LSTM_V3.0.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python2.7/dist-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from pandas import Series,DataFrame\n",
    "import numpy as np\n",
    "import os\n",
    "from sklearn.metrics import mean_squared_error\n",
    "import datetime\n",
    "import tensorflow as tf\n",
    "from tensorflow.contrib import learn\n",
    "from tensorflow.python.ops import rnn, rnn_cell\n",
    "import matplotlib.pyplot as plt\n",
    "import cPickle\n",
    "import gzip\n",
    "import math\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 提取主力合约数据，已经加入了MA、EMA、MACD等指标"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 提取主力合约数据\n",
    "path = '/home/chocolate/LSTM-source/'\n",
    "data = pd.read_csv(path + 'train_data_addMA.csv')\n",
    "data=data.ix[:,2:]\n",
    "data = data.fillna(0)\n",
    "# data['RaiseDown'] = (data['Latestprice'] - data['First.Latestprice'])/data['First.Latestprice']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Latestprice</th>\n",
       "      <th>MaxPrice</th>\n",
       "      <th>MinPrice</th>\n",
       "      <th>Stockup</th>\n",
       "      <th>Turnover</th>\n",
       "      <th>Position</th>\n",
       "      <th>Volume</th>\n",
       "      <th>First.Latestprice</th>\n",
       "      <th>Last.Buy1price</th>\n",
       "      <th>Last.Buy1quantity</th>\n",
       "      <th>...</th>\n",
       "      <th>Dis_Vol_MA5_26</th>\n",
       "      <th>Vol_EMA_5</th>\n",
       "      <th>Vol_EMA_12</th>\n",
       "      <th>Vol_EMA_26</th>\n",
       "      <th>Dis_Vol_EMA5_26</th>\n",
       "      <th>DIFF_12_26</th>\n",
       "      <th>DEA_12_26</th>\n",
       "      <th>MACD</th>\n",
       "      <th>MeanPrice</th>\n",
       "      <th>RaiseDown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3640.4</td>\n",
       "      <td>3646.0</td>\n",
       "      <td>3640.2</td>\n",
       "      <td>-63</td>\n",
       "      <td>115844340</td>\n",
       "      <td>28315</td>\n",
       "      <td>106</td>\n",
       "      <td>3644.8</td>\n",
       "      <td>3640.2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>106.000000</td>\n",
       "      <td>106.000000</td>\n",
       "      <td>106.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3642.85</td>\n",
       "      <td>-0.002704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3625.8</td>\n",
       "      <td>3641.4</td>\n",
       "      <td>3624.4</td>\n",
       "      <td>-79</td>\n",
       "      <td>228902820</td>\n",
       "      <td>28236</td>\n",
       "      <td>210</td>\n",
       "      <td>3640.4</td>\n",
       "      <td>3625.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>168.400000</td>\n",
       "      <td>162.333333</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>8.400000</td>\n",
       "      <td>-0.327564</td>\n",
       "      <td>-0.181980</td>\n",
       "      <td>-0.291168</td>\n",
       "      <td>3633.00</td>\n",
       "      <td>-0.003331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3618.8</td>\n",
       "      <td>3626.8</td>\n",
       "      <td>3612.6</td>\n",
       "      <td>-110</td>\n",
       "      <td>301745580</td>\n",
       "      <td>28126</td>\n",
       "      <td>278</td>\n",
       "      <td>3625.4</td>\n",
       "      <td>3619.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>220.315789</td>\n",
       "      <td>207.478060</td>\n",
       "      <td>202.396254</td>\n",
       "      <td>17.919535</td>\n",
       "      <td>-0.634451</td>\n",
       "      <td>-0.367419</td>\n",
       "      <td>-0.534064</td>\n",
       "      <td>3620.90</td>\n",
       "      <td>-0.000801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3617.0</td>\n",
       "      <td>3620.2</td>\n",
       "      <td>3616.0</td>\n",
       "      <td>-81</td>\n",
       "      <td>162817740</td>\n",
       "      <td>28045</td>\n",
       "      <td>150</td>\n",
       "      <td>3618.8</td>\n",
       "      <td>3616.8</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>191.107692</td>\n",
       "      <td>189.334483</td>\n",
       "      <td>187.748551</td>\n",
       "      <td>3.359141</td>\n",
       "      <td>-0.823469</td>\n",
       "      <td>-0.521908</td>\n",
       "      <td>-0.603123</td>\n",
       "      <td>3618.00</td>\n",
       "      <td>-0.000373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3618.0</td>\n",
       "      <td>3618.8</td>\n",
       "      <td>3613.0</td>\n",
       "      <td>-66</td>\n",
       "      <td>143205060</td>\n",
       "      <td>27979</td>\n",
       "      <td>132</td>\n",
       "      <td>3616.8</td>\n",
       "      <td>3618.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>168.417062</td>\n",
       "      <td>173.756909</td>\n",
       "      <td>174.820232</td>\n",
       "      <td>-6.403171</td>\n",
       "      <td>-0.869050</td>\n",
       "      <td>-0.625175</td>\n",
       "      <td>-0.487751</td>\n",
       "      <td>3616.65</td>\n",
       "      <td>0.000843</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3621.0</td>\n",
       "      <td>3621.6</td>\n",
       "      <td>3618.0</td>\n",
       "      <td>-47</td>\n",
       "      <td>111857580</td>\n",
       "      <td>27931</td>\n",
       "      <td>103</td>\n",
       "      <td>3618.2</td>\n",
       "      <td>3621.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>144.512782</td>\n",
       "      <td>156.559257</td>\n",
       "      <td>160.435213</td>\n",
       "      <td>-15.922431</td>\n",
       "      <td>-0.751834</td>\n",
       "      <td>-0.659506</td>\n",
       "      <td>-0.184655</td>\n",
       "      <td>3619.70</td>\n",
       "      <td>-0.000553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>3614.2</td>\n",
       "      <td>3621.4</td>\n",
       "      <td>3614.2</td>\n",
       "      <td>-46</td>\n",
       "      <td>120504420</td>\n",
       "      <td>27885</td>\n",
       "      <td>111</td>\n",
       "      <td>3621.0</td>\n",
       "      <td>3612.8</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>132.647402</td>\n",
       "      <td>146.392868</td>\n",
       "      <td>151.643416</td>\n",
       "      <td>-18.996015</td>\n",
       "      <td>-0.971763</td>\n",
       "      <td>-0.738530</td>\n",
       "      <td>-0.466465</td>\n",
       "      <td>3617.70</td>\n",
       "      <td>-0.002750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3601.2</td>\n",
       "      <td>3615.6</td>\n",
       "      <td>3600.0</td>\n",
       "      <td>-98</td>\n",
       "      <td>372121500</td>\n",
       "      <td>27787</td>\n",
       "      <td>344</td>\n",
       "      <td>3614.2</td>\n",
       "      <td>3601.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>205.958763</td>\n",
       "      <td>187.630431</td>\n",
       "      <td>182.636828</td>\n",
       "      <td>23.321935</td>\n",
       "      <td>-1.721885</td>\n",
       "      <td>-0.974849</td>\n",
       "      <td>-1.494072</td>\n",
       "      <td>3607.75</td>\n",
       "      <td>-0.001289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3602.2</td>\n",
       "      <td>3608.6</td>\n",
       "      <td>3600.4</td>\n",
       "      <td>-68</td>\n",
       "      <td>195712320</td>\n",
       "      <td>27719</td>\n",
       "      <td>181</td>\n",
       "      <td>3601.2</td>\n",
       "      <td>3602.2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>197.416984</td>\n",
       "      <td>186.318696</td>\n",
       "      <td>182.394214</td>\n",
       "      <td>15.022770</td>\n",
       "      <td>-2.165518</td>\n",
       "      <td>-1.249899</td>\n",
       "      <td>-1.831237</td>\n",
       "      <td>3603.10</td>\n",
       "      <td>0.000111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3604.6</td>\n",
       "      <td>3608.4</td>\n",
       "      <td>3599.0</td>\n",
       "      <td>-70</td>\n",
       "      <td>198897300</td>\n",
       "      <td>27649</td>\n",
       "      <td>184</td>\n",
       "      <td>3602.0</td>\n",
       "      <td>3603.4</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>192.865730</td>\n",
       "      <td>185.879304</td>\n",
       "      <td>182.615797</td>\n",
       "      <td>10.249933</td>\n",
       "      <td>-2.325090</td>\n",
       "      <td>-1.490804</td>\n",
       "      <td>-1.668572</td>\n",
       "      <td>3603.50</td>\n",
       "      <td>-0.000139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>3602.4</td>\n",
       "      <td>3606.0</td>\n",
       "      <td>3599.0</td>\n",
       "      <td>-39</td>\n",
       "      <td>127536900</td>\n",
       "      <td>27610</td>\n",
       "      <td>118</td>\n",
       "      <td>3604.6</td>\n",
       "      <td>3602.8</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>167.618604</td>\n",
       "      <td>173.459028</td>\n",
       "      <td>174.235108</td>\n",
       "      <td>-6.616503</td>\n",
       "      <td>-2.524999</td>\n",
       "      <td>-1.717080</td>\n",
       "      <td>-1.615838</td>\n",
       "      <td>3603.00</td>\n",
       "      <td>0.000375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>3605.2</td>\n",
       "      <td>3607.6</td>\n",
       "      <td>3602.2</td>\n",
       "      <td>-28</td>\n",
       "      <td>83269080</td>\n",
       "      <td>27582</td>\n",
       "      <td>77</td>\n",
       "      <td>3602.4</td>\n",
       "      <td>3603.4</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>137.177785</td>\n",
       "      <td>156.308929</td>\n",
       "      <td>162.288243</td>\n",
       "      <td>-25.110458</td>\n",
       "      <td>-2.483403</td>\n",
       "      <td>-1.881654</td>\n",
       "      <td>-1.203498</td>\n",
       "      <td>3604.35</td>\n",
       "      <td>-0.000014</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>3605.8</td>\n",
       "      <td>3605.8</td>\n",
       "      <td>3602.2</td>\n",
       "      <td>-9</td>\n",
       "      <td>65952540</td>\n",
       "      <td>27573</td>\n",
       "      <td>61</td>\n",
       "      <td>3603.4</td>\n",
       "      <td>3604.6</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>111.654043</td>\n",
       "      <td>139.759679</td>\n",
       "      <td>150.422343</td>\n",
       "      <td>-38.768300</td>\n",
       "      <td>-2.385718</td>\n",
       "      <td>-1.988332</td>\n",
       "      <td>-0.794773</td>\n",
       "      <td>3604.30</td>\n",
       "      <td>0.000166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>3605.0</td>\n",
       "      <td>3606.0</td>\n",
       "      <td>3602.8</td>\n",
       "      <td>-13</td>\n",
       "      <td>50825100</td>\n",
       "      <td>27560</td>\n",
       "      <td>47</td>\n",
       "      <td>3605.8</td>\n",
       "      <td>3604.2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>90.028618</td>\n",
       "      <td>123.965667</td>\n",
       "      <td>138.806783</td>\n",
       "      <td>-48.778166</td>\n",
       "      <td>-2.327979</td>\n",
       "      <td>-2.059386</td>\n",
       "      <td>-0.537186</td>\n",
       "      <td>3604.90</td>\n",
       "      <td>0.000055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>3606.0</td>\n",
       "      <td>3606.0</td>\n",
       "      <td>3604.2</td>\n",
       "      <td>-27</td>\n",
       "      <td>78953280</td>\n",
       "      <td>27533</td>\n",
       "      <td>73</td>\n",
       "      <td>3604.2</td>\n",
       "      <td>3605.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>84.339420</td>\n",
       "      <td>115.428045</td>\n",
       "      <td>131.688101</td>\n",
       "      <td>-47.348682</td>\n",
       "      <td>-2.195391</td>\n",
       "      <td>-2.087579</td>\n",
       "      <td>-0.215624</td>\n",
       "      <td>3605.10</td>\n",
       "      <td>0.000610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>3608.8</td>\n",
       "      <td>3609.0</td>\n",
       "      <td>3605.4</td>\n",
       "      <td>-23</td>\n",
       "      <td>102802020</td>\n",
       "      <td>27510</td>\n",
       "      <td>95</td>\n",
       "      <td>3606.0</td>\n",
       "      <td>3608.6</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>87.898365</td>\n",
       "      <td>112.052150</td>\n",
       "      <td>127.850225</td>\n",
       "      <td>-39.951860</td>\n",
       "      <td>-1.897371</td>\n",
       "      <td>-2.048436</td>\n",
       "      <td>0.302129</td>\n",
       "      <td>3607.30</td>\n",
       "      <td>0.000097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>3607.0</td>\n",
       "      <td>3609.2</td>\n",
       "      <td>3605.4</td>\n",
       "      <td>-19</td>\n",
       "      <td>60597360</td>\n",
       "      <td>27491</td>\n",
       "      <td>56</td>\n",
       "      <td>3609.0</td>\n",
       "      <td>3607.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>77.254774</td>\n",
       "      <td>102.893605</td>\n",
       "      <td>120.556799</td>\n",
       "      <td>-43.302026</td>\n",
       "      <td>-1.757935</td>\n",
       "      <td>-1.988997</td>\n",
       "      <td>0.462124</td>\n",
       "      <td>3607.65</td>\n",
       "      <td>-0.000998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>3601.4</td>\n",
       "      <td>3607.0</td>\n",
       "      <td>3601.4</td>\n",
       "      <td>-44</td>\n",
       "      <td>128675580</td>\n",
       "      <td>27447</td>\n",
       "      <td>119</td>\n",
       "      <td>3606.4</td>\n",
       "      <td>3601.4</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>91.179271</td>\n",
       "      <td>105.500394</td>\n",
       "      <td>120.402990</td>\n",
       "      <td>-29.223719</td>\n",
       "      <td>-1.982649</td>\n",
       "      <td>-1.987704</td>\n",
       "      <td>0.010111</td>\n",
       "      <td>3604.05</td>\n",
       "      <td>-0.000499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>3606.8</td>\n",
       "      <td>3606.8</td>\n",
       "      <td>3594.0</td>\n",
       "      <td>-16</td>\n",
       "      <td>176006520</td>\n",
       "      <td>27431</td>\n",
       "      <td>163</td>\n",
       "      <td>3601.4</td>\n",
       "      <td>3604.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>115.130318</td>\n",
       "      <td>114.732720</td>\n",
       "      <td>124.509958</td>\n",
       "      <td>-9.379640</td>\n",
       "      <td>-1.784824</td>\n",
       "      <td>-1.946535</td>\n",
       "      <td>0.323422</td>\n",
       "      <td>3602.25</td>\n",
       "      <td>0.000944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>3607.0</td>\n",
       "      <td>3607.8</td>\n",
       "      <td>3603.4</td>\n",
       "      <td>-4</td>\n",
       "      <td>97367760</td>\n",
       "      <td>27428</td>\n",
       "      <td>90</td>\n",
       "      <td>3604.4</td>\n",
       "      <td>3606.6</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>106.751026</td>\n",
       "      <td>110.788050</td>\n",
       "      <td>121.255407</td>\n",
       "      <td>-14.504382</td>\n",
       "      <td>-1.597776</td>\n",
       "      <td>-1.875970</td>\n",
       "      <td>0.556387</td>\n",
       "      <td>3605.65</td>\n",
       "      <td>0.001137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>3613.0</td>\n",
       "      <td>3613.6</td>\n",
       "      <td>3605.4</td>\n",
       "      <td>-35</td>\n",
       "      <td>123478800</td>\n",
       "      <td>27391</td>\n",
       "      <td>114</td>\n",
       "      <td>3607.0</td>\n",
       "      <td>3612.4</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>109.167835</td>\n",
       "      <td>111.297454</td>\n",
       "      <td>120.584737</td>\n",
       "      <td>-11.416902</td>\n",
       "      <td>-1.037152</td>\n",
       "      <td>-1.706644</td>\n",
       "      <td>1.338984</td>\n",
       "      <td>3609.75</td>\n",
       "      <td>0.000180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>3607.8</td>\n",
       "      <td>3613.6</td>\n",
       "      <td>3607.8</td>\n",
       "      <td>-30</td>\n",
       "      <td>72574440</td>\n",
       "      <td>27361</td>\n",
       "      <td>67</td>\n",
       "      <td>3612.4</td>\n",
       "      <td>3607.8</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>95.110011</td>\n",
       "      <td>104.305246</td>\n",
       "      <td>115.720827</td>\n",
       "      <td>-20.610816</td>\n",
       "      <td>-0.939550</td>\n",
       "      <td>-1.552085</td>\n",
       "      <td>1.225070</td>\n",
       "      <td>3610.40</td>\n",
       "      <td>-0.002839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>3593.8</td>\n",
       "      <td>3607.8</td>\n",
       "      <td>3591.2</td>\n",
       "      <td>-58</td>\n",
       "      <td>201761820</td>\n",
       "      <td>27303</td>\n",
       "      <td>187</td>\n",
       "      <td>3607.8</td>\n",
       "      <td>3592.2</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>125.742737</td>\n",
       "      <td>117.306330</td>\n",
       "      <td>122.084617</td>\n",
       "      <td>3.658120</td>\n",
       "      <td>-1.803848</td>\n",
       "      <td>-1.602737</td>\n",
       "      <td>-0.402222</td>\n",
       "      <td>3600.15</td>\n",
       "      <td>-0.002486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>3589.8</td>\n",
       "      <td>3594.4</td>\n",
       "      <td>3588.4</td>\n",
       "      <td>-39</td>\n",
       "      <td>170214000</td>\n",
       "      <td>27264</td>\n",
       "      <td>158</td>\n",
       "      <td>3592.2</td>\n",
       "      <td>3589.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>136.495797</td>\n",
       "      <td>123.682599</td>\n",
       "      <td>125.243107</td>\n",
       "      <td>11.252689</td>\n",
       "      <td>-2.728825</td>\n",
       "      <td>-1.829023</td>\n",
