1\n",
  129.       " - 2s - loss: 0.0254\n",
  130.       "Epoch 1/1\n",
  131.       " - 2s - loss: 0.0254\n",
  132.       "Epoch 1/1\n",
  133.       " - 2s - loss: 0.0254\n",
  134.       "Epoch 1/1\n",
  135.       " - 2s - loss: 0.0254\n",
  136.       "Epoch 1/1\n",
  137.       " - 2s - loss: 0.0253\n",
  138.       "Epoch 1/1\n",
  139.       " - 2s - loss: 0.0253\n",
  140.       "Epoch 1/1\n",
  141.       " - 2s - loss: 0.0253\n",
  142.       "Epoch 1/1\n",
  143.       " - 2s - loss: 0.0253\n",
  144.       "Epoch 1/1\n",
  145.       " - 2s - loss: 0.0253\n",
  146.       "Epoch 1/1\n",
  147.       " - 2s - loss: 0.0253\n",
  148.       "Epoch 1/1\n",
  149.       " - 2s - loss: 0.0253\n",
  150.       "Epoch 1/1\n",
  151.       " - 2s - loss: 0.0253\n",
  152.       "Epoch 1/1\n",
  153.       " - 2s - loss: 0.0253\n",
  154.       "Epoch 1/1\n",
  155.       " - 2s - loss: 0.0252\n",
  156.       "Epoch 1/1\n",
  157.       " - 2s - loss: 0.0252\n",
  158.       "Epoch 1/1\n",
  159.       " - 2s - loss: 0.0252\n",
  160.       "Epoch 1/1\n",
  161.       " - 2s - loss: 0.0252\n",
  162.       "Epoch 1/1\n",
  163.       " - 2s - loss: 0.0252\n",
  164.       "Epoch 1/1\n",
  165.       " - 2s - loss: 0.0252\n",
  166.       "Epoch 1/1\n",
  167.       " - 2s - loss: 0.0252\n",
  168.       "Epoch 1/1\n",
  169.       " - 2s - loss: 0.0252\n",
  170.       "Epoch 1/1\n",
  171.       " - 2s - loss: 0.0252\n",
  172.       "Epoch 1/1\n",
  173.       " - 2s - loss: 0.0252\n",
  174.       "Epoch 1/1\n",
  175.       " - 2s - loss: 0.0252\n",
  176.       "Epoch 1/1\n",
  177.       " - 2s - loss: 0.0252\n",
  178.       "Epoch 1/1\n",
  179.       " - 2s - loss: 0.0252\n",
  180.       "Epoch 1/1\n",
  181.       " - 2s - loss: 0.0252\n",
  182.       "Epoch 1/1\n",
  183.       " - 2s - loss: 0.0252\n",
  184.       "Epoch 1/1\n",
  185.       " - 2s - loss: 0.0252\n",
  186.       "Epoch 1/1\n",
  187.       " - 2s - loss: 0.0252\n",
  188.       "Epoch 1/1\n",
  189.       " - 2s - loss: 0.0252\n",
  190.       "Epoch 1/1\n",
  191.       " - 2s - loss: 0.0252\n",
  192.       "Epoch 1/1\n",
  193.       " - 2s - loss: 0.0252\n",
  194.       "Epoch 1/1\n",
  195.       " - 2s - loss: 0.0252\n",
  196.       "Epoch 1/1\n",
  197.       " - 2s - loss: 0.0252\n",
  198.       "Epoch 1/1\n",
  199.       " - 2s - loss: 0.0252\n",
  200.       "Epoch 1/1\n",
  201.       " - 2s - loss: 0.0252\n",
  202.       "Epoch 1/1\n",
  203.       " - 2s - loss: 0.0252\n",
  204.       "Epoch 1/1\n",
  205.       " - 2s - loss: 0.0252\n",
  206.       "Epoch 1/1\n",
  207.       " - 2s - loss: 0.0251\n",
  208.       "Epoch 1/1\n",
  209.       " - 2s - loss: 0.0251\n",
  210.       "Epoch 1/1\n",
  211.       " - 2s - loss: 0.0251\n",
  212.       "Epoch 1/1\n",
  213.       " - 2s - loss: 0.0251\n",
  214.       "Epoch 1/1\n",
  215.       " - 2s - loss: 0.0251\n",
  216.       "Epoch 1/1\n",
  217.       " - 2s - loss: 0.0251\n",
  218.       "Epoch 1/1\n",
  219.       " - 2s - loss: 0.0251\n",
  220.       "Epoch 1/1\n",
  221.       " - 2s - loss: 0.0251\n",
  222.       "Epoch 1/1\n",
  223.       " - 2s - loss: 0.0251\n",
  224.       "Epoch 1/1\n",