       "      <td>-1.799605</td>\n",
       "      <td>3591.20</td>\n",
       "      <td>0.000585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>3597.4</td>\n",
       "      <td>3597.8</td>\n",
       "      <td>3589.0</td>\n",
       "      <td>-37</td>\n",
       "      <td>95919060</td>\n",
       "      <td>27227</td>\n",
       "      <td>89</td>\n",
       "      <td>3589.0</td>\n",
       "      <td>3594.8</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>120.663237</td>\n",
       "      <td>118.263609</td>\n",
       "      <td>122.099394</td>\n",
       "      <td>-1.436157</td>\n",
       "      <td>-2.887792</td>\n",
       "      <td>-2.041580</td>\n",
       "      <td>-1.692425</td>\n",
       "      <td>3593.30</td>\n",
       "      <td>0.000153</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>3591.4</td>\n",
       "      <td>3597.4</td>\n",
       "      <td>3589.2</td>\n",
       "      <td>-7</td>\n",
       "      <td>89435100</td>\n",
       "      <td>27220</td>\n",
       "      <td>83</td>\n",
       "      <td>3597.4</td>\n",
       "      <td>3589.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>-13.815385</td>\n",
       "      <td>108.108493</td>\n",
       "      <td>112.767026</td>\n",
       "      <td>118.750345</td>\n",
       "      <td>-10.641852</td>\n",
       "      <td>-3.398996</td>\n",
       "      <td>-2.313886</td>\n",
       "      <td>-2.170221</td>\n",
       "      <td>3593.85</td>\n",
       "      <td>-0.001057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>3588.0</td>\n",
       "      <td>3592.8</td>\n",
       "      <td>3588.0</td>\n",
       "      <td>-13</td>\n",
       "      <td>91545060</td>\n",
       "      <td>27207</td>\n",
       "      <td>85</td>\n",
       "      <td>3591.4</td>\n",
       "      <td>3588.0</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>-9.407692</td>\n",
       "      <td>100.405527</td>\n",
       "      <td>108.447692</td>\n",
       "      <td>115.892563</td>\n",
       "      <td>-15.487036</td>\n",
       "      <td>-3.998578</td>\n",
       "      <td>-2.651641</td>\n",
       "      <td>-2.693873</td>\n",
       "      <td>3590.05</td>\n",
       "      <td>-0.001727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>3580.0</td>\n",
       "      <td>3588.6</td>\n",
       "      <td>3580.0</td>\n",
       "      <td>-4</td>\n",
       "      <td>167710680</td>\n",
       "      <td>27203</td>\n",
       "      <td>156</td>\n",
       "      <td>3586.8</td>\n",
       "      <td>3580.0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>-13.530769</td>\n",
       "      <td>118.937235</td>\n",
       "      <td>115.832122</td>\n",
       "      <td>119.253006</td>\n",
       "      <td>-0.315771</td>\n",
       "      <td>-4.991522</td>\n",
       "      <td>-3.120524</td>\n",
       "      <td>-3.741995</td>\n",
       "      <td>3583.85</td>\n",
       "      <td>-0.002776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>3568.0</td>\n",
       "      <td>3580.2</td>\n",
       "      <td>3567.2</td>\n",
       "      <td>-1</td>\n",
       "      <td>190875540</td>\n",
       "      <td>27202</td>\n",
       "      <td>178</td>\n",
       "      <td>3580.2</td>\n",
       "      <td>3567.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.684615</td>\n",
       "      <td>138.624978</td>\n",
       "      <td>125.472289</td>\n",
       "      <td>124.127839</td>\n",
       "      <td>14.497139</td>\n",
       "      <td>-6.573239</td>\n",
       "      <td>-3.812137</td>\n",
       "      <td>-5.522204</td>\n",
       "      <td>3573.90</td>\n",
       "      <td>-0.001679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>3568.0</td>\n",
       "      <td>3571.4</td>\n",
       "      <td>3565.0</td>\n",
       "      <td>-17</td>\n",
       "      <td>135944700</td>\n",
       "      <td>27185</td>\n",
       "      <td>127</td>\n",
       "      <td>3567.2</td>\n",
       "      <td>3566.6</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>134.749965</td>\n",
       "      <td>125.708897</td>\n",
       "      <td>124.364067</td>\n",
       "      <td>10.385898</td>\n",
       "      <td>-7.731735</td>\n",
       "      <td>-4.597029</td>\n",
       "      <td>-6.269413</td>\n",
       "      <td>3567.90</td>\n",
       "      <td>0.000070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35728</th>\n",
       "      <td>3301.6</td>\n",
       "      <td>3301.6</td>\n",
       "      <td>3299.2</td>\n",
       "      <td>14</td>\n",
       "      <td>30692160</td>\n",
       "      <td>37376</td>\n",
       "      <td>31</td>\n",
       "      <td>3300.8</td>\n",
       "      <td>3300.2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>-11.161538</td>\n",
       "      <td>26.889536</td>\n",
       "      <td>30.047989</td>\n",
       "      <td>33.567610</td>\n",
       "      <td>-6.678075</td>\n",
       "      <td>-0.404311</td>\n",
       "      <td>-0.697110</td>\n",
       "      <td>0.585599</td>\n",
       "      <td>3300.80</td>\n",
       "      <td>0.000409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35729</th>\n",
       "      <td>3302.4</td>\n",
       "      <td>3304.4</td>\n",
       "      <td>3300.2</td>\n",
       "      <td>29</td>\n",
       "      <td>61434300</td>\n",
       "      <td>37405</td>\n",
       "      <td>62</td>\n",
       "      <td>3301.6</td>\n",
       "      <td>3302.4</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.507692</td>\n",
       "      <td>38.593024</td>\n",
       "      <td>34.963683</td>\n",
       "      <td>35.673713</td>\n",
       "      <td>2.919311</td>\n",
       "      <td>-0.271863</td>\n",
       "      <td>-0.612061</td>\n",
       "      <td>0.680396</td>\n",
       "      <td>3302.15</td>\n",
       "      <td>0.000424</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35730</th>\n",
       "      <td>3304.4</td>\n",
       "      <td>3305.0</td>\n",
       "      <td>3302.4</td>\n",
       "      <td>22</td>\n",
       "      <td>42620280</td>\n",
       "      <td>37427</td>\n",
       "      <td>43</td>\n",
       "      <td>3302.4</td>\n",
       "      <td>3303.2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.184615</td>\n",
       "      <td>40.062016</td>\n",
       "      <td>36.200039</td>\n",
       "      <td>36.216401</td>\n",
       "      <td>3.845615</td>\n",
       "      <td>-0.005450</td>\n",
       "      <td>-0.490739</td>\n",
       "      <td>0.970576</td>\n",
       "      <td>3303.55</td>\n",
       "      <td>0.000288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35731</th>\n",
       "      <td>3304.6</td>\n",
       "      <td>3305.6</td>\n",
       "      <td>3303.2</td>\n",
       "      <td>22</td>\n",
       "      <td>75345540</td>\n",
       "      <td>37449</td>\n",
       "      <td>76</td>\n",
       "      <td>3304.6</td>\n",
       "      <td>3304.6</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>7.276923</td>\n",
       "      <td>52.041344</td>\n",
       "      <td>42.323110</td>\n",
       "      <td>39.163334</td>\n",
       "      <td>12.878010</td>\n",
       "      <td>0.219294</td>\n",
       "      <td>-0.348732</td>\n",
       "      <td>1.136052</td>\n",
       "      <td>3304.50</td>\n",
       "      <td>-0.000151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35732</th>\n",
       "      <td>3304.6</td>\n",
       "      <td>3305.4</td>\n",
       "      <td>3301.4</td>\n",
       "      <td>40</td>\n",
       "      <td>85243080</td>\n",
       "      <td>37489</td>\n",
       "      <td>86</td>\n",
       "      <td>3304.6</td>\n",
       "      <td>3304.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>19.292308</td>\n",
       "      <td>63.360896</td>\n",
       "      <td>49.042632</td>\n",
       "      <td>42.632717</td>\n",
       "      <td>20.728179</td>\n",
       "      <td>0.392876</td>\n",
       "      <td>-0.200410</td>\n",
       "      <td>1.186573</td>\n",
       "      <td>3304.00</td>\n",
       "      <td>-0.000030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35733</th>\n",
       "      <td>3303.6</td>\n",
       "      <td>3304.6</td>\n",
       "      <td>3303.2</td>\n",
       "      <td>7</td>\n",
       "      <td>30726180</td>\n",
       "      <td>37496</td>\n",
       "      <td>31</td>\n",
       "      <td>3304.2</td>\n",
       "      <td>3303.6</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>21.407692</td>\n",
       "      <td>52.573931</td>\n",
       "      <td>46.266842</td>\n",
       "      <td>41.771034</td>\n",
       "      <td>10.802896</td>\n",
       "      <td>0.444624</td>\n",
       "      <td>-0.071403</td>\n",
       "      <td>1.032056</td>\n",
       "      <td>3303.90</td>\n",
       "      <td>-0.000030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35734</th>\n",
       "      <td>3304.2</td>\n",
       "      <td>3304.2</td>\n",
       "      <td>3303.2</td>\n",
       "      <td>8</td>\n",
       "      <td>21803700</td>\n",
       "      <td>37504</td>\n",
       "      <td>22</td>\n",
       "      <td>3303.6</td>\n",
       "      <td>3303.2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>13.792308</td>\n",
       "      <td>42.382620</td>\n",
       "      <td>42.533482</td>\n",
       "      <td>40.306513</td>\n",
       "      <td>2.076107</td>\n",
       "      <td>0.527964</td>\n",
       "      <td>0.048470</td>\n",
       "      <td>0.958988</td>\n",
       "      <td>3303.80</td>\n",
       "      <td>0.000636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35735</th>\n",
       "      <td>3307.0</td>\n",
       "      <td>3309.0</td>\n",
       "      <td>3303.4</td>\n",
       "      <td>9</td>\n",
       "      <td>61494420</td>\n",
       "      <td>37513</td>\n",
       "      <td>62</td>\n",
       "      <td>3304.2</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>17.015385</td>\n",
       "      <td>48.921747</td>\n",
       "      <td>45.528331</td>\n",
       "      <td>41.913438</td>\n",
       "      <td>7.008309</td>\n",
       "      <td>0.810604</td>\n",
       "      <td>0.200897</td>\n",
       "      <td>1.219415</td>\n",
       "      <td>3305.90</td>\n",
       "      <td>0.000514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35736</th>\n",
       "      <td>3308.8</td>\n",
       "      <td>3309.0</td>\n",
       "      <td>3305.6</td>\n",
       "      <td>12</td>\n",
       "      <td>47622960</td>\n",
       "      <td>37525</td>\n",
       "      <td>48</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>3307.6</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>10.992308</td>\n",
       "      <td>48.614498</td>\n",
       "      <td>45.908588</td>\n",
       "      <td>42.364294</td>\n",
       "      <td>6.250203</td>\n",
       "      <td>1.166398</td>\n",
       "      <td>0.393997</td>\n",
       "      <td>1.544801</td>\n",
       "      <td>3307.60</td>\n",
       "      <td>0.000363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35737</th>\n",
       "      <td>3309.8</td>\n",
       "      <td>3309.8</td>\n",
       "      <td>3306.8</td>\n",
       "      <td>26</td>\n",
       "      <td>70471860</td>\n",
       "      <td>37551</td>\n",
       "      <td>71</td>\n",
       "      <td>3308.8</td>\n",
       "      <td>3308.6</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>6.376923</td>\n",
       "      <td>56.076332</td>\n",
       "      <td>49.768805</td>\n",
       "      <td>44.485458</td>\n",
       "      <td>11.590874</td>\n",
       "      <td>1.511633</td>\n",
       "      <td>0.617524</td>\n",
       "      <td>1.788218</td>\n",
       "      <td>3308.80</td>\n",
       "      <td>0.000015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35738</th>\n",
       "      <td>3308.0</td>\n",
       "      <td>3309.8</td>\n",
       "      <td>3307.8</td>\n",
       "      <td>8</td>\n",
       "      <td>37720740</td>\n",
       "      <td>37559</td>\n",
       "      <td>38</td>\n",
       "      <td>3309.8</td>\n",
       "      <td>3308.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>7.161538</td>\n",
       "      <td>50.050888</td>\n",
       "      <td>47.958220</td>\n",
       "      <td>44.005054</td>\n",
       "      <td>6.045834</td>\n",
       "      <td>1.621301</td>\n",
       "      <td>0.818280</td>\n",
       "      <td>1.606042</td>\n",
       "      <td>3308.85</td>\n",
       "      <td>-0.000348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35739</th>\n",
       "      <td>3307.2</td>\n",
       "      <td>3308.6</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>-3</td>\n",
       "      <td>19846320</td>\n",
       "      <td>37556</td>\n",
       "      <td>20</td>\n",
       "      <td>3308.0</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>6.953846</td>\n",
       "      <td>40.033925</td>\n",
       "      <td>43.656955</td>\n",
       "      <td>42.226901</td>\n",
       "      <td>-2.192976</td>\n",
       "      <td>1.624928</td>\n",
       "      <td>0.979609</td>\n",
       "      <td>1.290638</td>\n",
       "      <td>3307.70</td>\n",
       "      <td>-0.000151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35740</th>\n",
       "      <td>3306.8</td>\n",
       "      <td>3308.2</td>\n",
       "      <td>3306.6</td>\n",
       "      <td>5</td>\n",
       "      <td>33737580</td>\n",
       "      <td>37561</td>\n",
       "      <td>34</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>3306.6</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1.315385</td>\n",
       "      <td>38.022617</td>\n",
       "      <td>42.171270</td>\n",
       "      <td>41.617501</td>\n",
       "      <td>-3.594884</td>\n",
       "      <td>1.577344</td>\n",
       "      <td>1.099156</td>\n",
       "      <td>0.956375</td>\n",
       "      <td>3307.20</td>\n",
       "      <td>-0.000030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35741</th>\n",
       "      <td>3307.0</td>\n",
       "      <td>3308.0</td>\n",
       "      <td>3306.4</td>\n",
       "      <td>1</td>\n",
       "      <td>38690940</td>\n",
       "      <td>37562</td>\n",
       "      <td>39</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>3306.4</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.984615</td>\n",
       "      <td>38.348411</td>\n",
       "      <td>41.683382</td>\n",
       "      <td>41.423612</td>\n",
       "      <td>-3.075201</td>\n",
       "      <td>1.538042</td>\n",
       "      <td>1.186933</td>\n",
       "      <td>0.702217</td>\n",
       "      <td>3307.10</td>\n",
       "      <td>-0.000121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35742</th>\n",
       "      <td>3307.0</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>3306.2</td>\n",
       "      <td>6</td>\n",
       "      <td>31743660</td>\n",
       "      <td>37568</td>\n",
       "      <td>32</td>\n",
       "      <td>3306.4</td>\n",
       "      <td>3306.4</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>-7.784615</td>\n",
       "      <td>36.232274</td>\n",
       "      <td>40.193631</td>\n",
       "      <td>40.725567</td>\n",
       "      <td>-4.493293</td>\n",
       "      <td>1.489722</td>\n",
       "      <td>1.247491</td>\n",
       "      <td>0.484462</td>\n",
       "      <td>3306.70</td>\n",
       "      <td>0.000151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35743</th>\n",
       "      <td>3307.0</td>\n",
       "      <td>3308.0</td>\n",
       "      <td>3306.6</td>\n",
       "      <td>4</td>\n",
       "      <td>30758820</td>\n",
       "      <td>37572</td>\n",
       "      <td>31</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>3306.2</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>-8.723077</td>\n",
       "      <td>34.488183</td>\n",
       "      <td>38.779226</td>\n",
       "      <td>40.005155</td>\n",
       "      <td>-5.516972</td>\n",
       "      <td>1.434888</td>\n",
       "      <td>1.284970</td>\n",
       "      <td>0.299834</td>\n",
       "      <td>3307.20</td>\n",
       "      <td>-0.000348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35744</th>\n",
       "      <td>3305.2</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>3305.0</td>\n",
       "      <td>10</td>\n",
       "      <td>32726820</td>\n",
       "      <td>37582</td>\n",
       "      <td>33</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>3305.0</td>\n",
       "      <td>6</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.546154</td>\n",
       "      <td>33.992122</td>\n",
       "      <td>37.890114</td>\n",
       "      <td>39.486254</td>\n",
       "      <td>-5.494132</td>\n",
       "      <td>1.231984</td>\n",
       "      <td>1.274373</td>\n",
       "      <td>-0.084778</td>\n",
       "      <td>3306.05</td>\n",
       "      <td>0.000060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35745</th>\n",
       "      <td>3306.6</td>\n",
       "      <td>3307.6</td>\n",
       "      <td>3305.4</td>\n",
       "      <td>9</td>\n",
       "      <td>32736780</td>\n",
       "      <td>37593</td>\n",
       "      <td>33</td>\n",
       "      <td>3305.4</td>\n",
       "      <td>3306.6</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>-5.823077</td>\n",
       "      <td>33.661415</td>\n",
       "      <td>37.137789</td>\n",
       "      <td>39.005791</td>\n",
       "      <td>-5.344376</td>\n",
       "      <td>1.170656</td>\n",
       "      <td>1.253630</td>\n",
       "      <td>-0.165948</td>\n",
       "      <td>3306.25</td>\n",
       "      <td>-0.000106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35746</th>\n",
       "      <td>3306.0</td>\n",
       "      <td>3306.6</td>\n",
       "      <td>3305.2</td>\n",
       "      <td>2</td>\n",
       "      <td>27770820</td>\n",
       "      <td>37584</td>\n",
       "      <td>28</td>\n",
       "      <td>3305.8</td>\n",
       "      <td>3305.4</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>-7.100000</td>\n",
       "      <td>31.774276</td>\n",
       "      <td>35.731975</td>\n",
       "      <td>38.190547</td>\n",
       "      <td>-6.416271</td>\n",
       "      <td>1.061403</td>\n",
       "      <td>1.215184</td>\n",
       "      <td>-0.307564</td>\n",
       "      <td>3305.90</td>\n",
       "      <td>0.000151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35747</th>\n",
       "      <td>3305.8</td>\n",
       "      <td>3307.4</td>\n",
       "      <td>3305.8</td>\n",
       "      <td>13</td>\n",
       "      <td>38687640</td>\n",
       "      <td>37606</td>\n",
       "      <td>39</td>\n",
       "      <td>3306.6</td>\n",
       "      <td>3306.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>-6.392308</td>\n",
       "      <td>34.182851</td>\n",
       "      <td>36.234748</td>\n",
       "      <td>38.250507</td>\n",
       "      <td>-4.067656</td>\n",
       "      <td>0.947755</td>\n",
       "      <td>1.161698</td>\n",
       "      <td>-0.427887</td>\n",
       "      <td>3306.40</td>\n",
       "      <td>-0.000030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35748</th>\n",
       "      <td>3307.0</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>3305.4</td>\n",
       "      <td>17</td>\n",
       "      <td>47606100</td>\n",
       "      <td>37623</td>\n",
       "      <td>48</td>\n",
       "      <td>3305.8</td>\n",
       "      <td>3306.4</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.761538</td>\n",
       "      <td>38.788567</td>\n",
       "      <td>38.044787</td>\n",
       "      <td>38.972691</td>\n",
       "      <td>-0.184124</td>\n",
       "      <td>0.943641</td>\n",
       "      <td>1.118087</td>\n",
       "      <td>-0.348892</td>\n",
       "      <td>3306.30</td>\n",
       "      <td>0.000257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35749</th>\n",
       "      <td>3307.2</td>\n",
       "      <td>3308.0</td>\n",
       "      <td>3306.4</td>\n",
       "      <td>16</td>\n",
       "      <td>47622480</td>\n",
       "      <td>37639</td>\n",
       "      <td>48</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.107692</td>\n",
       "      <td>41.859045</td>\n",
       "      <td>39.576358</td>\n",
       "      <td>39.641381</td>\n",
       "      <td>2.217664</td>\n",
       "      <td>0.945618</td>\n",
       "      <td>1.083593</td>\n",
       "      <td>-0.275950</td>\n",
       "      <td>3307.15</td>\n",
       "      <td>-0.000166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35750</th>\n",
       "      <td>3306.2</td>\n",
       "      <td>3307.4</td>\n",
       "      <td>3305.6</td>\n",
       "      <td>27</td>\n",
       "      <td>60509280</td>\n",
       "      <td>37666</td>\n",
       "      <td>61</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>3305.8</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>3.184615</td>\n",
       "      <td>48.239363</td>\n",
       "      <td>42.872303</td>\n",
       "      <td>41.223501</td>\n",
       "      <td>7.015862</td>\n",
       "      <td>0.856619</td>\n",
       "      <td>1.038198</td>\n",
       "      <td>-0.363159</td>\n",
       "      <td>3306.60</td>\n",
       "      <td>-0.000076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35751</th>\n",
       "      <td>3306.2</td>\n",
       "      <td>3307.0</td>\n",
       "      <td>3306.0</td>\n",
       "      <td>13</td>\n",
       "      <td>39678300</td>\n",
       "      <td>37679</td>\n",
       "      <td>40</td>\n",
       "      <td>3306.2</td>\n",
       "      <td>3306.0</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>4.853846</td>\n",
       "      <td>45.492909</td>\n",
       "      <td>42.430410</td>\n",
       "      <td>41.132871</td>\n",
       "      <td>4.360038</td>\n",
       "      <td>0.777128</td>\n",
       "      <td>0.985984</td>\n",
       "      <td>-0.417712</td>\n",
       "      <td>3306.35</td>\n",
       "      <td>0.000121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35752</th>\n",
       "      <td>3306.6</td>\n",
       "      <td>3307.4</td>\n",
       "      <td>3305.8</td>\n",
       "      <td>6</td>\n",
       "      <td>46622640</td>\n",
       "      <td>37685</td>\n",
       "      <td>47</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>3306.2</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>5.646154</td>\n",
       "      <td>45.995273</td>\n",
       "      <td>43.133424</td>\n",
       "      <td>41.567473</td>\n",
       "      <td>4.427799</td>\n",
       "      <td>0.737902</td>\n",
       "      <td>0.936368</td>\n",
       "      <td>-0.396932</td>\n",
       "      <td>3306.75</td>\n",
       "      <td>-0.000091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35753</th>\n",
       "      <td>3306.6</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>3305.8</td>\n",
       "      <td>4</td>\n",
       "      <td>38684280</td>\n",
       "      <td>37689</td>\n",
       "      <td>39</td>\n",
       "      <td>3306.2</td>\n",
       "      <td>3306.6</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>3.076923</td>\n",
       "      <td>43.663515</td>\n",
       "      <td>42.497513</td>\n",
       "      <td>41.377290</td>\n",
       "      <td>2.286225</td>\n",
       "      <td>0.698759</td>\n",
       "      <td>0.888846</td>\n",
       "      <td>-0.380173</td>\n",
       "      <td>3306.45</td>\n",
       "      <td>-0.000015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35754</th>\n",
       "      <td>3306.2</td>\n",
       "      <td>3307.2</td>\n",
       "      <td>3305.6</td>\n",
       "      <td>7</td>\n",
       "      <td>51579780</td>\n",
       "      <td>37696</td>\n",
       "      <td>52</td>\n",
       "      <td>3306.6</td>\n",
       "      <td>3305.8</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>3.069231</td>\n",
       "      <td>46.442343</td>\n",
       "      <td>43.959434</td>\n",
       "      <td>42.164157</td>\n",
       "      <td>4.278186</td>\n",
       "      <td>0.628221</td>\n",
       "      <td>0.836721</td>\n",
       "      <td>-0.417001</td>\n",
       "      <td>3306.40</td>\n",
       "      <td>-0.000348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35755</th>\n",
       "      <td>3304.2</td>\n",
       "      <td>3306.8</td>\n",
       "      <td>3304.2</td>\n",
       "      <td>29</td>\n",
       "      <td>80316660</td>\n",
       "      <td>37725</td>\n",
       "      <td>81</td>\n",
       "      <td>3305.8</td>\n",
       "      <td>3304.2</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>6.338462</td>\n",
       "      <td>57.961562</td>\n",
       "      <td>49.657983</td>\n",
       "      <td>45.040887</td>\n",
       "      <td>12.920676</td>\n",
       "      <td>0.406252</td>\n",
       "      <td>0.750627</td>\n",
       "      <td>-0.688751</td>\n",
       "      <td>3305.25</td>\n",
       "      <td>-0.000212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35756</th>\n",
       "      <td>3304.8</td>\n",
       "      <td>3305.4</td>\n",
       "      <td>3303.8</td>\n",
       "      <td>11</td>\n",
       "      <td>60475860</td>\n",
       "      <td>37736</td>\n",
       "      <td>61</td>\n",
       "      <td>3304.2</td>\n",
       "      <td>3304.8</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>9.846154</td>\n",
       "      <td>58.974375</td>\n",
       "      <td>51.402908</td>\n",
       "      <td>46.223043</td>\n",
       "      <td>12.751332</td>\n",
       "      <td>0.275578</td>\n",
       "      <td>0.655617</td>\n",
       "      <td>-0.760078</td>\n",
       "      <td>3304.55</td>\n",
       "      <td>0.000151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35757</th>\n",
       "      <td>3305.0</td>\n",
       "      <td>3306.2</td>\n",
       "      <td>3304.0</td>\n",
       "      <td>12</td>\n",
       "      <td>74366280</td>\n",
       "      <td>37748</td>\n",
       "      <td>75</td>\n",
       "      <td>3305.0</td>\n",
       "      <td>3305.0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>15.484615</td>\n",
       "      <td>64.316250</td>\n",
       "      <td>55.033230</td>\n",
       "      <td>48.354670</td>\n",
       "      <td>15.961580</td>\n",
       "      <td>0.186013</td>\n",
       "      <td>0.561696</td>\n",
       "      <td>-0.751368</td>\n",
       "      <td>3305.05</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>35758 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Latestprice  MaxPrice  MinPrice  Stockup   Turnover  Position  Volume  \\\n",
       "0           3640.4    3646.0    3640.2      -63  115844340     28315     106   \n",
       "1           3625.8    3641.4    3624.4      -79  228902820     28236     210   \n",
       "2           3618.8    3626.8    3612.6     -110  301745580     28126     278   \n",
       "3           3617.0    3620.2    3616.0      -81  162817740     28045     150   \n",
       "4           3618.0    3618.8    3613.0      -66  143205060     27979     132   \n",
       "5           3621.0    3621.6    3618.0      -47  111857580     27931     103   \n",
       "6           3614.2    3621.4    3614.2      -46  120504420     27885     111   \n",
       "7           3601.2    3615.6    3600.0      -98  372121500     27787     344   \n",
       "8           3602.2    3608.6    3600.4      -68  195712320     27719     181   \n",
       "9           3604.6    3608.4    3599.0      -70  198897300     27649     184   \n",
       "10          3602.4    3606.0    3599.0      -39  127536900     27610     118   \n",
       "11          3605.2    3607.6    3602.2      -28   83269080     27582      77   \n",
       "12          3605.8    3605.8    3602.2       -9   65952540     27573      61   \n",
       "13          3605.0    3606.0    3602.8      -13   50825100     27560      47   \n",
       "14          3606.0    3606.0    3604.2      -27   78953280     27533      73   \n",
       "15          3608.8    3609.0    3605.4      -23  102802020     27510      95   \n",
       "16          3607.0    3609.2    3605.4      -19   60597360     27491      56   \n",
       "17          3601.4    3607.0    3601.4      -44  128675580     27447     119   \n",
       "18          3606.8    3606.8    3594.0      -16  176006520     27431     163   \n",
       "19          3607.0    3607.8    3603.4       -4   97367760     27428      90   \n",
       "20          3613.0    3613.6    3605.4      -35  123478800     27391     114   \n",
       "21          3607.8    3613.6    3607.8      -30   72574440     27361      67   \n",
       "22          3593.8    3607.8    3591.2      -58  201761820     27303     187   \n",
       "23          3589.8    3594.4    3588.4      -39  170214000     27264     158   \n",
       "24          3597.4    3597.8    3589.0      -37   95919060     27227      89   \n",
       "25          3591.4    3597.4    3589.2       -7   89435100     27220      83   \n",
       "26          3588.0    3592.8    3588.0      -13   91545060     27207      85   \n",
       "27          3580.0    3588.6    3580.0       -4  167710680     27203     156   \n",
       "28          3568.0    3580.2    3567.2       -1  190875540     27202     178   \n",
       "29          3568.0    3571.4    3565.0      -17  135944700     27185     127   \n",
       "...            ...       ...       ...      ...        ...       ...     ...   \n",
       "35728       3301.6    3301.6    3299.2       14   30692160     37376      31   \n",
       "35729       3302.4    3304.4    3300.2       29   61434300     37405      62   \n",
       "35730       3304.4    3305.0    3302.4       22   42620280     37427      43   \n",
       "35731       3304.6    3305.6    3303.2       22   75345540     37449      76   \n",
       "35732       3304.6    3305.4    3301.4       40   85243080     37489      86   \n",
       "35733       3303.6    3304.6    3303.2        7   30726180     37496      31   \n",
       "35734       3304.2    3304.2    3303.2        8   21803700     37504      22   \n",
       "35735       3307.0    3309.0    3303.4        9   61494420     37513      62   \n",
       "35736       3308.8    3309.0    3305.6       12   47622960     37525      48   \n",
       "35737       3309.8    3309.8    3306.8       26   70471860     37551      71   \n",
       "35738       3308.0    3309.8    3307.8        8   37720740     37559      38   \n",
       "35739       3307.2    3308.6    3307.0       -3   19846320     37556      20   \n",
       "35740       3306.8    3308.2    3306.6        5   33737580     37561      34   \n",
       "35741       3307.0    3308.0    3306.4        1   38690940     37562      39   \n",
       "35742       3307.0    3307.2    3306.2        6   31743660     37568      32   \n",
       "35743       3307.0    3308.0    3306.6        4   30758820     37572      31   \n",
       "35744       3305.2    3307.0    3305.0       10   32726820     37582      33   \n",
       "35745       3306.6    3307.6    3305.4        9   32736780     37593      33   \n",
       "35746       3306.0    3306.6    3305.2        2   27770820     37584      28   \n",
       "35747       3305.8    3307.4    3305.8       13   38687640     37606      39   \n",
       "35748       3307.0    3307.0    3305.4       17   47606100     37623      48   \n",
       "35749       3307.2    3308.0    3306.4       16   47622480     37639      48   \n",
       "35750       3306.2    3307.4    3305.6       27   60509280     37666      61   \n",
       "35751       3306.2    3307.0    3306.0       13   39678300     37679      40   \n",
       "35752       3306.6    3307.4    3305.8        6   46622640     37685      47   \n",
       "35753       3306.6    3307.2    3305.8        4   38684280     37689      39   \n",
       "35754       3306.2    3307.2    3305.6        7   51579780     37696      52   \n",
       "35755       3304.2    3306.8    3304.2       29   80316660     37725      81   \n",
       "35756       3304.8    3305.4    3303.8       11   60475860     37736      61   \n",
       "35757       3305.0    3306.2    3304.0       12   74366280     37748      75   \n",
       "\n",
       "       First.Latestprice  Last.Buy1price  Last.Buy1quantity    ...      \\\n",
       "0                 3644.8          3640.2                  2    ...       \n",
       "1                 3640.4          3625.0                  4    ...       \n",
       "2                 3625.4          3619.2                  1    ...       \n",
       "3                 3618.8          3616.8                  5    ...       \n",
       "4                 3616.8          3618.0                  2    ...       \n",
       "5                 3618.2          3621.0                  1    ...       \n",
       "6                 3621.0          3612.8                  2    ...       \n",
       "7                 3614.2          3601.0                  1    ...       \n",
       "8                 3601.2          3602.2                  2    ...       \n",
       "9                 3602.0          3603.4                  1    ...       \n",
       "10                3604.6          3602.8                  1    ...       \n",
       "11                3602.4          3603.4                  2    ...       \n",
       "12                3603.4          3604.6                  1    ...       \n",
       "13                3605.8          3604.2                  5    ...       \n",
       "14                3604.2          3605.2                  1    ...       \n",
       "15                3606.0          3608.6                  1    ...       \n",
       "16                3609.0          3607.0                  1    ...       \n",
       "17                3606.4          3601.4                  3    ...       \n",
       "18                3601.4          3604.2                  1    ...       \n",
       "19                3604.4          3606.6                  3    ...       \n",
       "20                3607.0          3612.4                  1    ...       \n",
       "21                3612.4          3607.8                  3    ...       \n",
       "22                3607.8          3592.2                  3    ...       \n",
       "23                3592.2          3589.2                  1    ...       \n",
       "24                3589.0          3594.8                  1    ...       \n",
       "25                3597.4          3589.2                  1    ...       \n",
       "26                3591.4          3588.0                  5    ...       \n",
       "27                3586.8          3580.0                  2    ...       \n",
       "28                3580.2          3567.2                  1    ...       \n",
       "29                3567.2          3566.6                  1    ...       \n",
       "...                  ...             ...                ...    ...       \n",
       "35728             3300.8          3300.2                  2    ...       \n",
       "35729             3301.6          3302.4                  1    ...       \n",
       "35730             3302.4          3303.2                  2    ...       \n",
       "35731             3304.6          3304.6                  2    ...       \n",
       "35732             3304.6          3304.2                  1    ...       \n",
       "35733             3304.2          3303.6                  2    ...       \n",
       "35734             3303.6          3303.2                  2    ...       \n",
       "35735             3304.2          3307.2                  1    ...       \n",
       "35736             3307.0          3307.6                  3    ...       \n",
       "35737             3308.8          3308.6                  2    ...       \n",