  225.       " - 2s - loss: 0.0251\n",
  226.       "Epoch 1/1\n",
  227.       " - 2s - loss: 0.0251\n",
  228.       "Epoch 1/1\n",
  229.       " - 2s - loss: 0.0251\n",
  230.       "Epoch 1/1\n",
  231.       " - 2s - loss: 0.0251\n",
  232.       "Epoch 1/1\n",
  233.       " - 2s - loss: 0.0251\n",
  234.       "Epoch 1/1\n",
  235.       " - 2s - loss: 0.0251\n",
  236.       "Epoch 1/1\n",
  237.       " - 2s - loss: 0.0251\n",
  238.       "Epoch 1/1\n",
  239.       " - 2s - loss: 0.0251\n",
  240.       "Epoch 1/1\n",
  241.       " - 3s - loss: 0.0251\n",
  242.       "Epoch 1/1\n",
  243.       " - 2s - loss: 0.0251\n",
  244.       "Epoch 1/1\n",
  245.       " - 2s - loss: 0.0251\n",
  246.       "Epoch 1/1\n",
  247.       " - 2s - loss: 0.0251\n",
  248.       "Epoch 1/1\n",
  249.       " - 2s - loss: 0.0251\n",
  250.       "Epoch 1/1\n",
  251.       " - 2s - loss: 0.0251\n",
  252.       "Epoch 1/1\n",
  253.       " - 2s - loss: 0.0251\n",
  254.       "Epoch 1/1\n",
  255.       " - 2s - loss: 0.0251\n",
  256.       "Epoch 1/1\n",
  257.       " - 2s - loss: 0.0251\n",
  258.       "Epoch 1/1\n",
  259.       " - 2s - loss: 0.0251\n",
  260.       "Epoch 1/1\n",
  261.       " - 2s - loss: 0.0251\n",
  262.       "Epoch 1/1\n",
  263.       " - 2s - loss: 0.0251\n",
  264.       "Epoch 1/1\n",
  265.       " - 2s - loss: 0.0251\n",
  266.       "Epoch 1/1\n",
  267.       " - 2s - loss: 0.0251\n",
  268.       "Epoch 1/1\n",
  269.       " - 2s - loss: 0.0251\n",
  270.       "Epoch 1/1\n",
  271.       " - 2s - loss: 0.0251\n",
  272.       "Epoch 1/1\n",
  273.       " - 2s - loss: 0.0251\n",
  274.       "Epoch 1/1\n",
  275.       " - 3s - loss: 0.0251\n",
  276.       "Epoch 1/1\n",
  277.       " - 3s - loss: 0.0251\n",
  278.       "Epoch 1/1\n",
  279.       " - 2s - loss: 0.0251\n",
  280.       "Epoch 1/1\n",
  281.       " - 2s - loss: 0.0251\n",
  282.       "Epoch 1/1\n",
  283.       " - 2s - loss: 0.0251\n",
  284.       "Epoch 1/1\n",
  285.       " - 2s - loss: 0.0250\n",
  286.       "Epoch 1/1\n",
  287.       " - 2s - loss: 0.0250\n",
  288.       "Epoch 1/1\n",
  289.       " - 2s - loss: 0.0250\n",
  290.       "Epoch 1/1\n",
  291.       " - 2s - loss: 0.0250\n",
  292.       "Epoch 1/1\n",
  293.       " - 2s - loss: 0.0250\n",
  294.       "Epoch 1/1\n",
  295.       " - 2s - loss: 0.0250\n"
  296.      ]
  297.     }
  298.    ],
  299.    "source": [
  300.     "# create and fit the LSTM network\n",
  301.     "batch_size = 1\n",
  302.     "model = Sequential()\n",
  303.     "model.add(LSTM(4, batch_input_shape=(batch_size, look_back, 1), stateful=True, return_sequences=True))\n",
  304.     "model.add(LSTM(4, batch_input_shape=(batch_size, look_back, 1), stateful=True))\n",
  305.     "model.add(Dense(1))\n",
  306.     "model.compile(loss='mean_squared_error', optimizer='adam')\n",
  307.     "for i in range(100):\n",
  308.     "\tmodel.fit(trainX, trainY, epochs=1, batch_size=batch_size, verbose=2, shuffle=False)\n",
  309.     "\tmodel.reset_states()"
  310.    ]
  311.   },
  312.   {
  313.    "cell_type": "code",
  314.    "execution_count": 26,
  315.    "metadata": {},
  316.    "outputs": [
  317.     {