       "35738             3309.8          3308.0                  1    ...       \n",
       "35739             3308.0          3307.0                  1    ...       \n",
       "35740             3307.2          3306.6                  1    ...       \n",
       "35741             3307.0          3306.4                  3    ...       \n",
       "35742             3306.4          3306.4                  2    ...       \n",
       "35743             3307.2          3306.2                  2    ...       \n",
       "35744             3307.0          3305.0                  6    ...       \n",
       "35745             3305.4          3306.6                  1    ...       \n",
       "35746             3305.8          3305.4                  2    ...       \n",
       "35747             3306.6          3306.0                  1    ...       \n",
       "35748             3305.8          3306.4                  2    ...       \n",
       "35749             3307.0          3307.2                  1    ...       \n",
       "35750             3307.2          3305.8                  1    ...       \n",
       "35751             3306.2          3306.0                  4    ...       \n",
       "35752             3307.2          3306.2                  3    ...       \n",
       "35753             3306.2          3306.6                  1    ...       \n",
       "35754             3306.6          3305.8                  2    ...       \n",
       "35755             3305.8          3304.2                  3    ...       \n",
       "35756             3304.2          3304.8                  1    ...       \n",
       "35757             3305.0          3305.0                  1    ...       \n",
       "\n",
       "       Dis_Vol_MA5_26   Vol_EMA_5  Vol_EMA_12  Vol_EMA_26  Dis_Vol_EMA5_26  \\\n",
       "0            0.000000  106.000000  106.000000  106.000000         0.000000   \n",
       "1            0.000000  168.400000  162.333333  160.000000         8.400000   \n",
       "2            0.000000  220.315789  207.478060  202.396254        17.919535   \n",
       "3            0.000000  191.107692  189.334483  187.748551         3.359141   \n",
       "4            0.000000  168.417062  173.756909  174.820232        -6.403171   \n",
       "5            0.000000  144.512782  156.559257  160.435213       -15.922431   \n",
       "6            0.000000  132.647402  146.392868  151.643416       -18.996015   \n",
       "7            0.000000  205.958763  187.630431  182.636828        23.321935   \n",
       "8            0.000000  197.416984  186.318696  182.394214        15.022770   \n",
       "9            0.000000  192.865730  185.879304  182.615797        10.249933   \n",
       "10           0.000000  167.618604  173.459028  174.235108        -6.616503   \n",
       "11           0.000000  137.177785  156.308929  162.288243       -25.110458   \n",
       "12           0.000000  111.654043  139.759679  150.422343       -38.768300   \n",
       "13           0.000000   90.028618  123.965667  138.806783       -48.778166   \n",
       "14           0.000000   84.339420  115.428045  131.688101       -47.348682   \n",
       "15           0.000000   87.898365  112.052150  127.850225       -39.951860   \n",
       "16           0.000000   77.254774  102.893605  120.556799       -43.302026   \n",
       "17           0.000000   91.179271  105.500394  120.402990       -29.223719   \n",
       "18           0.000000  115.130318  114.732720  124.509958        -9.379640   \n",
       "19           0.000000  106.751026  110.788050  121.255407       -14.504382   \n",
       "20           0.000000  109.167835  111.297454  120.584737       -11.416902   \n",
       "21           0.000000   95.110011  104.305246  115.720827       -20.610816   \n",
       "22           0.000000  125.742737  117.306330  122.084617         3.658120   \n",
       "23           0.000000  136.495797  123.682599  125.243107        11.252689   \n",
       "24           0.000000  120.663237  118.263609  122.099394        -1.436157   \n",
       "25         -13.815385  108.108493  112.767026  118.750345       -10.641852   \n",
       "26          -9.407692  100.405527  108.447692  115.892563       -15.487036   \n",
       "27         -13.530769  118.937235  115.832122  119.253006        -0.315771   \n",
       "28          -5.684615  138.624978  125.472289  124.127839        14.497139   \n",
       "29           2.800000  134.749965  125.708897  124.364067        10.385898   \n",
       "...               ...         ...         ...         ...              ...   \n",
       "35728      -11.161538   26.889536   30.047989   33.567610        -6.678075   \n",
       "35729       -5.507692   38.593024   34.963683   35.673713         2.919311   \n",
       "35730       -1.184615   40.062016   36.200039   36.216401         3.845615   \n",
       "35731        7.276923   52.041344   42.323110   39.163334        12.878010   \n",
       "35732       19.292308   63.360896   49.042632   42.632717        20.728179   \n",
       "35733       21.407692   52.573931   46.266842   41.771034        10.802896   \n",
       "35734       13.792308   42.382620   42.533482   40.306513         2.076107   \n",
       "35735       17.015385   48.921747   45.528331   41.913438         7.008309   \n",
       "35736       10.992308   48.614498   45.908588   42.364294         6.250203   \n",
       "35737        6.376923   56.076332   49.768805   44.485458        11.590874   \n",
       "35738        7.161538   50.050888   47.958220   44.005054         6.045834   \n",
       "35739        6.953846   40.033925   43.656955   42.226901        -2.192976   \n",
       "35740        1.315385   38.022617   42.171270   41.617501        -3.594884   \n",
       "35741       -0.984615   38.348411   41.683382   41.423612        -3.075201   \n",
       "35742       -7.784615   36.232274   40.193631   40.725567        -4.493293   \n",
       "35743       -8.723077   34.488183   38.779226   40.005155        -5.516972   \n",
       "35744       -5.546154   33.992122   37.890114   39.486254        -5.494132   \n",
       "35745       -5.823077   33.661415   37.137789   39.005791        -5.344376   \n",
       "35746       -7.100000   31.774276   35.731975   38.190547        -6.416271   \n",
       "35747       -6.392308   34.182851   36.234748   38.250507        -4.067656   \n",
       "35748       -3.761538   38.788567   38.044787   38.972691        -0.184124   \n",
       "35749       -1.107692   41.859045   39.576358   39.641381         2.217664   \n",
       "35750        3.184615   48.239363   42.872303   41.223501         7.015862   \n",
       "35751        4.853846   45.492909   42.430410   41.132871         4.360038   \n",
       "35752        5.646154   45.995273   43.133424   41.567473         4.427799   \n",
       "35753        3.076923   43.663515   42.497513   41.377290         2.286225   \n",
       "35754        3.069231   46.442343   43.959434   42.164157         4.278186   \n",
       "35755        6.338462   57.961562   49.657983   45.040887        12.920676   \n",
       "35756        9.846154   58.974375   51.402908   46.223043        12.751332   \n",
       "35757       15.484615   64.316250   55.033230   48.354670        15.961580   \n",
       "\n",
       "       DIFF_12_26  DEA_12_26      MACD  MeanPrice  RaiseDown  \n",
       "0        0.000000   0.000000  0.000000    3642.85  -0.002704  \n",
       "1       -0.327564  -0.181980 -0.291168    3633.00  -0.003331  \n",
       "2       -0.634451  -0.367419 -0.534064    3620.90  -0.000801  \n",
       "3       -0.823469  -0.521908 -0.603123    3618.00  -0.000373  \n",
       "4       -0.869050  -0.625175 -0.487751    3616.65   0.000843  \n",
       "5       -0.751834  -0.659506 -0.184655    3619.70  -0.000553  \n",
       "6       -0.971763  -0.738530 -0.466465    3617.70  -0.002750  \n",
       "7       -1.721885  -0.974849 -1.494072    3607.75  -0.001289  \n",
       "8       -2.165518  -1.249899 -1.831237    3603.10   0.000111  \n",
       "9       -2.325090  -1.490804 -1.668572    3603.50  -0.000139  \n",
       "10      -2.524999  -1.717080 -1.615838    3603.00   0.000375  \n",
       "11      -2.483403  -1.881654 -1.203498    3604.35  -0.000014  \n",
       "12      -2.385718  -1.988332 -0.794773    3604.30   0.000166  \n",
       "13      -2.327979  -2.059386 -0.537186    3604.90   0.000055  \n",
       "14      -2.195391  -2.087579 -0.215624    3605.10   0.000610  \n",
       "15      -1.897371  -2.048436  0.302129    3607.30   0.000097  \n",
       "16      -1.757935  -1.988997  0.462124    3607.65  -0.000998  \n",
       "17      -1.982649  -1.987704  0.010111    3604.05  -0.000499  \n",
       "18      -1.784824  -1.946535  0.323422    3602.25   0.000944  \n",
       "19      -1.597776  -1.875970  0.556387    3605.65   0.001137  \n",
       "20      -1.037152  -1.706644  1.338984    3609.75   0.000180  \n",
       "21      -0.939550  -1.552085  1.225070    3610.40  -0.002839  \n",
       "22      -1.803848  -1.602737 -0.402222    3600.15  -0.002486  \n",
       "23      -2.728825  -1.829023 -1.799605    3591.20   0.000585  \n",
       "24      -2.887792  -2.041580 -1.692425    3593.30   0.000153  \n",
       "25      -3.398996  -2.313886 -2.170221    3593.85  -0.001057  \n",
       "26      -3.998578  -2.651641 -2.693873    3590.05  -0.001727  \n",
       "27      -4.991522  -3.120524 -3.741995    3583.85  -0.002776  \n",
       "28      -6.573239  -3.812137 -5.522204    3573.90  -0.001679  \n",
       "29      -7.731735  -4.597029 -6.269413    3567.90   0.000070  \n",
       "...           ...        ...       ...        ...        ...  \n",
       "35728   -0.404311  -0.697110  0.585599    3300.80   0.000409  \n",
       "35729   -0.271863  -0.612061  0.680396    3302.15   0.000424  \n",
       "35730   -0.005450  -0.490739  0.970576    3303.55   0.000288  \n",
       "35731    0.219294  -0.348732  1.136052    3304.50  -0.000151  \n",
       "35732    0.392876  -0.200410  1.186573    3304.00  -0.000030  \n",
       "35733    0.444624  -0.071403  1.032056    3303.90  -0.000030  \n",
       "35734    0.527964   0.048470  0.958988    3303.80   0.000636  \n",
       "35735    0.810604   0.200897  1.219415    3305.90   0.000514  \n",
       "35736    1.166398   0.393997  1.544801    3307.60   0.000363  \n",
       "35737    1.511633   0.617524  1.788218    3308.80   0.000015  \n",
       "35738    1.621301   0.818280  1.606042    3308.85  -0.000348  \n",
       "35739    1.624928   0.979609  1.290638    3307.70  -0.000151  \n",
       "35740    1.577344   1.099156  0.956375    3307.20  -0.000030  \n",
       "35741    1.538042   1.186933  0.702217    3307.10  -0.000121  \n",
       "35742    1.489722   1.247491  0.484462    3306.70   0.000151  \n",
       "35743    1.434888   1.284970  0.299834    3307.20  -0.000348  \n",
       "35744    1.231984   1.274373 -0.084778    3306.05   0.000060  \n",
       "35745    1.170656   1.253630 -0.165948    3306.25  -0.000106  \n",
       "35746    1.061403   1.215184 -0.307564    3305.90   0.000151  \n",
       "35747    0.947755   1.161698 -0.427887    3306.40  -0.000030  \n",
       "35748    0.943641   1.118087 -0.348892    3306.30   0.000257  \n",
       "35749    0.945618   1.083593 -0.275950    3307.15  -0.000166  \n",
       "35750    0.856619   1.038198 -0.363159    3306.60  -0.000076  \n",
       "35751    0.777128   0.985984 -0.417712    3306.35   0.000121  \n",
       "35752    0.737902   0.936368 -0.396932    3306.75  -0.000091  \n",
       "35753    0.698759   0.888846 -0.380173    3306.45  -0.000015  \n",
       "35754    0.628221   0.836721 -0.417001    3306.40  -0.000348  \n",
       "35755    0.406252   0.750627 -0.688751    3305.25  -0.000212  \n",
       "35756    0.275578   0.655617 -0.760078    3304.55   0.000151  \n",
       "35757    0.186013   0.561696 -0.751368    3305.05   0.000000  \n",
       "\n",
       "[35758 rows x 33 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 为主力合约数据添加两个特征：平均价和下一分钟涨跌幅\n",
    "# 其中，平均价为当前分钟的四个价格的平均值\n",
    "# 下一分钟的涨跌幅为（下一分钟平均价 - 当前分钟平均价）/当前分钟平均价\n",
    "data['MeanPrice'] = (data['Latestprice'] + data['First.Latestprice'] +data['MaxPrice']+data['MinPrice'])/4\n",
    "ser1 = data['MeanPrice'][1:]\n",
    "ser1.index = range(len(ser1))\n",
    "ser2 = data['MeanPrice'][:-1]\n",
    "ser2.index = range(len(ser2))\n",
    "data['RaiseDown'] = (ser1 -ser2)/ser2\n",
    "data = data.fillna(0)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.026806102525068671, -0.033535366458236689)"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max(data['RaiseDown']),min(data['RaiseDown'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  3.64040000e+03,   3.64600000e+03,   3.64020000e+03, ...,\n",
       "          0.00000000e+00,   3.64285000e+03,  -2.70392687e-03],\n",
       "       [  3.62580000e+03,   3.64140000e+03,   3.62440000e+03, ...,\n",
       "         -2.91168091e-01,   3.63300000e+03,  -3.33058079e-03],\n",
       "       [  3.61880000e+03,   3.62680000e+03,   3.61260000e+03, ...,\n",
       "         -5.34064377e-01,   3.62090000e+03,  -8.00905852e-04],\n",
       "       ..., \n",
       "       [  3.30420000e+03,   3.30680000e+03,   3.30420000e+03, ...,\n",
       "         -6.88750506e-01,   3.30525000e+03,  -2.11784283e-04],\n",
       "       [  3.30480000e+03,   3.30540000e+03,   3.30380000e+03, ...,\n",
       "         -7.60078129e-01,   3.30455000e+03,   1.51306532e-04],\n",
       "       [  3.30500000e+03,   3.30620000e+03,   3.30400000e+03, ...,\n",
       "         -7.51367561e-01,   3.30505000e+03,   0.00000000e+00]])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_array = np.array(data)\n",
    "data_array"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 对数据进行标准化"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.        ,  1.        ,  1.        , ...,  0.65315817,\n",
       "         1.        ,  0.51094944],\n",
       "       [ 0.98349537,  0.9947952 ,  0.98259529, ...,  0.64915289,\n",
       "         0.98885873,  0.50056431],\n",
       "       [ 0.97558218,  0.97827563,  0.96959683, ...,  0.64581164,\n",
       "         0.97517249,  0.54248697],\n",
       "       ..., \n",
       "       [ 0.61994122,  0.61620276,  0.62987442, ...,  0.64368379,\n",
       "         0.61814274,  0.5522501 ],\n",
       "       [ 0.62061949,  0.61461869,  0.6294338 , ...,  0.64270261,\n",
       "         0.61735098,  0.55826737],\n",
       "       [ 0.62084558,  0.61552387,  0.62965411, ...,  0.64282244,\n",
       "         0.61791653,  0.55575986]])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import preprocessing\n",
    "min_max_scaler = preprocessing.MinMaxScaler()\n",
    "data_array = min_max_scaler.fit_transform(data_array)\n",
    "data_array"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 生成5分钟的时间序列，变为3维数组"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(35753, 5, 33)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "seq_5 = []\n",
    "for j in range(len(data_array)):\n",
    "    if j+5 < len(data_array):\n",
    "        seq_5.append(data_array[j:j+5])\n",
    "seq_5 = np.array(seq_5)\n",
    "seq_5.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 进行涨跌情况的分类"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 提取训练集、测试集。根据下一分钟的涨跌幅分别打标签，分为五类"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 5, 33)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 取前面10000个数据作为训练集\n",
    "# 前面的数据点因有na值填充，会对准确性造成一定影响，因此从第1000个点开始\n",
    "train_data = seq_5[1000:11000]\n",
    "train_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 5)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data1 = data[1000:11000]\n",
    "train_label = []\n",
    "for i in data1['RaiseDown']:\n",
    "    if i > 0.002:\n",
    "        train_label.append([1,0,0,0,0])   #大涨\n",
    "    elif i > 0.0005:\n",
    "        train_label.append([0,1,0,0,0])   #小涨\n",
    "    elif i> -0.0005: \n",
    "        train_label.append([0,0,1,0,0])   #平稳\n",
    "    elif i > -0.002:\n",
    "        train_label.append([0,0,0,1,0])   #小跌\n",
    "    else:\n",
    "        train_label.append([0,0,0,0,1])   #大跌\n",
    "train_label = np.array(train_label)\n",
    "train_label.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "大涨： 244\n",
      "小涨： 2136\n",
      "平稳： 5124\n",
      "小跌： 2313\n",
      "大跌： 183\n"
     ]
    }
   ],
   "source": [
    "# 统计五类情况各自的数量\n",
    "train_label_list = train_label.tolist()\n",
    "print u'大涨：',train_label_list.count([1,0,0,0,0])\n",
    "print u'小涨：',train_label_list.count([0,1,0,0,0])\n",
    "print u'平稳：',train_label_list.count([0,0,1,0,0])\n",
    "print u'小跌：',train_label_list.count([0,0,0,1,0])\n",
    "print u'大跌：',train_label_list.count([0,0,0,0,1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# 取后面2000个分钟数据作为测试集"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2000, 5, 33)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data = seq_5[12000:14000]\n",
    "test_data = np.array(test_data)\n",
    "test_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2000, 5)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data2 = data[12000:14000]\n",
    "test_label = []\n",
    "for i in data2['RaiseDown']:\n",
    "    if i > 0.002:\n",
    "        test_label.append([1,0,0,0,0])\n",
    "    elif i > 0.0005:\n",
    "        test_label.append([0,1,0,0,0])\n",
    "    elif i> -0.0005:\n",
    "        test_label.append([0,0,1,0,0])\n",
    "    elif i > -0.002:\n",
    "        test_label.append([0,0,0,1,0])\n",
    "    else:\n",
    "        test_label.append([0,0,0,0,1])\n",
    "test_label = np.array(test_label)\n",
    "test_label.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "大涨： 17\n",
      "小涨： 320\n",
      "平稳： 1304\n",
      "小跌： 349\n",
      "大跌： 10\n"
     ]
    }
   ],
   "source": [
    "test_label_list = test_label.tolist()\n",
    "# 测试集五类样本的数量\n",
    "print u'大涨：', test_label_list.count([1,0,0,0,0])\n",
    "print u'小涨：', test_label_list.count([0,1,0,0,0])\n",
    "print u'平稳：', test_label_list.count([0,0,1,0,0])\n",
    "print u'小跌：', test_label_list.count([0,0,0,1,0])\n",
    "print u'大跌：', test_label_list.count([0,0,0,0,1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 开始构建LSTM网络，进行分类准确率测试"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "learning_rate = 0.001 # 学习速率\n",
    "training_iters = 10000  # 训练总步长\n",
    "batch_size = 50  # 每小段步长\n",
    "display_step = 10  # 每隔十小段显示输出\n",
    "\n",
    "# 参数设置\n",
    "n_input = 33 # 特征数量\n",
    "n_steps = 5 # 时间序列长度\n",
    "n_hidden = 500 # 隐藏层神经元个数\n",
    "n_classes = 5 # 分类数量\n",
    "\n",
    "tf.reset_default_graph()  # 重置流图\n",
    "# 设置输入输出格式大小\n",
    "xtr = tf.placeholder(\"float\", [None, n_steps, n_input])\n",
    "ytr = tf.placeholder(\"float\", [None,n_classes])\n",
    "# xte = tf.placeholder(\"float\", [None, n_steps, n_input])\n",
    "# yte = tf.placeholder(\"float\", [None,n_classes])\n",
    "xtr1 = tf.placeholder(\"float\", [None, n_input*n_steps])  \n",
    "xte1 = tf.placeholder(\"float\", [n_input*n_steps])  \n",
    "# 定义权重和偏移量\n",
    "weights = {\n",
    "    'out': tf.Variable(tf.random_normal([n_hidden, n_classes]))\n",
    "}\n",
    "biases = {\n",
    "    'out': tf.Variable(tf.random_normal([n_classes]))\n",
    "}\n",
    "\n",
    "# def pred():\n",
    "    \n",
    "def RNN(x, weights, biases):\n",
    "\n",
    "    # 更改数据集的格式，使其满足RNN网络的输入需求\n",
    "    # 现在输入格式: (batch_size, n_steps, n_input)\n",
    "    # 需要的格式: 'n_steps' 个tensor，每个的格式为 (batch_size, n_input)\n",
    "\n",
    "    # 变换batch_size 和 n_steps的位置\n",
    "    x = tf.transpose(x, [1, 0, 2])\n",
    "    # 重新调整格式 (n_steps*batch_size, n_input)\n",
    "    x = tf.reshape(x, [-1, n_input])\n",
    "    # 分割成一系列的 'n_steps' 个tensors，每个的格式为 (batch_size, n_input)\n",
    "    x = tf.split(0, n_steps, x)\n",
    "\n",
    "    # 定义LSTM神经网络结构\n",
    "    lstm_cell = rnn_cell.BasicLSTMCell(n_hidden, forget_bias=1.0)\n",
    "    # 获取输出和状态\n",
    "    outputs, states = rnn.rnn(lstm_cell, x, dtype=tf.float32)\n",
    "    # 返回输出\n",
    "    return tf.matmul(outputs[-1], weights['out']) + biases['out']\n",
    "# 获取预测值\n",
    "pred = RNN(xtr, weights, biases)\n",
    "# train_label = tf.split(0, batch_size, train_label)\n",
    "# 取下一小段的输入\n",
    "def next_batch_tr(step, batch_size):\n",
    "    return train_data[(step-1)*batch_size:step*batch_size], train_label[(step-1)*batch_size:step*batch_size]\n",
    "def next_batch_te(step, batch_size):\n",
    "    return test_data[(step-1)*batch_size:step*batch_size], test_label[(step-1)*batch_size:step*batch_size]\n",
    "\n",
    "Xtr1 = np.reshape(train_data, newshape=(-1, n_input*n_steps))  \n",
    "Xte1 = np.reshape(test_data, newshape=(-1, n_input*n_steps)) \n",
    "Xtr2 = map(map, [float, float], Xtr1)\n",
    "Xte2 = map(map, [float, float], Xte1)\n",
    "Xtr3 = np.array(Xtr2)\n",
    "Xte3 = np.array(Xte2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Exception AssertionError: AssertionError() in <bound method InteractiveSession.__del__ of <tensorflow.python.client.session.InteractiveSession object at 0x7f7b2a378610>> ignored\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Iter 500, Minibatch Loss= 3.195077, Training Accuracy= 0.36000\n",
      "Iter 1000, Minibatch Loss= 1.577735, Training Accuracy= 0.38000\n",
      "Iter 1500, Minibatch Loss= 1.418900, Training Accuracy= 0.48000\n",
      "Iter 2000, Minibatch Loss= 1.398080, Training Accuracy= 0.40000\n",
      "Iter 2500, Minibatch Loss= 1.507281, Training Accuracy= 0.30000\n",
      "Iter 3000, Minibatch Loss= 1.288328, Training Accuracy= 0.38000\n",
      "Iter 3500, Minibatch Loss= 0.855918, Training Accuracy= 0.70000\n",
      "Iter 4000, Minibatch Loss= 1.138813, Training Accuracy= 0.58000\n",
      "Iter 4500, Minibatch Loss= 1.086167, Training Accuracy= 0.52000\n",
      "Iter 5000, Minibatch Loss= 0.851407, Training Accuracy= 0.72000\n",
      "Iter 5500, Minibatch Loss= 1.126116, Training Accuracy= 0.54000\n",
      "Iter 6000, Minibatch Loss= 1.005334, Training Accuracy= 0.72000\n",
      "Iter 6500, Minibatch Loss= 1.412037, Training Accuracy= 0.38000\n",
      "Iter 7000, Minibatch Loss= 1.066712, Training Accuracy= 0.54000\n",
      "Iter 7500, Minibatch Loss= 1.789981, Training Accuracy= 0.26000\n",
      "Iter 8000, Minibatch Loss= 1.216005, Training Accuracy= 0.48000\n",
      "Iter 8500, Minibatch Loss= 1.204752, Training Accuracy= 0.58000\n",
      "Iter 9000, Minibatch Loss= 1.209887, Training Accuracy= 0.48000\n",
      "Iter 9500, Minibatch Loss= 1.231871, Training Accuracy= 0.50000\n",
      "Optimization Finished!\n",
      "('Testing Accuracy:', 0.69999999)\n",
      "Test 0 Prediction: 3 True Class: 2\n",
      "Test 1 Prediction: 2 True Class: 2\n",
      "Test 2 Prediction: 3 True Class: 3\n",
      "Test 3 Prediction: 3 True Class: 3\n",
      "Test 4 Prediction: 2 True Class: 3\n",
      "Test 5 Prediction: 3 True Class: 3\n",
      "Test 6 Prediction: 3 True Class: 2\n",
      "Test 7 Prediction: 3 True Class: 1\n",
      "Test 8 Prediction: 1 True Class: 1\n",
      "Test 9 Prediction: 2 True Class: 2\n",
      "Test 10 Prediction: 2 True Class: 2\n",
      "Test 11 Prediction: 2 True Class: 2\n",
      "Test 12 Prediction: 2 True Class: 2\n",
      "Test 13 Prediction: 2 True Class: 2\n",
      "Test 14 Prediction: 2 True Class: 2\n",
      "Test 15 Prediction: 2 True Class: 2\n",
      "Test 16 Prediction: 2 True Class: 1\n",
      "Test 17 Prediction: 3 True Class: 2\n",
      "Test 18 Prediction: 2 True Class: 3\n",
      "Test 19 Prediction: 4 True Class: 3\n",
      "Test 20 Prediction: 3 True Class: 2\n",
      "Test 21 Prediction: 1 True Class: 1\n",
      "Test 22 Prediction: 3 True Class: 2\n",
      "Test 23 Prediction: 3 True Class: 2\n",
      "Test 24 Prediction: 2 True Class: 2\n",
      "Test 25 Prediction: 1 True Class: 2\n",
      "Test 26 Prediction: 1 True Class: 1\n",
      "Test 27 Prediction: 3 True Class: 1\n",
      "Test 28 Prediction: 2 True Class: 2\n",
      "Test 29 Prediction: 2 True Class: 2\n",
      "Test 30 Prediction: 2 True Class: 2\n",
      "Test 31 Prediction: 1 True Class: 2\n",
      "Test 32 Prediction: 2 True Class: 2\n",
      "Test 33 Prediction: 3 True Class: 2\n",
      "Test 34 Prediction: 2 True Class: 2\n",
      "Test 35 Prediction: 2 True Class: 2\n",
      "Test 36 Prediction: 2 True Class: 2\n",
      "Test 37 Prediction: 2 True Class: 2\n",
      "Test 38 Prediction: 1 True Class: 1\n",
      "Test 39 Prediction: 3 True Class: 1\n",
      "Test 40 Prediction: 2 True Class: 2\n",
      "Test 41 Prediction: 2 True Class: 1\n",
      "Test 42 Prediction: 3 True Class: 2\n",
      "Test 43 Prediction: 2 True Class: 2\n",
      "Test 44 Prediction: 2 True Class: 2\n",
      "Test 45 Prediction: 1 True Class: 2\n",
      "Test 46 Prediction: 1 True Class: 1\n",
      "Test 47 Prediction: 2 True Class: 1\n",
      "Test 48 Prediction: 3 True Class: 2\n",
      "Test 49 Prediction: 3 True Class: 2\n",
      "Test 50 Prediction: 3 True Class: 3\n",
      "Test 51 Prediction: 2 True Class: 2\n",
      "Test 52 Prediction: 2 True Class: 2\n",
      "Test 53 Prediction: 3 True Class: 2\n",
      "Test 54 Prediction: 3 True Class: 2\n",
      "Test 55 Prediction: 2 True Class: 2\n",
      "Test 56 Prediction: 2 True Class: 2\n",
      "Test 57 Prediction: 2 True Class: 2\n",
      "Test 58 Prediction: 2 True Class: 2\n",
      "Test 59 Prediction: 2 True Class: 2\n",
      "Test 60 Prediction: 0 True Class: 2\n",
      "Test 61 Prediction: 2 True Class: 2\n",
      "Test 62 Prediction: 2 True Class: 2\n",
      "Test 63 Prediction: 2 True Class: 2\n",
      "Test 64 Prediction: 2 True Class: 3\n",
      "Test 65 Prediction: 2 True Class: 3\n",
      "Test 66 Prediction: 3 True Class: 2\n",
      "Test 67 Prediction: 2 True Class: 3\n",
      "Test 68 Prediction: 1 True Class: 1\n",
      "Test 69 Prediction: 1 True Class: 2\n",
      "Test 70 Prediction: 1 True Class: 2\n",
      "Test 71 Prediction: 3 True Class: 2\n",
      "Test 72 Prediction: 1 True Class: 2\n",
      "Test 73 Prediction: 1 True Class: 2\n",
      "Test 74 Prediction: 2 True Class: 3\n",
      "Test 75 Prediction: 3 True Class: 2\n",
      "Test 76 Prediction: 2 True Class: 2\n",
      "Test 77 Prediction: 2 True Class: 2\n",
      "Test 78 Prediction: 2 True Class: 2\n",
      "Test 79 Prediction: 2 True Class: 1\n",
      "Test 80 Prediction: 2 True Class: 2\n",
      "Test 81 Prediction: 2 True Class: 2\n",
      "Test 82 Prediction: 2 True Class: 2\n",
      "Test 83 Prediction: 2 True Class: 2\n",
      "Test 84 Prediction: 2 True Class: 2\n",
      "Test 85 Prediction: 2 True Class: 1\n",
      "Test 86 Prediction: 2 True Class: 2\n",
      "Test 87 Prediction: 1 True Class: 2\n",
      "Test 88 Prediction: 2 True Class: 2\n",
      "Test 89 Prediction: 1 True Class: 3\n",
      "Test 90 Prediction: 1 True Class: 3\n",
      "Test 91 Prediction: 2 True Class: 3\n",
      "Test 92 Prediction: 2 True Class: 2\n",
      "Test 93 Prediction: 2 True Class: 2\n",
      "Test 94 Prediction: 1 True Class: 2\n",
      "Test 95 Prediction: 1 True Class: 2\n",
      "Test 96 Prediction: 2 True Class: 3\n",
      "Test 97 Prediction: 1 True Class: 2\n",
      "Test 98 Prediction: 2 True Class: 2\n",
      "Test 99 Prediction: 1 True Class: 1\n",
      "Done!\n"
     ]
    }
   ],
   "source": [
    "# 定义代价函数和最优化方法\n",
    "cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(pred, ytr))\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)\n",
    "\n",
    "# 评估模型准确率\n",
    "correct_pred = tf.equal(tf.argmax(pred, 1), tf.argmax(ytr, 1))\n",
    "accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))\n",
    "\n",
    "# 定义相似度，即预测值与训练值的距离\n",
    "distance = tf.reduce_sum(tf.abs(tf.add(xtr1, tf.neg(xte1))), reduction_indices=1)  \n",
    "# 预测值: 获取相似度最高的训练标签，作为预测标签\n",
    "pred1 = tf.arg_min(distance, 0)  \n",
    "# 变量初始化\n",
    "init = tf.initialize_all_variables()\n",
    "\n",
    "# 创建会话\n",
    "sess = tf.InteractiveSession()\n",
    "sess.run(init)\n",
    "step = 1\n",
    "    # 一直迭代，直到最大步长\n",
    "while step * batch_size < training_iters:\n",
    "    batch_x, batch_y = next_batch_tr(step, batch_size)\n",
    "    batch_x = batch_x.reshape((batch_size, n_steps, n_input))\n",
    "        # 利用最优化方法调整权值\n",
    "    sess.run(optimizer, feed_dict={xtr: batch_x, ytr: batch_y})\n",
    "    if step % display_step == 0:\n",
    "            # 计算准确率和损失函数\n",
    "        acc = sess.run(accuracy, feed_dict={xtr: batch_x, ytr: batch_y})\n",
    "        loss = sess.run(cost, feed_dict={xtr: batch_x, ytr: batch_y})\n",
    "        # 每隔一定步长显示训练准确率和损失率\n",
    "        print(\"Iter \" + str(step * batch_size) + \", Minibatch Loss= \" +\n",
    "                \"{:.6f}\".format(loss) + \", Training Accuracy= \" +\n",
    "                \"{:.5f}\".format(acc))\n",
    "    step += 1\n",
    "print(\"Optimization Finished!\")\n",
    "\n",
    "    # 计算n个预测值的预测准确率\n",
    "test_len = 100\n",
    "test_data = test_data[:test_len].reshape((-1, n_steps, n_input))\n",
    "test_label = test_label[:test_len]\n",
    "print(\"Testing Accuracy:\",\n",
    "        sess.run(accuracy, feed_dict={xtr: test_data, ytr: test_label}))\n",
    "\n",
    "for i in range(test_len):  \n",
    "        # 索引即为相似度最大的标签值\n",
    "    nn_index = sess.run(pred1, feed_dict={xtr1: Xtr3, xte1: Xte3[i,:]})  \n",
    "        # 打印出对应的预测标签和实际标签\n",
    "    print \"Test\", i, \"Prediction:\", np.argmax(train_label[nn_index]), \"True Class:\", np.argmax(test_label[i])  \n",
    "\n",
    "print \"Done!\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 作出预测值和真实标签的对比"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f7af6ed5c50>"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7b2b57ba90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhwAAAGLCAYAAACMdUsaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7b2b87a9d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#!/usr/bin/env python\n",
    "# -*- coding: utf-8 -*-  \n",
    "# 画出对比图，绿色表示实际值，红色表示预测值\n",
    "import numpy as np    \n",
    "import matplotlib.mlab as mlab    \n",
    "import matplotlib.pyplot as plt    \n",
    "# import ch\n",
    "%matplotlib inline\n",
    "\n",
    "# from pylab import *\n",
    "\n",
    "# mpl.rcParams['font.sans-serif'] = ['SimHei']\n",
    "# mpl.rcParams['axes.unicode_minus'] = False\n",
    "Y1=[]\n",
    "for i in range(15,65):\n",
    "    nn_index = sess.run(pred1, feed_dict={xtr1: Xtr3, xte1: Xte3[i,:]})  \n",
    "    Y1.append(2 - np.argmax(train_label[nn_index]))\n",
    "Y2=[]\n",
    "for i in range(15,65):\n",
    "    Y2.append(2 - np.argmax(test_label[i]))\n",
    "\n",
    "Y3 = pd.concat([Series(Y1),Series(Y2)],axis=1)\n",
    "# set_ch()\n",
    "Y3.columns = ['Predict','Real']\n",
    "X=range(15,65)  \n",
    "fig = plt.figure()  \n",
    "Y3.plot(color=['red','green'],kind='bar',use_index=False)  \n",
    "#plt.bar(X,Y2,0.4,color=\"red\")  \n",
    "plt.xlabel(\"time/min\")  \n",
    "plt.ylabel(\"Class\")  \n",
    "plt.title(\"Contrast\")  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 进行最高价的分类"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(20000, 5, 32)"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train1_data = seq_5[2000:22000]\n",
    "train1_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(20000, 10)"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data3 = data[2000:22000]\n",
    "train1_label = []\n",
    "for i in data3['MaxPrice']:\n",
    "    if i < 3100:\n",
    "        train1_label.append([1,0,0,0,0,0,0,0,0,0])   # 最高价 < 2800\n",
    "    elif i < 3110:\n",
    "        train1_label.append([0,1,0,0,0,0,0,0,0,0])   # 最高价 2800~2900\n",
    "    elif i < 3120: \n",
    "        train1_label.append([0,0,1,0,0,0,0,0,0,0])   # 最高价 2900~3000\n",
    "    elif i < 3130:\n",
    "        train1_label.append([0,0,0,1,0,0,0,0,0,0])   # 最高价 3000~3100\n",
    "    elif i < 3140:\n",
    "        train1_label.append([0,0,0,0,1,0,0,0,0,0])   # 最高价 3100~3200\n",
    "    elif i < 3150:\n",
    "        train1_label.append([0,0,0,0,0,1,0,0,0,0])   # 最高价 3200~3300\n",
    "    elif i < 3160:\n",
    "        train1_label.append([0,0,0,0,0,0,1,0,0,0])   # 最高价 3300~3400\n",
    "    elif i < 3170:\n",
    "        train1_label.append([0,0,0,0,0,0,0,1,0,0])   # 最高价 3400~3500\n",
    "    elif i < 3180:\n",
    "        train1_label.append([0,0,0,0,0,0,0,0,1,0])   # 最高价 3500~3600\n",
    "    else:\n",
    "        train1_label.append([0,0,0,0,0,0,0,0,0,1])   # 最高价 > 3600\n",
    "    \n",
    "train1_label = np.array(train1_label)\n",
    "train1_label.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11615\n",
      "267\n",
      "421\n",
      "655\n",
      "1086\n",
      "598\n",
      "403\n",
      "400\n",
      "587\n",
      "3968\n"
     ]
    }
   ],
   "source": [
    "train1_label_list = train1_label.tolist()\n",
    "print train1_label_list.count([1,0,0,0,0,0,0,0,0,0])  \n",
    "print train1_label_list.count([0,1,0,0,0,0,0,0,0,0]) \n",
    "print train1_label_list.count([0,0,1,0,0,0,0,0,0,0]) \n",
    "print train1_label_list.count([0,0,0,1,0,0,0,0,0,0])\n",
    "print train1_label_list.count([0,0,0,0,1,0,0,0,0,0]) \n",
    "print train1_label_list.count([0,0,0,0,0,1,0,0,0,0]) \n",
    "print train1_label_list.count([0,0,0,0,0,0,1,0,0,0]) \n",
    "print train1_label_list.count([0,0,0,0,0,0,0,1,0,0]) \n",
    "print train1_label_list.count([0,0,0,0,0,0,0,0,1,0]) \n",
    "print train1_label_list.count([0,0,0,0,0,0,0,0,0,1]) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1000, 5, 32)"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test1_data = seq_5[22000:23000]\n",
    "test1_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1000, 10)"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data4 = data[22000:23000]\n",
    "test1_label = []\n",
    "for i in data4['MaxPrice']:\n",
    "    if i < 3100:\n",
    "        test1_label.append([1,0,0,0,0,0,0,0,0,0])   # 最高价 < 2800\n",
    "    elif i < 3110:\n",
    "        test1_label.append([0,1,0,0,0,0,0,0,0,0])   # 最高价 2800~2900\n",
    "    elif i < 3120: \n",
    "        test1_label.append([0,0,1,0,0,0,0,0,0,0])   # 最高价 2900~3000\n",
    "    elif i < 3130:\n",
    "        test1_label.append([0,0,0,1,0,0,0,0,0,0])   # 最高价 3000~3100\n",
    "    elif i < 3140:\n",
    "        test1_label.append([0,0,0,0,1,0,0,0,0,0])   # 最高价 3100~3200\n",
    "    elif i < 3150:\n",
    "        test1_label.append([0,0,0,0,0,1,0,0,0,0])   # 最高价 3200~3300\n",
    "    elif i < 3160:\n",
    "        test1_label.append([0,0,0,0,0,0,1,0,0,0])   # 最高价 3300~3400\n",
    "    elif i < 3170:\n",
    "        test1_label.append([0,0,0,0,0,0,0,1,0,0])   # 最高价 3400~3500\n",
    "    elif i < 3180:\n",
    "        test1_label.append([0,0,0,0,0,0,0,0,1,0])   # 最高价 3500~3600\n",
    "    else:\n",
    "        test1_label.append([0,0,0,0,0,0,0,0,0,1])   # 最高价 > 3600\n",
    "    \n",
    "test1_label = np.array(test1_label)\n",
    "test1_label.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3182.8000000000002, 3043.8000000000002)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "max(data4['MaxPrice']),min(data4['MaxPrice'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "0\n",
      "5\n",
      "7\n",
      "234\n",
      "162\n",
      "386\n",
      "157\n",
      "42\n",
      "7\n"
     ]
    }
   ],
   "source": [
    "test1_label_list = test1_label.tolist()\n",
    "# 测试集五类样本的数量\n",
    "print test1_label_list.count([1,0,0,0,0,0,0,0,0,0])  \n",
    "print test1_label_list.count([0,1,0,0,0,0,0,0,0,0]) \n",
    "print test1_label_list.count([0,0,1,0,0,0,0,0,0,0]) \n",
    "print test1_label_list.count([0,0,0,1,0,0,0,0,0,0])\n",
    "print test1_label_list.count([0,0,0,0,1,0,0,0,0,0]) \n",