  318.      "name": "stdout",
  319.      "output_type": "stream",
  320.      "text": [
  321.       "Train Score: 29.89 RMSE\n",
  322.       "Test Score: 33.18 RMSE\n"
  323.      ]
  324.     }
  325.    ],
  326.    "source": [
  327.     "# make predictions\n",
  328.     "trainPredict = model.predict(trainX, batch_size=batch_size)\n",
  329.     "model.reset_states()\n",
  330.     "testPredict = model.predict(testX, batch_size=batch_size)\n",
  331.     "# invert predictions\n",
  332.     "trainPredict = scaler.inverse_transform(trainPredict)\n",
  333.     "trainY = scaler.inverse_transform([trainY])\n",
  334.     "testPredict = scaler.inverse_transform(testPredict)\n",
  335.     "testY = scaler.inverse_transform([testY])\n",
  336.     "# calculate root mean squared error\n",
  337.     "trainScore = math.sqrt(mean_squared_error(trainY[0], trainPredict[:,0]))\n",
  338.     "print('Train Score: %.2f RMSE' % (trainScore))\n",
  339.     "testScore = math.sqrt(mean_squared_error(testY[0], testPredict[:,0]))\n",
  340.     "print('Test Score: %.2f RMSE' % (testScore))\n",
  341.     "# shift train predictions for plotting\n",
  342.     "trainPredictPlot = numpy.empty_like(dataset)\n",
  343.     "trainPredictPlot[:, :] = numpy.nan\n",
  344.     "trainPredictPlot[look_back:len(trainPredict)+look_back, :] = trainPredict\n",
  345.     "# shift test predictions for plotting\n",
  346.     "testPredictPlot = numpy.empty_like(dataset)\n",
  347.     "testPredictPlot[:, :] = numpy.nan\n",
  348.     "testPredictPlot[len(trainPredict)+(look_back*2)+1:len(dataset)-1, :] = testPredict"
  349.    ]
  350.   },
  351.   {
  352.    "cell_type": "code",
  353.    "execution_count": 27,
  354.    "metadata": {},
  355.    "outputs": [
  356.     {
  357.      "data": {
  358.       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD8CAYAAAB5Pm/hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
  359.       "text/plain": [
  360.        "<matplotlib.figure.Figure at 0x16049b70>"
  361.       ]
  362.      },
  363.      "metadata": {},
  364.      "output_type": "display_data"
  365.     }
  366.    ],
  367.    "source": [
  368.     "# plot baseline and predictions\n",
  369.     "plt.plot(scaler.inverse_transform(dataset))\n",
  370.     "plt.plot(trainPredictPlot)\n",
  371.     "plt.plot(testPredictPlot)\n",
  372.     "plt.show()"
  373.    ]
  374.   },
  375.   {
  376.    "cell_type": "code",
  377.    "execution_count": null,
  378.    "metadata": {
  379.     "collapsed": true
  380.    },
  381.    "outputs": [],
  382.    "source": []
  383.   }
  384.  ],
  385.  "metadata": {
  386.   "kernelspec": {
  387.    "display_name": "Python 3",
  388.    "language": "python",
  389.    "name": "python3"
  390.   },
  391.   "language_info": {
  392.    "codemirror_mode": {
  393.     "name": "ipython",
  394.     "version": 3
  395.    },
  396.    "file_extension": ".py",
  397.    "mimetype": "text/x-python",
  398.    "name": "python",
  399.    "nbconvert_exporter": "python",
  400.    "pygments_lexer": "ipython3",
  401.    "version": "3.6.2"
  402.   }
  403.  },
  404.  "nbformat": 4,
  405.  "nbformat_minor": 2
  406. }
RAW Paste Data
Top