    "print test1_label_list.count([0,0,0,0,0,1,0,0,0,0]) \n",
    "print test1_label_list.count([0,0,0,0,0,0,1,0,0,0]) \n",
    "print test1_label_list.count([0,0,0,0,0,0,0,1,0,0]) \n",
    "print test1_label_list.count([0,0,0,0,0,0,0,0,1,0]) \n",
    "print test1_label_list.count([0,0,0,0,0,0,0,0,0,1]) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "learning_rate = 0.001 # 学习速率\n",
    "training_iters = 20000  # 训练总步长\n",
    "batch_size = 200  # 每小段步长\n",
    "display_step = 10  # 每隔十小段显示输出\n",
    "\n",
    "# 参数设置\n",
    "n_input = 32 # 特征数量\n",
    "n_steps = 5 # 时间序列长度\n",
    "n_hidden = 500 # 隐藏层神经元个数\n",
    "n_classes = 10 # 分类数量\n",
    "\n",
    "tf.reset_default_graph()  # 重置流图\n",
    "# 设置输入输出格式大小\n",
    "xtr = tf.placeholder(\"float\", [None, n_steps, n_input])\n",
    "ytr = tf.placeholder(\"float\", [None,n_classes])\n",
    "# xte = tf.placeholder(\"float\", [None, n_steps, n_input])\n",
    "# yte = tf.placeholder(\"float\", [None,n_classes])\n",
    "xtr1 = tf.placeholder(\"float\", [None, n_input*n_steps])  \n",
    "xte1 = tf.placeholder(\"float\", [n_input*n_steps])  \n",
    "# 定义权重和偏移量\n",
    "weights = {\n",
    "    'out': tf.Variable(tf.random_normal([n_hidden, n_classes]))\n",
    "}\n",
    "biases = {\n",
    "    'out': tf.Variable(tf.random_normal([n_classes]))\n",
    "}\n",
    "\n",
    "# def pred():\n",
    "    \n",
    "def RNN(x, weights, biases):\n",
    "\n",
    "    # 更改数据集的格式，使其满足RNN网络的输入需求\n",
    "    # 现在输入格式: (batch_size, n_steps, n_input)\n",
    "    # 需要的格式: 'n_steps' 个tensor，每个的格式为 (batch_size, n_input)\n",
    "\n",
    "    # 变换batch_size 和 n_steps的位置\n",
    "    x = tf.transpose(x, [1, 0, 2])\n",
    "    # 重新调整格式 (n_steps*batch_size, n_input)\n",
    "    x = tf.reshape(x, [-1, n_input])\n",
    "    # 分割成一系列的 'n_steps' 个tensors，每个的格式为 (batch_size, n_input)\n",
    "    x = tf.split(0, n_steps, x)\n",
    "\n",
    "    # 定义LSTM神经网络结构\n",
    "    lstm_cell = rnn_cell.BasicLSTMCell(n_hidden, forget_bias=1.0)\n",
    "    # 获取输出和状态\n",
    "    outputs, states = rnn.rnn(lstm_cell, x, dtype=tf.float32)\n",
    "    # 返回输出\n",
    "    return tf.matmul(outputs[-1], weights['out']) + biases['out']\n",
    "# 获取预测值\n",
    "pred = RNN(xtr, weights, biases)\n",
    "# train_label = tf.split(0, batch_size, train_label)\n",
    "# 取下一小段的输入\n",
    "def next_batch_tr(step, batch_size):\n",
    "    return train1_data[(step-1)*batch_size:step*batch_size], train1_label[(step-1)*batch_size:step*batch_size]\n",
    "def next_batch_te(step, batch_size):\n",
    "    return test1_data[(step-1)*batch_size:step*batch_size], test1_label[(step-1)*batch_size:step*batch_size]\n",
    "\n",
    "Xtr1 = np.reshape(train1_data, newshape=(-1, n_input*n_steps))  \n",
    "Xte1 = np.reshape(test1_data, newshape=(-1, n_input*n_steps)) \n",
    "Xtr2 = map(map, [float, float], Xtr1)\n",
    "Xte2 = map(map, [float, float], Xte1)\n",
    "Xtr3 = np.array(Xtr2)\n",
    "Xte3 = np.array(Xte2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Exception AssertionError: AssertionError() in <bound method InteractiveSession.__del__ of <tensorflow.python.client.session.InteractiveSession object at 0x7f7b282ed710>> ignored\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization Finished!\n",
      "Test 0 Prediction: 4 True Class: 4\n",
      "Test 1 Prediction: 4 True Class: 4\n",
      "Test 2 Prediction: 4 True Class: 4\n",
      "Test 3 Prediction: 4 True Class: 4\n",
      "Test 4 Prediction: 4 True Class: 4\n",
      "Test 5 Prediction: 4 True Class: 4\n",
      "Test 6 Prediction: 4 True Class: 4\n",
      "Test 7 Prediction: 4 True Class: 4\n",
      "Test 8 Prediction: 4 True Class: 4\n",
      "Test 9 Prediction: 4 True Class: 4\n",
      "Test 10 Prediction: 4 True Class: 4\n",
      "Test 11 Prediction: 4 True Class: 4\n",
      "Test 12 Prediction: 4 True Class: 4\n",
      "Test 13 Prediction: 4 True Class: 4\n",
      "Test 14 Prediction: 4 True Class: 4\n",
      "Test 15 Prediction: 4 True Class: 4\n",
      "Test 16 Prediction: 4 True Class: 4\n",
      "Test 17 Prediction: 5 True Class: 4\n",
      "Test 18 Prediction: 4 True Class: 4\n",
      "Test 19 Prediction: 4 True Class: 4\n",
      "Test 20 Prediction: 4 True Class: 4\n",
      "Test 21 Prediction: 4 True Class: 4\n",
      "Test 22 Prediction: 4 True Class: 4\n",
      "Test 23 Prediction: 4 True Class: 4\n",
      "Test 24 Prediction: 4 True Class: 4\n",
      "Test 25 Prediction: 4 True Class: 4\n",
      "Test 26 Prediction: 4 True Class: 4\n",
      "Test 27 Prediction: 4 True Class: 4\n",
      "Test 28 Prediction: 4 True Class: 4\n",
      "Test 29 Prediction: 4 True Class: 4\n",
      "Test 30 Prediction: 4 True Class: 4\n",
      "Test 31 Prediction: 4 True Class: 4\n",
      "Test 32 Prediction: 4 True Class: 4\n",
      "Test 33 Prediction: 5 True Class: 4\n",
      "Test 34 Prediction: 4 True Class: 4\n",
      "Test 35 Prediction: 4 True Class: 4\n",
      "Test 36 Prediction: 4 True Class: 4\n",
      "Test 37 Prediction: 4 True Class: 4\n",
      "Test 38 Prediction: 4 True Class: 4\n",
      "Test 39 Prediction: 4 True Class: 4\n",
      "Test 40 Prediction: 4 True Class: 4\n",
      "Test 41 Prediction: 4 True Class: 4\n",
      "Test 42 Prediction: 4 True Class: 4\n",
      "Test 43 Prediction: 4 True Class: 4\n",
      "Test 44 Prediction: 4 True Class: 4\n",
      "Test 45 Prediction: 4 True Class: 4\n",
      "Test 46 Prediction: 3 True Class: 4\n",
      "Test 47 Prediction: 4 True Class: 4\n",
      "Test 48 Prediction: 4 True Class: 4\n",
      "Test 49 Prediction: 4 True Class: 4\n",
      "Test 50 Prediction: 4 True Class: 4\n",
      "Test 51 Prediction: 4 True Class: 4\n",
      "Test 52 Prediction: 4 True Class: 4\n",
      "Test 53 Prediction: 4 True Class: 4\n",
      "Test 54 Prediction: 4 True Class: 4\n",
      "Test 55 Prediction: 4 True Class: 4\n",
      "Test 56 Prediction: 4 True Class: 4\n",
      "Test 57 Prediction: 4 True Class: 4\n",
      "Test 58 Prediction: 4 True Class: 4\n",
      "Test 59 Prediction: 4 True Class: 4\n",
      "Test 60 Prediction: 4 True Class: 4\n",
      "Test 61 Prediction: 4 True Class: 4\n",
      "Test 62 Prediction: 4 True Class: 4\n",
      "Test 63 Prediction: 4 True Class: 4\n",
      "Test 64 Prediction: 4 True Class: 4\n",
      "Test 65 Prediction: 4 True Class: 4\n",
      "Test 66 Prediction: 4 True Class: 4\n",
      "Test 67 Prediction: 4 True Class: 4\n",
      "Test 68 Prediction: 4 True Class: 4\n",
      "Test 69 Prediction: 4 True Class: 4\n",
      "Test 70 Prediction: 4 True Class: 4\n",
      "Test 71 Prediction: 4 True Class: 4\n",
      "Test 72 Prediction: 4 True Class: 4\n",
      "Test 73 Prediction: 4 True Class: 4\n",
      "Test 74 Prediction: 4 True Class: 4\n",
      "Test 75 Prediction: 4 True Class: 4\n",
      "Test 76 Prediction: 4 True Class: 4\n",
      "Test 77 Prediction: 4 True Class: 4\n",
      "Test 78 Prediction: 4 True Class: 4\n",
      "Test 79 Prediction: 4 True Class: 4\n",
      "Test 80 Prediction: 4 True Class: 4\n",
      "Test 81 Prediction: 4 True Class: 4\n",
      "Test 82 Prediction: 4 True Class: 4\n",
      "Test 83 Prediction: 4 True Class: 4\n",
      "Test 84 Prediction: 4 True Class: 4\n",
      "Test 85 Prediction: 4 True Class: 4\n",
      "Test 86 Prediction: 4 True Class: 4\n",
      "Test 87 Prediction: 4 True Class: 4\n",
      "Test 88 Prediction: 4 True Class: 4\n",
      "Test 89 Prediction: 4 True Class: 4\n",
      "Test 90 Prediction: 4 True Class: 4\n",
      "Test 91 Prediction: 4 True Class: 4\n",
      "Test 92 Prediction: 4 True Class: 3\n",
      "Test 93 Prediction: 4 True Class: 3\n",
      "Test 94 Prediction: 4 True Class: 3\n",
      "Test 95 Prediction: 4 True Class: 3\n",
      "Test 96 Prediction: 4 True Class: 3\n",
      "Test 97 Prediction: 4 True Class: 4\n",
      "Test 98 Prediction: 4 True Class: 4\n",
      "Test 99 Prediction: 4 True Class: 4\n",
      "Test 100 Prediction: 4 True Class: 4\n",
      "Test 101 Prediction: 4 True Class: 4\n",
      "Test 102 Prediction: 4 True Class: 4\n",
      "Test 103 Prediction: 4 True Class: 4\n",
      "Test 104 Prediction: 4 True Class: 4\n",
      "Test 105 Prediction: 4 True Class: 4\n",
      "Test 106 Prediction: 4 True Class: 4\n",
      "Test 107 Prediction: 4 True Class: 4\n",
      "Test 108 Prediction: 4 True Class: 4\n",
      "Test 109 Prediction: 4 True Class: 4\n",
      "Test 110 Prediction: 4 True Class: 4\n",
      "Test 111 Prediction: 5 True Class: 4\n",
      "Test 112 Prediction: 4 True Class: 4\n",
      "Test 113 Prediction: 6 True Class: 4\n",
      "Test 114 Prediction: 6 True Class: 5\n",
      "Test 115 Prediction: 5 True Class: 5\n",
      "Test 116 Prediction: 5 True Class: 5\n",
      "Test 117 Prediction: 5 True Class: 5\n",
      "Test 118 Prediction: 5 True Class: 5\n",
      "Test 119 Prediction: 5 True Class: 5\n",
      "Test 120 Prediction: 6 True Class: 5\n",
      "Test 121 Prediction: 5 True Class: 5\n",
      "Test 122 Prediction: 5 True Class: 5\n",
      "Test 123 Prediction: 5 True Class: 5\n",
      "Test 124 Prediction: 5 True Class: 4\n",
      "Test 125 Prediction: 5 True Class: 4\n",
      "Test 126 Prediction: 5 True Class: 4\n",
      "Test 127 Prediction: 5 True Class: 4\n",
      "Test 128 Prediction: 4 True Class: 4\n",
      "Test 129 Prediction: 4 True Class: 4\n",
      "Test 130 Prediction: 4 True Class: 4\n",
      "Test 131 Prediction: 4 True Class: 4\n",
      "Test 132 Prediction: 4 True Class: 4\n",
      "Test 133 Prediction: 4 True Class: 4\n",
      "Test 134 Prediction: 4 True Class: 4\n",
      "Test 135 Prediction: 4 True Class: 4\n",
      "Test 136 Prediction: 4 True Class: 4\n",
      "Test 137 Prediction: 4 True Class: 4\n",
      "Test 138 Prediction: 5 True Class: 4\n",
      "Test 139 Prediction: 5 True Class: 5\n",
      "Test 140 Prediction: 4 True Class: 4\n",
      "Test 141 Prediction: 4 True Class: 5\n",
      "Test 142 Prediction: 4 True Class: 5\n",
      "Test 143 Prediction: 4 True Class: 5\n",
      "Test 144 Prediction: 5 True Class: 5\n",
      "Test 145 Prediction: 5 True Class: 5\n",
      "Test 146 Prediction: 5 True Class: 5\n",
      "Test 147 Prediction: 5 True Class: 5\n",
      "Test 148 Prediction: 5 True Class: 5\n",
      "Test 149 Prediction: 4 True Class: 5\n",
      "Test 150 Prediction: 4 True Class: 5\n",
      "Test 151 Prediction: 4 True Class: 5\n",
      "Test 152 Prediction: 5 True Class: 5\n",
      "Test 153 Prediction: 4 True Class: 4\n",
      "Test 154 Prediction: 5 True Class: 5\n",
      "Test 155 Prediction: 5 True Class: 5\n",
      "Test 156 Prediction: 5 True Class: 5\n",
      "Test 157 Prediction: 4 True Class: 4\n",
      "Test 158 Prediction: 4 True Class: 4\n",
      "Test 159 Prediction: 4 True Class: 4\n",
      "Test 160 Prediction: 4 True Class: 4\n",
      "Test 161 Prediction: 4 True Class: 4\n",
      "Test 162 Prediction: 4 True Class: 4\n",
      "Test 163 Prediction: 5 True Class: 4\n",
      "Test 164 Prediction: 4 True Class: 4\n",
      "Test 165 Prediction: 5 True Class: 5\n",
      "Test 166 Prediction: 4 True Class: 5\n",
      "Test 167 Prediction: 4 True Class: 5\n",
      "Test 168 Prediction: 4 True Class: 5\n",
      "Test 169 Prediction: 4 True Class: 5\n",
      "Test 170 Prediction: 4 True Class: 5\n",
      "Test 171 Prediction: 4 True Class: 5\n",
      "Test 172 Prediction: 4 True Class: 5\n",
      "Test 173 Prediction: 4 True Class: 5\n",
      "Test 174 Prediction: 4 True Class: 5\n",
      "Test 175 Prediction: 4 True Class: 5\n",
      "Test 176 Prediction: 5 True Class: 5\n",
      "Test 177 Prediction: 4 True Class: 5\n",
      "Test 178 Prediction: 5 True Class: 5\n",
      "Test 179 Prediction: 5 True Class: 5\n",
      "Test 180 Prediction: 5 True Class: 5\n",
      "Test 181 Prediction: 5 True Class: 5\n",
      "Test 182 Prediction: 5 True Class: 5\n",
      "Test 183 Prediction: 5 True Class: 5\n",
      "Test 184 Prediction: 5 True Class: 5\n",
      "Test 185 Prediction: 5 True Class: 5\n",
      "Test 186 Prediction: 4 True Class: 5\n",
      "Test 187 Prediction: 5 True Class: 5\n",
      "Test 188 Prediction: 5 True Class: 5\n",
      "Test 189 Prediction: 4 True Class: 5\n",
      "Test 190 Prediction: 5 True Class: 5\n",
      "Test 191 Prediction: 5 True Class: 5\n",
      "Test 192 Prediction: 4 True Class: 5\n",
      "Test 193 Prediction: 4 True Class: 5\n",
      "Test 194 Prediction: 4 True Class: 5\n",
      "Test 195 Prediction: 4 True Class: 5\n",
      "Test 196 Prediction: 4 True Class: 5\n",
      "Test 197 Prediction: 5 True Class: 5\n",
      "Test 198 Prediction: 5 True Class: 5\n",
      "Test 199 Prediction: 4 True Class: 5\n",
      "Test 200 Prediction: 4 True Class: 5\n",
      "Test 201 Prediction: 4 True Class: 5\n",
      "Test 202 Prediction: 4 True Class: 5\n",
      "Test 203 Prediction: 5 True Class: 5\n",
      "Test 204 Prediction: 5 True Class: 4\n",
      "Test 205 Prediction: 5 True Class: 4\n",
      "Test 206 Prediction: 5 True Class: 4\n",
      "Test 207 Prediction: 5 True Class: 5\n",
      "Test 208 Prediction: 5 True Class: 4\n",
      "Test 209 Prediction: 5 True Class: 4\n",
      "Test 210 Prediction: 5 True Class: 4\n",
      "Test 211 Prediction: 5 True Class: 4\n",
      "Test 212 Prediction: 5 True Class: 4\n",
      "Test 213 Prediction: 5 True Class: 4\n",
      "Test 214 Prediction: 5 True Class: 4\n",
      "Test 215 Prediction: 5 True Class: 4\n",
      "Test 216 Prediction: 5 True Class: 4\n",
      "Test 217 Prediction: 5 True Class: 5\n",
      "Test 218 Prediction: 5 True Class: 5\n",
      "Test 219 Prediction: 5 True Class: 5\n",
      "Test 220 Prediction: 5 True Class: 5\n",
      "Test 221 Prediction: 4 True Class: 5\n",
      "Test 222 Prediction: 4 True Class: 5\n",
      "Test 223 Prediction: 4 True Class: 5\n",
      "Test 224 Prediction: 5 True Class: 5\n",
      "Test 225 Prediction: 4 True Class: 5\n",
      "Test 226 Prediction: 5 True Class: 5\n",
      "Test 227 Prediction: 5 True Class: 5\n",
      "Test 228 Prediction: 5 True Class: 5\n",
      "Test 229 Prediction: 5 True Class: 5\n",
      "Test 230 Prediction: 5 True Class: 5\n",
      "Test 231 Prediction: 5 True Class: 5\n",
      "Test 232 Prediction: 5 True Class: 5\n",
      "Test 233 Prediction: 5 True Class: 5\n",
      "Test 234 Prediction: 5 True Class: 5\n",
      "Test 235 Prediction: 6 True Class: 6\n",
      "Test 236 Prediction: 7 True Class: 8\n",
      "Test 237 Prediction: 7 True Class: 7\n",
      "Test 238 Prediction: 7 True Class: 7\n",
      "Test 239 Prediction: 7 True Class: 7\n",
      "Test 240 Prediction: 7 True Class: 7\n",
      "Test 241 Prediction: 7 True Class: 7\n",
      "Test 242 Prediction: 7 True Class: 6\n",
      "Test 243 Prediction: 7 True Class: 6\n",
      "Test 244 Prediction: 7 True Class: 6\n",
      "Test 245 Prediction: 7 True Class: 6\n",
      "Test 246 Prediction: 6 True Class: 6\n",
      "Test 247 Prediction: 6 True Class: 6\n",
      "Test 248 Prediction: 6 True Class: 6\n",
      "Test 249 Prediction: 7 True Class: 6\n",
      "Test 250 Prediction: 7 True Class: 6\n",
      "Test 251 Prediction: 7 True Class: 6\n",
      "Test 252 Prediction: 7 True Class: 6\n",
      "Test 253 Prediction: 7 True Class: 6\n",
      "Test 254 Prediction: 6 True Class: 6\n",
      "Test 255 Prediction: 6 True Class: 6\n",
      "Test 256 Prediction: 6 True Class: 6\n",
      "Test 257 Prediction: 7 True Class: 6\n",
      "Test 258 Prediction: 7 True Class: 6\n",
      "Test 259 Prediction: 7 True Class: 6\n",
      "Test 260 Prediction: 6 True Class: 6\n",
      "Test 261 Prediction: 6 True Class: 6\n",
      "Test 262 Prediction: 7 True Class: 6\n",
      "Test 263 Prediction: 6 True Class: 6\n",
      "Test 264 Prediction: 6 True Class: 6\n",
      "Test 265 Prediction: 6 True Class: 6\n",
      "Test 266 Prediction: 6 True Class: 6\n",
      "Test 267 Prediction: 7 True Class: 6\n",
      "Test 268 Prediction: 7 True Class: 6\n",
      "Test 269 Prediction: 7 True Class: 7\n",
      "Test 270 Prediction: 7 True Class: 7\n",
      "Test 271 Prediction: 7 True Class: 7\n",
      "Test 272 Prediction: 7 True Class: 7\n",
      "Test 273 Prediction: 6 True Class: 7\n",
      "Test 274 Prediction: 6 True Class: 7\n",
      "Test 275 Prediction: 7 True Class: 7\n",
      "Test 276 Prediction: 7 True Class: 7\n",
      "Test 277 Prediction: 7 True Class: 6\n",
      "Test 278 Prediction: 7 True Class: 7\n",
      "Test 279 Prediction: 7 True Class: 6\n",
      "Test 280 Prediction: 6 True Class: 6\n",
      "Test 281 Prediction: 7 True Class: 6\n",
      "Test 282 Prediction: 6 True Class: 6\n",
      "Test 283 Prediction: 6 True Class: 6\n",
      "Test 284 Prediction: 7 True Class: 7\n",
      "Test 285 Prediction: 7 True Class: 7\n",
      "Test 286 Prediction: 8 True Class: 8\n",
      "Test 287 Prediction: 9 True Class: 8\n",
      "Test 288 Prediction: 8 True Class: 8\n",
      "Test 289 Prediction: 9 True Class: 8\n",
      "Test 290 Prediction: 8 True Class: 8\n",
      "Test 291 Prediction: 9 True Class: 8\n",
      "Test 292 Prediction: 9 True Class: 8\n",
      "Test 293 Prediction: 9 True Class: 8\n",
      "Test 294 Prediction: 9 True Class: 8\n",
      "Test 295 Prediction: 9 True Class: 9\n",
      "Test 296 Prediction: 9 True Class: 8\n",
      "Test 297 Prediction: 9 True Class: 8\n",
      "Test 298 Prediction: 9 True Class: 9\n",
      "Test 299 Prediction: 9 True Class: 9\n",
      "Test 300 Prediction: 9 True Class: 9\n",
      "Test 301 Prediction: 9 True Class: 9\n",
      "Test 302 Prediction: 9 True Class: 9\n",
      "Test 303 Prediction: 9 True Class: 9\n",
      "Test 304 Prediction: 7 True Class: 8\n",
      "Test 305 Prediction: 7 True Class: 8\n",
      "Test 306 Prediction: 7 True Class: 8\n",
      "Test 307 Prediction: 7 True Class: 8\n",
      "Test 308 Prediction: 7 True Class: 8\n",
      "Test 309 Prediction: 7 True Class: 8\n",
      "Test 310 Prediction: 7 True Class: 7\n",
      "Test 311 Prediction: 7 True Class: 7\n",
      "Test 312 Prediction: 7 True Class: 7\n",
      "Test 313 Prediction: 7 True Class: 7\n",
      "Test 314 Prediction: 7 True Class: 7\n",
      "Test 315 Prediction: 7 True Class: 7\n",
      "Test 316 Prediction: 8 True Class: 7\n",
      "Test 317 Prediction: 8 True Class: 7\n",
      "Test 318 Prediction: 8 True Class: 7\n",
      "Test 319 Prediction: 7 True Class: 7\n",
      "Test 320 Prediction: 8 True Class: 7\n",
      "Test 321 Prediction: 8 True Class: 7\n",
      "Test 322 Prediction: 8 True Class: 7\n",
      "Test 323 Prediction: 8 True Class: 8\n",
      "Test 324 Prediction: 8 True Class: 8\n",
      "Test 325 Prediction: 9 True Class: 8\n",
      "Test 326 Prediction: 9 True Class: 8\n",
      "Test 327 Prediction: 8 True Class: 8\n",
      "Test 328 Prediction: 8 True Class: 8\n",
      "Test 329 Prediction: 9 True Class: 8\n",
      "Test 330 Prediction: 7 True Class: 8\n",
      "Test 331 Prediction: 6 True Class: 8\n",
      "Test 332 Prediction: 6 True Class: 8\n",
      "Test 333 Prediction: 9 True Class: 8\n",
      "Test 334 Prediction: 9 True Class: 8\n",
      "Test 335 Prediction: 9 True Class: 8\n",
      "Test 336 Prediction: 9 True Class: 8\n",
      "Test 337 Prediction: 9 True Class: 8\n",
      "Test 338 Prediction: 9 True Class: 8\n",
      "Test 339 Prediction: 7 True Class: 8\n",
      "Test 340 Prediction: 7 True Class: 8\n",
      "Test 341 Prediction: 7 True Class: 8\n",
      "Test 342 Prediction: 8 True Class: 8\n",
      "Test 343 Prediction: 8 True Class: 8\n",
      "Test 344 Prediction: 8 True Class: 8\n",
      "Test 345 Prediction: 8 True Class: 8\n",
      "Test 346 Prediction: 7 True Class: 7\n",
      "Test 347 Prediction: 7 True Class: 8\n",
      "Test 348 Prediction: 7 True Class: 7\n",
      "Test 349 Prediction: 7 True Class: 7\n",
      "Test 350 Prediction: 7 True Class: 7\n",
      "Test 351 Prediction: 7 True Class: 7\n",
      "Test 352 Prediction: 7 True Class: 7\n",
      "Test 353 Prediction: 7 True Class: 7\n",
      "Test 354 Prediction: 7 True Class: 7\n",
      "Test 355 Prediction: 7 True Class: 7\n",
      "Test 356 Prediction: 7 True Class: 7\n",
      "Test 357 Prediction: 7 True Class: 7\n",
      "Test 358 Prediction: 7 True Class: 7\n",
      "Test 359 Prediction: 7 True Class: 7\n",
      "Test 360 Prediction: 7 True Class: 7\n",
      "Test 361 Prediction: 6 True Class: 7\n",
      "Test 362 Prediction: 6 True Class: 7\n",
      "Test 363 Prediction: 7 True Class: 7\n",
      "Test 364 Prediction: 6 True Class: 7\n",
      "Test 365 Prediction: 6 True Class: 7\n",
      "Test 366 Prediction: 7 True Class: 7\n",
      "Test 367 Prediction: 7 True Class: 7\n",
      "Test 368 Prediction: 7 True Class: 7\n",
      "Test 369 Prediction: 7 True Class: 7\n",
      "Test 370 Prediction: 7 True Class: 7\n",
      "Test 371 Prediction: 7 True Class: 7\n",
      "Test 372 Prediction: 7 True Class: 7\n",
      "Test 373 Prediction: 7 True Class: 7\n",
      "Test 374 Prediction: 7 True Class: 7\n",
      "Test 375 Prediction: 7 True Class: 7\n",
      "Test 376 Prediction: 7 True Class: 7\n",
      "Test 377 Prediction: 7 True Class: 7\n",
      "Test 378 Prediction: 7 True Class: 7\n",
      "Test 379 Prediction: 7 True Class: 7\n",
      "Test 380 Prediction: 7 True Class: 7\n",
      "Test 381 Prediction: 7 True Class: 7\n",
      "Test 382 Prediction: 7 True Class: 6\n",
      "Test 383 Prediction: 7 True Class: 7\n",
      "Test 384 Prediction: 7 True Class: 7\n",
      "Test 385 Prediction: 7 True Class: 7\n",
      "Test 386 Prediction: 7 True Class: 7\n",
      "Test 387 Prediction: 7 True Class: 7\n",
      "Test 388 Prediction: 7 True Class: 7\n",
      "Test 389 Prediction: 7 True Class: 7\n",
      "Test 390 Prediction: 7 True Class: 7\n",
      "Test 391 Prediction: 7 True Class: 7\n",
      "Test 392 Prediction: 7 True Class: 7\n",
      "Test 393 Prediction: 7 True Class: 7\n",
      "Test 394 Prediction: 6 True Class: 7\n",
      "Test 395 Prediction: 6 True Class: 6\n",
      "Test 396 Prediction: 6 True Class: 6\n",
      "Test 397 Prediction: 6 True Class: 6\n",
      "Test 398 Prediction: 6 True Class: 6\n",
      "Test 399 Prediction: 6 True Class: 6\n",
      "Test 400 Prediction: 6 True Class: 6\n",
      "Test 401 Prediction: 6 True Class: 6\n",
      "Test 402 Prediction: 6 True Class: 6\n",
      "Test 403 Prediction: 6 True Class: 6\n",
      "Test 404 Prediction: 6 True Class: 6\n",
      "Test 405 Prediction: 6 True Class: 6\n",
      "Test 406 Prediction: 6 True Class: 6\n",
      "Test 407 Prediction: 6 True Class: 6\n",
      "Test 408 Prediction: 7 True Class: 6\n",
      "Test 409 Prediction: 7 True Class: 6\n",
      "Test 410 Prediction: 6 True Class: 6\n",
      "Test 411 Prediction: 6 True Class: 6\n",
      "Test 412 Prediction: 6 True Class: 7\n",
      "Test 413 Prediction: 6 True Class: 7\n",
      "Test 414 Prediction: 6 True Class: 6\n",
      "Test 415 Prediction: 6 True Class: 6\n",
      "Test 416 Prediction: 6 True Class: 6\n",
      "Test 417 Prediction: 6 True Class: 7\n",
      "Test 418 Prediction: 6 True Class: 6\n",
      "Test 419 Prediction: 6 True Class: 7\n",
      "Test 420 Prediction: 6 True Class: 7\n",
      "Test 421 Prediction: 7 True Class: 7\n",
      "Test 422 Prediction: 7 True Class: 6\n",
      "Test 423 Prediction: 7 True Class: 6\n",
      "Test 424 Prediction: 6 True Class: 6\n",
      "Test 425 Prediction: 6 True Class: 7\n",
      "Test 426 Prediction: 7 True Class: 7\n",
      "Test 427 Prediction: 7 True Class: 7\n",
      "Test 428 Prediction: 7 True Class: 7\n",
      "Test 429 Prediction: 7 True Class: 7\n",
      "Test 430 Prediction: 7 True Class: 7\n",
      "Test 431 Prediction: 6 True Class: 7\n",
      "Test 432 Prediction: 6 True Class: 7\n",
      "Test 433 Prediction: 6 True Class: 7\n",
      "Test 434 Prediction: 6 True Class: 7\n",
      "Test 435 Prediction: 6 True Class: 7\n",
      "Test 436 Prediction: 7 True Class: 7\n",
      "Test 437 Prediction: 7 True Class: 7\n",
      "Test 438 Prediction: 7 True Class: 7\n",
      "Test 439 Prediction: 7 True Class: 7\n",
      "Test 440 Prediction: 7 True Class: 7\n",
      "Test 441 Prediction: 7 True Class: 7\n",
      "Test 442 Prediction: 6 True Class: 7\n",
      "Test 443 Prediction: 6 True Class: 6\n",
      "Test 444 Prediction: 6 True Class: 7\n",
      "Test 445 Prediction: 6 True Class: 7\n",
      "Test 446 Prediction: 6 True Class: 7\n",
      "Test 447 Prediction: 6 True Class: 7\n",
      "Test 448 Prediction: 6 True Class: 6\n",
      "Test 449 Prediction: 6 True Class: 6\n",
      "Test 450 Prediction: 7 True Class: 6\n",
      "Test 451 Prediction: 7 True Class: 6\n",
      "Test 452 Prediction: 7 True Class: 6\n",
      "Test 453 Prediction: 6 True Class: 6\n",
      "Test 454 Prediction: 6 True Class: 6\n",
      "Test 455 Prediction: 6 True Class: 6\n",
      "Test 456 Prediction: 6 True Class: 6\n",
      "Test 457 Prediction: 6 True Class: 6\n",
      "Test 458 Prediction: 6 True Class: 6\n",
      "Test 459 Prediction: 6 True Class: 6\n",
      "Test 460 Prediction: 6 True Class: 6\n",
      "Test 461 Prediction: 7 True Class: 6\n",
      "Test 462 Prediction: 6 True Class: 6\n",
      "Test 463 Prediction: 6 True Class: 6\n",
      "Test 464 Prediction: 5 True Class: 6\n",
      "Test 465 Prediction: 5 True Class: 6\n",
      "Test 466 Prediction: 5 True Class: 6\n",
      "Test 467 Prediction: 5 True Class: 6\n",
      "Test 468 Prediction: 5 True Class: 6\n",
      "Test 469 Prediction: 6 True Class: 6\n",
      "Test 470 Prediction: 4 True Class: 6\n",
      "Test 471 Prediction: 4 True Class: 6\n",
      "Test 472 Prediction: 6 True Class: 6\n",
      "Test 473 Prediction: 6 True Class: 6\n",
      "Test 474 Prediction: 6 True Class: 6\n",
      "Test 475 Prediction: 6 True Class: 6\n",
      "Test 476 Prediction: 6 True Class: 6\n",
      "Test 477 Prediction: 6 True Class: 6\n",
      "Test 478 Prediction: 6 True Class: 6\n",
      "Test 479 Prediction: 6 True Class: 6\n",
      "Test 480 Prediction: 6 True Class: 6\n",
      "Test 481 Prediction: 6 True Class: 6\n",
      "Test 482 Prediction: 6 True Class: 6\n",
      "Test 483 Prediction: 6 True Class: 6\n",
      "Test 484 Prediction: 6 True Class: 6\n",
      "Test 485 Prediction: 7 True Class: 6\n",
      "Test 486 Prediction: 6 True Class: 6\n",
      "Test 487 Prediction: 7 True Class: 6\n",
      "Test 488 Prediction: 7 True Class: 6\n",
      "Test 489 Prediction: 6 True Class: 6\n",
      "Test 490 Prediction: 6 True Class: 6\n",
      "Test 491 Prediction: 6 True Class: 6\n",
      "Test 492 Prediction: 6 True Class: 6\n",
      "Test 493 Prediction: 6 True Class: 6\n",
      "Test 494 Prediction: 6 True Class: 6\n",
      "Test 495 Prediction: 6 True Class: 6\n",
      "Test 496 Prediction: 6 True Class: 6\n",
      "Test 497 Prediction: 6 True Class: 6\n",
      "Test 498 Prediction: 5 True Class: 6\n",
      "Test 499 Prediction: 5 True Class: 6\n",
      "Test 500 Prediction: 7 True Class: 6\n",
      "Test 501 Prediction: 6 True Class: 6\n",
      "Test 502 Prediction: 5 True Class: 6\n",
      "Test 503 Prediction: 5 True Class: 6\n",
      "Test 504 Prediction: 5 True Class: 6\n",
      "Test 505 Prediction: 5 True Class: 6\n",
      "Test 506 Prediction: 5 True Class: 6\n",
      "Test 507 Prediction: 5 True Class: 6\n",
      "Test 508 Prediction: 5 True Class: 6\n",
      "Test 509 Prediction: 5 True Class: 6\n",
      "Test 510 Prediction: 5 True Class: 6\n",
      "Test 511 Prediction: 5 True Class: 6\n",
      "Test 512 Prediction: 5 True Class: 6\n",
      "Test 513 Prediction: 6 True Class: 6\n",
      "Test 514 Prediction: 6 True Class: 6\n",
      "Test 515 Prediction: 6 True Class: 6\n",
      "Test 516 Prediction: 6 True Class: 6\n",
      "Test 517 Prediction: 6 True Class: 6\n",
      "Test 518 Prediction: 6 True Class: 6\n",
      "Test 519 Prediction: 6 True Class: 6\n",
      "Test 520 Prediction: 6 True Class: 6\n",
      "Test 521 Prediction: 6 True Class: 6\n",
      "Test 522 Prediction: 6 True Class: 6\n",
      "Test 523 Prediction: 6 True Class: 6\n",
      "Test 524 Prediction: 6 True Class: 6\n",
      "Test 525 Prediction: 5 True Class: 6\n",
      "Test 526 Prediction: 5 True Class: 6\n",
      "Test 527 Prediction: 5 True Class: 6\n",
      "Test 528 Prediction: 7 True Class: 6\n",
      "Test 529 Prediction: 7 True Class: 6\n",
      "Test 530 Prediction: 7 True Class: 6\n",
      "Test 531 Prediction: 6 True Class: 6\n",
      "Test 532 Prediction: 6 True Class: 6\n",
      "Test 533 Prediction: 6 True Class: 6\n",
      "Test 534 Prediction: 6 True Class: 6\n",
      "Test 535 Prediction: 6 True Class: 6\n",
      "Test 536 Prediction: 6 True Class: 6\n",
      "Test 537 Prediction: 6 True Class: 6\n",
      "Test 538 Prediction: 6 True Class: 6\n",
      "Test 539 Prediction: 6 True Class: 6\n",
      "Test 540 Prediction: 6 True Class: 6\n",
      "Test 541 Prediction: 6 True Class: 6\n",
      "Test 542 Prediction: 7 True Class: 6\n",
      "Test 543 Prediction: 6 True Class: 6\n",
      "Test 544 Prediction: 6 True Class: 6\n",
      "Test 545 Prediction: 6 True Class: 6\n",
      "Test 546 Prediction: 7 True Class: 6\n",
      "Test 547 Prediction: 6 True Class: 6\n",
      "Test 548 Prediction: 6 True Class: 6\n",
      "Test 549 Prediction: 6 True Class: 6\n",
      "Test 550 Prediction: 7 True Class: 6\n",
      "Test 551 Prediction: 6 True Class: 6\n",
      "Test 552 Prediction: 7 True Class: 7\n",
      "Test 553 Prediction: 7 True Class: 7\n",
      "Test 554 Prediction: 7 True Class: 7\n",
      "Test 555 Prediction: 7 True Class: 7\n",
      "Test 556 Prediction: 7 True Class: 7\n",
      "Test 557 Prediction: 7 True Class: 7\n",
      "Test 558 Prediction: 7 True Class: 7\n",
      "Test 559 Prediction: 6 True Class: 6\n",
      "Test 560 Prediction: 6 True Class: 6\n",
      "Test 561 Prediction: 6 True Class: 6\n",
      "Test 562 Prediction: 6 True Class: 6\n",
      "Test 563 Prediction: 6 True Class: 6\n",
      "Test 564 Prediction: 6 True Class: 6\n",
      "Test 565 Prediction: 6 True Class: 6\n",
      "Test 566 Prediction: 6 True Class: 6\n",
      "Test 567 Prediction: 6 True Class: 6\n",
      "Test 568 Prediction: 6 True Class: 6\n",
      "Test 569 Prediction: 6 True Class: 6\n",
      "Test 570 Prediction: 6 True Class: 6\n",
      "Test 571 Prediction: 6 True Class: 6\n",
      "Test 572 Prediction: 6 True Class: 6\n",
      "Test 573 Prediction: 6 True Class: 6\n",
      "Test 574 Prediction: 6 True Class: 6\n",
      "Test 575 Prediction: 7 True Class: 6\n",
      "Test 576 Prediction: 6 True Class: 6\n",
      "Test 577 Prediction: 7 True Class: 6\n",
      "Test 578 Prediction: 7 True Class: 6\n",
      "Test 579 Prediction: 7 True Class: 6\n",
      "Test 580 Prediction: 7 True Class: 6\n",
      "Test 581 Prediction: 7 True Class: 6\n",
      "Test 582 Prediction: 7 True Class: 6\n",
      "Test 583 Prediction: 7 True Class: 6\n",
      "Test 584 Prediction: 6 True Class: 6\n",
      "Test 585 Prediction: 6 True Class: 6\n",
      "Test 586 Prediction: 6 True Class: 6\n",
      "Test 587 Prediction: 6 True Class: 6\n",
      "Test 588 Prediction: 6 True Class: 6\n",
      "Test 589 Prediction: 6 True Class: 6\n",
      "Test 590 Prediction: 6 True Class: 6\n",
      "Test 591 Prediction: 7 True Class: 6\n",
      "Test 592 Prediction: 6 True Class: 6\n",
      "Test 593 Prediction: 7 True Class: 6\n",
      "Test 594 Prediction: 6 True Class: 6\n",
      "Test 595 Prediction: 7 True Class: 6\n",
      "Test 596 Prediction: 6 True Class: 6\n",
      "Test 597 Prediction: 6 True Class: 6\n",
      "Test 598 Prediction: 6 True Class: 6\n",
      "Test 599 Prediction: 6 True Class: 6\n",
      "Test 600 Prediction: 6 True Class: 6\n",
      "Test 601 Prediction: 6 True Class: 6\n",
      "Test 602 Prediction: 6 True Class: 6\n",
      "Test 603 Prediction: 7 True Class: 7\n",
      "Test 604 Prediction: 6 True Class: 7\n",
      "Test 605 Prediction: 6 True Class: 7\n",
      "Test 606 Prediction: 6 True Class: 7\n",
      "Test 607 Prediction: 6 True Class: 7\n",
      "Test 608 Prediction: 7 True Class: 7\n",
      "Test 609 Prediction: 7 True Class: 7\n",
      "Test 610 Prediction: 6 True Class: 6\n",
      "Test 611 Prediction: 6 True Class: 6\n",
      "Test 612 Prediction: 6 True Class: 6\n",
      "Test 613 Prediction: 6 True Class: 6\n",
      "Test 614 Prediction: 6 True Class: 6\n",
      "Test 615 Prediction: 6 True Class: 6\n",
      "Test 616 Prediction: 7 True Class: 6\n",
      "Test 617 Prediction: 6 True Class: 6\n",
      "Test 618 Prediction: 6 True Class: 6\n",
      "Test 619 Prediction: 6 True Class: 7\n",
      "Test 620 Prediction: 6 True Class: 6\n",
      "Test 621 Prediction: 6 True Class: 6\n",
      "Test 622 Prediction: 6 True Class: 6\n",
      "Test 623 Prediction: 6 True Class: 6\n",
      "Test 624 Prediction: 6 True Class: 6\n",
      "Test 625 Prediction: 6 True Class: 6\n",
      "Test 626 Prediction: 6 True Class: 6\n",
      "Test 627 Prediction: 6 True Class: 6\n",
      "Test 628 Prediction: 6 True Class: 6\n",
      "Test 629 Prediction: 6 True Class: 6\n",
      "Test 630 Prediction: 6 True Class: 6\n",
      "Test 631 Prediction: 4 True Class: 6\n",
      "Test 632 Prediction: 6 True Class: 6\n",
      "Test 633 Prediction: 6 True Class: 6\n",
      "Test 634 Prediction: 6 True Class: 6\n",
      "Test 635 Prediction: 6 True Class: 6\n",
      "Test 636 Prediction: 6 True Class: 6\n",
      "Test 637 Prediction: 6 True Class: 6\n",
      "Test 638 Prediction: 6 True Class: 6\n",
      "Test 639 Prediction: 6 True Class: 6\n",
      "Test 640 Prediction: 6 True Class: 6\n",
      "Test 641 Prediction: 6 True Class: 6\n",
      "Test 642 Prediction: 6 True Class: 6\n",
      "Test 643 Prediction: 6 True Class: 6\n",
      "Test 644 Prediction: 6 True Class: 6\n",
      "Test 645 Prediction: 6 True Class: 6\n",
      "Test 646 Prediction: 5 True Class: 6\n",
      "Test 647 Prediction: 5 True Class: 6\n",
      "Test 648 Prediction: 5 True Class: 6\n",
      "Test 649 Prediction: 6 True Class: 6\n",
      "Test 650 Prediction: 6 True Class: 6\n",
      "Test 651 Prediction: 6 True Class: 6\n",
      "Test 652 Prediction: 6 True Class: 6\n",
      "Test 653 Prediction: 6 True Class: 6\n",
      "Test 654 Prediction: 6 True Class: 6\n",
      "Test 655 Prediction: 7 True Class: 6\n",
      "Test 656 Prediction: 6 True Class: 6\n",
      "Test 657 Prediction: 6 True Class: 6\n",
      "Test 658 Prediction: 6 True Class: 6\n",
      "Test 659 Prediction: 6 True Class: 6\n",
      "Test 660 Prediction: 6 True Class: 6\n",
      "Test 661 Prediction: 6 True Class: 6\n",
      "Test 662 Prediction: 6 True Class: 6\n",
      "Test 663 Prediction: 6 True Class: 6\n",
      "Test 664 Prediction: 6 True Class: 6\n",
      "Test 665 Prediction: 6 True Class: 7\n",
      "Test 666 Prediction: 4 True Class: 6\n",
      "Test 667 Prediction: 4 True Class: 6\n",
      "Test 668 Prediction: 5 True Class: 6\n",
      "Test 669 Prediction: 5 True Class: 6\n",
      "Test 670 Prediction: 5 True Class: 6\n",
      "Test 671 Prediction: 5 True Class: 6\n",
      "Test 672 Prediction: 5 True Class: 6\n",
      "Test 673 Prediction: 5 True Class: 6\n",
      "Test 674 Prediction: 5 True Class: 6\n",
      "Test 675 Prediction: 5 True Class: 6\n",
      "Test 676 Prediction: 5 True Class: 6\n",
      "Test 677 Prediction: 6 True Class: 6\n",
      "Test 678 Prediction: 7 True Class: 6\n",
      "Test 679 Prediction: 7 True Class: 6\n",
      "Test 680 Prediction: 7 True Class: 7\n",
      "Test 681 Prediction: 7 True Class: 7\n",
      "Test 682 Prediction: 7 True Class: 7\n",
      "Test 683 Prediction: 7 True Class: 7\n",
      "Test 684 Prediction: 7 True Class: 7\n",
      "Test 685 Prediction: 7 True Class: 7\n",
      "Test 686 Prediction: 6 True Class: 7\n",
      "Test 687 Prediction: 6 True Class: 7\n",
      "Test 688 Prediction: 6 True Class: 7\n",
      "Test 689 Prediction: 8 True Class: 7\n",
      "Test 690 Prediction: 6 True Class: 7\n",
      "Test 691 Prediction: 8 True Class: 7\n",
      "Test 692 Prediction: 6 True Class: 7\n",
      "Test 693 Prediction: 8 True Class: 7\n",
      "Test 694 Prediction: 6 True Class: 7\n",
      "Test 695 Prediction: 6 True Class: 7\n",
      "Test 696 Prediction: 7 True Class: 7\n",
      "Test 697 Prediction: 7 True Class: 7\n",
      "Test 698 Prediction: 8 True Class: 7\n",
      "Test 699 Prediction: 6 True Class: 7\n",
      "Test 700 Prediction: 6 True Class: 7\n",
      "Test 701 Prediction: 6 True Class: 7\n",
      "Test 702 Prediction: 6 True Class: 7\n",
      "Test 703 Prediction: 6 True Class: 7\n",
      "Test 704 Prediction: 6 True Class: 6\n",
      "Test 705 Prediction: 6 True Class: 6\n",
      "Test 706 Prediction: 4 True Class: 6\n",
      "Test 707 Prediction: 8 True Class: 6\n",
      "Test 708 Prediction: 8 True Class: 6\n",
      "Test 709 Prediction: 8 True Class: 6\n",
      "Test 710 Prediction: 8 True Class: 6\n",
      "Test 711 Prediction: 8 True Class: 6\n",
      "Test 712 Prediction: 8 True Class: 6\n",
      "Test 713 Prediction: 8 True Class: 6\n",
      "Test 714 Prediction: 8 True Class: 6\n",
      "Test 715 Prediction: 8 True Class: 6\n",
      "Test 716 Prediction: 8 True Class: 6\n",
      "Test 717 Prediction: 8 True Class: 6\n",
      "Test 718 Prediction: 8 True Class: 6\n",
      "Test 719 Prediction: 7 True Class: 7\n",
      "Test 720 Prediction: 6 True Class: 7\n",
      "Test 721 Prediction: 7 True Class: 7\n",
      "Test 722 Prediction: 7 True Class: 7\n",
      "Test 723 Prediction: 7 True Class: 7\n",
      "Test 724 Prediction: 6 True Class: 7\n",
      "Test 725 Prediction: 6 True Class: 7\n",
      "Test 726 Prediction: 6 True Class: 6\n",
      "Test 727 Prediction: 6 True Class: 7\n",
      "Test 728 Prediction: 7 True Class: 6\n",
      "Test 729 Prediction: 7 True Class: 6\n",
      "Test 730 Prediction: 6 True Class: 6\n",
      "Test 731 Prediction: 6 True Class: 6\n",
      "Test 732 Prediction: 6 True Class: 6\n",
      "Test 733 Prediction: 6 True Class: 6\n",
      "Test 734 Prediction: 6 True Class: 6\n",
      "Test 735 Prediction: 6 True Class: 6\n",
      "Test 736 Prediction: 6 True Class: 6\n",
      "Test 737 Prediction: 6 True Class: 6\n",
      "Test 738 Prediction: 6 True Class: 6\n",
      "Test 739 Prediction: 6 True Class: 6\n",
      "Test 740 Prediction: 6 True Class: 6\n",
      "Test 741 Prediction: 7 True Class: 6\n",
      "Test 742 Prediction: 7 True Class: 6\n",
      "Test 743 Prediction: 6 True Class: 6\n",
      "Test 744 Prediction: 6 True Class: 6\n",
      "Test 745 Prediction: 6 True Class: 6\n",
      "Test 746 Prediction: 6 True Class: 6\n",
      "Test 747 Prediction: 6 True Class: 6\n",
      "Test 748 Prediction: 7 True Class: 6\n",
      "Test 749 Prediction: 7 True Class: 6\n",
      "Test 750 Prediction: 6 True Class: 6\n",
      "Test 751 Prediction: 6 True Class: 6\n",
      "Test 752 Prediction: 7 True Class: 6\n",
      "Test 753 Prediction: 7 True Class: 6\n",
      "Test 754 Prediction: 7 True Class: 6\n",
      "Test 755 Prediction: 7 True Class: 7\n",
      "Test 756 Prediction: 7 True Class: 7\n",
      "Test 757 Prediction: 7 True Class: 7\n",
      "Test 758 Prediction: 7 True Class: 7\n",
      "Test 759 Prediction: 6 True Class: 7\n",
      "Test 760 Prediction: 6 True Class: 6\n",
      "Test 761 Prediction: 6 True Class: 6\n",
      "Test 762 Prediction: 6 True Class: 6\n",
      "Test 763 Prediction: 6 True Class: 6\n",
      "Test 764 Prediction: 6 True Class: 6\n",
      "Test 765 Prediction: 6 True Class: 6\n",
      "Test 766 Prediction: 6 True Class: 6\n",
      "Test 767 Prediction: 6 True Class: 6\n",
      "Test 768 Prediction: 7 True Class: 6\n",
      "Test 769 Prediction: 7 True Class: 6\n",
      "Test 770 Prediction: 6 True Class: 6\n",
      "Test 771 Prediction: 6 True Class: 6\n",
      "Test 772 Prediction: 6 True Class: 6\n",
      "Test 773 Prediction: 6 True Class: 5\n",
      "Test 774 Prediction: 6 True Class: 5\n",
      "Test 775 Prediction: 6 True Class: 5\n",
      "Test 776 Prediction: 6 True Class: 5\n",
      "Test 777 Prediction: 7 True Class: 5\n",
      "Test 778 Prediction: 7 True Class: 5\n",
      "Test 779 Prediction: 6 True Class: 5\n",
      "Test 780 Prediction: 5 True Class: 5\n",
      "Test 781 Prediction: 5 True Class: 5\n",
      "Test 782 Prediction: 5 True Class: 5\n",
      "Test 783 Prediction: 5 True Class: 5\n",
      "Test 784 Prediction: 5 True Class: 5\n",
      "Test 785 Prediction: 5 True Class: 5\n",
      "Test 786 Prediction: 5 True Class: 4\n",
      "Test 787 Prediction: 5 True Class: 4\n",
      "Test 788 Prediction: 5 True Class: 4\n",
      "Test 789 Prediction: 5 True Class: 4\n",
      "Test 790 Prediction: 5 True Class: 4\n",
      "Test 791 Prediction: 4 True Class: 4\n",
      "Test 792 Prediction: 2 True Class: 4\n",
      "Test 793 Prediction: 2 True Class: 4\n",
      "Test 794 Prediction: 5 True Class: 4\n",
      "Test 795 Prediction: 4 True Class: 4\n",
      "Test 796 Prediction: 4 True Class: 4\n",
      "Test 797 Prediction: 4 True Class: 4\n",
      "Test 798 Prediction: 4 True Class: 4\n",
      "Test 799 Prediction: 4 True Class: 4\n",
      "Test 800 Prediction: 4 True Class: 4\n",
      "Test 801 Prediction: 4 True Class: 4\n",
      "Test 802 Prediction: 4 True Class: 4\n",
      "Test 803 Prediction: 4 True Class: 4\n",
      "Test 804 Prediction: 4 True Class: 4\n",
      "Test 805 Prediction: 5 True Class: 4\n",
      "Test 806 Prediction: 5 True Class: 4\n",
      "Test 807 Prediction: 5 True Class: 4\n",
      "Test 808 Prediction: 5 True Class: 4\n",
      "Test 809 Prediction: 4 True Class: 4\n",
      "Test 810 Prediction: 4 True Class: 4\n",
      "Test 811 Prediction: 4 True Class: 5\n",
      "Test 812 Prediction: 4 True Class: 5\n",
      "Test 813 Prediction: 4 True Class: 4\n",
      "Test 814 Prediction: 5 True Class: 4\n",
      "Test 815 Prediction: 4 True Class: 4\n",
      "Test 816 Prediction: 4 True Class: 4\n",
      "Test 817 Prediction: 4 True Class: 4\n",
      "Test 818 Prediction: 4 True Class: 4\n",
      "Test 819 Prediction: 3 True Class: 4\n",
      "Test 820 Prediction: 3 True Class: 4\n",
      "Test 821 Prediction: 3 True Class: 4\n",
      "Test 822 Prediction: 3 True Class: 4\n",
      "Test 823 Prediction: 3 True Class: 4\n",
      "Test 824 Prediction: 3 True Class: 4\n",
      "Test 825 Prediction: 3 True Class: 4\n",
      "Test 826 Prediction: 3 True Class: 4\n",
      "Test 827 Prediction: 4 True Class: 4\n",
      "Test 828 Prediction: 4 True Class: 4\n",
      "Test 829 Prediction: 4 True Class: 4\n",
      "Test 830 Prediction: 4 True Class: 4\n",
      "Test 831 Prediction: 4 True Class: 4\n",
      "Test 832 Prediction: 4 True Class: 4\n",
      "Test 833 Prediction: 4 True Class: 4\n",
      "Test 834 Prediction: 4 True Class: 4\n",
      "Test 835 Prediction: 4 True Class: 4\n",
      "Test 836 Prediction: 4 True Class: 4\n",
      "Test 837 Prediction: 4 True Class: 4\n",
      "Test 838 Prediction: 4 True Class: 4\n",
      "Test 839 Prediction: 4 True Class: 4\n",
      "Test 840 Prediction: 4 True Class: 4\n",
      "Test 841 Prediction: 4 True Class: 4\n",
      "Test 842 Prediction: 5 True Class: 4\n",
      "Test 843 Prediction: 4 True Class: 4\n",
      "Test 844 Prediction: 4 True Class: 4\n",
      "Test 845 Prediction: 4 True Class: 4\n",
      "Test 846 Prediction: 4 True Class: 4\n",
      "Test 847 Prediction: 4 True Class: 4\n",
      "Test 848 Prediction: 4 True Class: 4\n",
      "Test 849 Prediction: 4 True Class: 4\n",
      "Test 850 Prediction: 4 True Class: 4\n",
      "Test 851 Prediction: 4 True Class: 4\n",
      "Test 852 Prediction: 4 True Class: 4\n",
      "Test 853 Prediction: 4 True Class: 4\n",
      "Test 854 Prediction: 4 True Class: 4\n",
      "Test 855 Prediction: 4 True Class: 4\n",
      "Test 856 Prediction: 4 True Class: 4\n",
      "Test 857 Prediction: 4 True Class: 4\n",
      "Test 858 Prediction: 4 True Class: 4\n",
      "Test 859 Prediction: 4 True Class: 4\n",
      "Test 860 Prediction: 4 True Class: 4\n",
      "Test 861 Prediction: 4 True Class: 4\n",
      "Test 862 Prediction: 4 True Class: 4\n",
      "Test 863 Prediction: 4 True Class: 4\n",
      "Test 864 Prediction: 4 True Class: 4\n",
      "Test 865 Prediction: 4 True Class: 4\n",
      "Test 866 Prediction: 4 True Class: 4\n",
      "Test 867 Prediction: 4 True Class: 4\n",
      "Test 868 Prediction: 4 True Class: 4\n",
      "Test 869 Prediction: 4 True Class: 4\n",
      "Test 870 Prediction: 5 True Class: 4\n",
      "Test 871 Prediction: 5 True Class: 4\n",
      "Test 872 Prediction: 5 True Class: 4\n",
      "Test 873 Prediction: 5 True Class: 5\n",
      "Test 874 Prediction: 5 True Class: 4\n",
      "Test 875 Prediction: 4 True Class: 4\n",
      "Test 876 Prediction: 4 True Class: 4\n",
      "Test 877 Prediction: 5 True Class: 5\n",
      "Test 878 Prediction: 5 True Class: 5\n",
      "Test 879 Prediction: 6 True Class: 5\n",
      "Test 880 Prediction: 5 True Class: 5\n",
      "Test 881 Prediction: 5 True Class: 5\n",
      "Test 882 Prediction: 5 True Class: 5\n",
      "Test 883 Prediction: 5 True Class: 5\n",
      "Test 884 Prediction: 5 True Class: 5\n",
      "Test 885 Prediction: 5 True Class: 5\n",
      "Test 886 Prediction: 5 True Class: 5\n",
      "Test 887 Prediction: 5 True Class: 5\n",
      "Test 888 Prediction: 5 True Class: 5\n",
      "Test 889 Prediction: 5 True Class: 5\n",
      "Test 890 Prediction: 5 True Class: 5\n",
      "Test 891 Prediction: 5 True Class: 5\n",
      "Test 892 Prediction: 5 True Class: 5\n",
      "Test 893 Prediction: 5 True Class: 5\n",
      "Test 894 Prediction: 4 True Class: 5\n",
      "Test 895 Prediction: 5 True Class: 5\n",
      "Test 896 Prediction: 4 True Class: 5\n",
      "Test 897 Prediction: 4 True Class: 5\n",
      "Test 898 Prediction: 5 True Class: 5\n",
      "Test 899 Prediction: 4 True Class: 5\n",
      "Test 900 Prediction: 4 True Class: 5\n",
      "Test 901 Prediction: 5 True Class: 5\n",
      "Test 902 Prediction: 4 True Class: 5\n",
      "Test 903 Prediction: 5 True Class: 5\n",
      "Test 904 Prediction: 4 True Class: 5\n",
      "Test 905 Prediction: 4 True Class: 5\n",
      "Test 906 Prediction: 4 True Class: 5\n",
      "Test 907 Prediction: 4 True Class: 5\n",
      "Test 908 Prediction: 5 True Class: 5\n",
      "Test 909 Prediction: 5 True Class: 5\n",
      "Test 910 Prediction: 4 True Class: 5\n",
      "Test 911 Prediction: 4 True Class: 5\n",
      "Test 912 Prediction: 5 True Class: 5\n",
      "Test 913 Prediction: 5 True Class: 5\n",
      "Test 914 Prediction: 5 True Class: 5\n",
      "Test 915 Prediction: 5 True Class: 5\n",
      "Test 916 Prediction: 5 True Class: 5\n",
      "Test 917 Prediction: 5 True Class: 5\n",
      "Test 918 Prediction: 4 True Class: 5\n",
      "Test 919 Prediction: 5 True Class: 5\n",
      "Test 920 Prediction: 5 True Class: 5\n",
      "Test 921 Prediction: 5 True Class: 5\n",
      "Test 922 Prediction: 5 True Class: 5\n",
      "Test 923 Prediction: 5 True Class: 5\n",
      "Test 924 Prediction: 5 True Class: 5\n",
      "Test 925 Prediction: 5 True Class: 5\n",
      "Test 926 Prediction: 5 True Class: 5\n",
      "Test 927 Prediction: 4 True Class: 5\n",
      "Test 928 Prediction: 5 True Class: 5\n",
      "Test 929 Prediction: 5 True Class: 5\n",
      "Test 930 Prediction: 4 True Class: 5\n",
      "Test 931 Prediction: 6 True Class: 5\n",
      "Test 932 Prediction: 6 True Class: 5\n",
      "Test 933 Prediction: 6 True Class: 6\n",
      "Test 934 Prediction: 6 True Class: 6\n",
      "Test 935 Prediction: 6 True Class: 6\n",
      "Test 936 Prediction: 6 True Class: 6\n",
      "Test 937 Prediction: 7 True Class: 6\n",
      "Test 938 Prediction: 7 True Class: 6\n",
      "Test 939 Prediction: 5 True Class: 6\n",
      "Test 940 Prediction: 6 True Class: 6\n",
      "Test 941 Prediction: 6 True Class: 6\n",
      "Test 942 Prediction: 6 True Class: 6\n",
      "Test 943 Prediction: 6 True Class: 6\n",
      "Test 944 Prediction: 6 True Class: 6\n",
      "Test 945 Prediction: 6 True Class: 6\n",
      "Test 946 Prediction: 6 True Class: 6\n",
      "Test 947 Prediction: 6 True Class: 6\n",
      "Test 948 Prediction: 6 True Class: 6\n",
      "Test 949 Prediction: 6 True Class: 6\n",
      "Test 950 Prediction: 5 True Class: 6\n",
      "Test 951 Prediction: 6 True Class: 6\n",
      "Test 952 Prediction: 5 True Class: 5\n",
      "Test 953 Prediction: 5 True Class: 6\n",
      "Test 954 Prediction: 5 True Class: 5\n",
      "Test 955 Prediction: 5 True Class: 5\n",
      "Test 956 Prediction: 5 True Class: 6\n",
      "Test 957 Prediction: 5 True Class: 6\n",
      "Test 958 Prediction: 5 True Class: 6\n",
      "Test 959 Prediction: 5 True Class: 5\n",
      "Test 960 Prediction: 4 True Class: 5\n",
      "Test 961 Prediction: 4 True Class: 6\n",
      "Test 962 Prediction: 4 True Class: 6\n",
      "Test 963 Prediction: 4 True Class: 6\n",
      "Test 964 Prediction: 5 True Class: 6\n",
      "Test 965 Prediction: 5 True Class: 6\n",
      "Test 966 Prediction: 5 True Class: 6\n",
      "Test 967 Prediction: 4 True Class: 6\n",
      "Test 968 Prediction: 5 True Class: 5\n",
      "Test 969 Prediction: 5 True Class: 6\n",
      "Test 970 Prediction: 4 True Class: 6\n",
      "Test 971 Prediction: 6 True Class: 6\n",
      "Test 972 Prediction: 5 True Class: 6\n",
      "Test 973 Prediction: 5 True Class: 6\n",
      "Test 974 Prediction: 5 True Class: 6\n",
      "Test 975 Prediction: 5 True Class: 6\n",
      "Test 976 Prediction: 5 True Class: 6\n",
      "Test 977 Prediction: 5 True Class: 6\n",
      "Test 978 Prediction: 5 True Class: 6\n",
      "Test 979 Prediction: 5 True Class: 6\n",
      "Test 980 Prediction: 6 True Class: 6\n",
      "Test 981 Prediction: 6 True Class: 6\n",
      "Test 982 Prediction: 6 True Class: 6\n",
      "Test 983 Prediction: 6 True Class: 6\n",
      "Test 984 Prediction: 6 True Class: 6\n",
      "Test 985 Prediction: 6 True Class: 6\n",
      "Test 986 Prediction: 6 True Class: 6\n",
      "Test 987 Prediction: 6 True Class: 6\n",
      "Test 988 Prediction: 6 True Class: 6\n",
      "Test 989 Prediction: 6 True Class: 6\n",
      "Test 990 Prediction: 6 True Class: 6\n",
      "Test 991 Prediction: 4 True Class: 6\n",
      "Test 992 Prediction: 4 True Class: 6\n",
      "Test 993 Prediction: 3 True Class: 2\n",
      "Test 994 Prediction: 4 True Class: 2\n",
      "Test 995 Prediction: 3 True Class: 2\n",
      "Test 996 Prediction: 3 True Class: 2\n",
      "Test 997 Prediction: 3 True Class: 3\n",
      "Test 998 Prediction: 2 True Class: 3\n",
      "Test 999 Prediction: 3 True Class: 2\n",
      "Done!\n",
      "Accuracy: 0.65\n"
     ]
    }
   ],
   "source": [
    "# 定义代价函数和最优化方法\n",
    "cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(pred, ytr))\n",
    "optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)\n",
    "\n",
    "# 评估模型准确率\n",
    "correct_pred = tf.equal(tf.argmax(pred, 1), tf.argmax(ytr, 1))\n",
    "accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))\n",
    "\n",
    "# 定义相似度，即预测值与训练值的距离\n",
    "distance = tf.reduce_sum(tf.abs(tf.add(xtr1, tf.neg(xte1))), reduction_indices=1)  \n",
    "# 预测值: 获取相似度最高的训练标签，作为预测标签\n",
    "pred1 = tf.arg_min(distance, 0)  \n",
    "accuracy1 = 0.  \n",
    "# 变量初始化\n",
    "init = tf.initialize_all_variables()\n",
    "\n",
    "# 创建会话\n",
    "sess = tf.InteractiveSession()\n",
    "sess.run(init)\n",
    "step = 1\n",
    "    # 一直迭代，直到最大步长\n",
    "while step * batch_size < training_iters:\n",
    "    batch_x, batch_y = next_batch_tr(step, batch_size)\n",
    "    batch_x = batch_x.reshape((batch_size, n_steps, n_input))\n",
    "        # 利用最优化方法调整权值\n",
    "    sess.run(optimizer, feed_dict={xtr: batch_x, ytr: batch_y})\n",
    "    if step % display_step == 0:\n",
    "            # 计算准确率和损失函数\n",
    "        acc = sess.run(accuracy, feed_dict={xtr: batch_x, ytr: batch_y})\n",
    "        loss = sess.run(cost, feed_dict={xtr: batch_x, ytr: batch_y})\n",
    "        # 每隔一定步长显示训练准确率和损失率\n",
    "#         print(\"Iter \" + str(step * batch_size) + \", Minibatch Loss= \" +\n",
    "#                 \"{:.6f}\".format(loss) + \", Training Accuracy= \" +\n",
    "#                 \"{:.5f}\".format(acc))\n",
    "    step += 1\n",
    "print(\"Optimization Finished!\")\n",
    "\n",
    "    # 计算n个预测值的预测准确率\n",
    "test_len = 1000\n",
    "test1_data = test1_data[:test_len].reshape((-1, n_steps, n_input))\n",
    "test1_label = test1_label[:test_len]\n",
    "# print(\"Testing Accuracy:\",\n",
    "#         sess.run(accuracy, feed_dict={xtr: test1_data, ytr: test1_label}))\n",
    "\n",
    "for i in range(test_len):  \n",
    "        # 索引即为相似度最大的标签值\n",
    "    nn_index = sess.run(pred1, feed_dict={xtr1: Xtr3, xte1: Xte3[i,:]})  \n",
    "        # 打印出对应的预测标签和实际标签\n",
    "    print \"Test\", i, \"Prediction:\", np.argmax(train1_label[nn_index]), \"True Class:\", np.argmax(test1_label[i])  \n",
    "      \n",
    "    if np.argmax(train1_label[nn_index]) == np.argmax(test1_label[i]):  \n",
    "        accuracy1 += 1./test_len \n",
    "print \"Done!\"  \n",
    "print \"Accuracy:\", accuracy1  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7f7b21350150>"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7b21e05350>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhUAAAGHCAYAAAAHoqCrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzt3Xt8HHW9//HXJ5dtkyxUobQUsBTKRRRBW6GiUKD4gwoI\nykVIAQvFFkS5FI9wRAWkWjwcbBU5IN6gyCF6jqjggSpiQeQm2nARaJFLsUCFUpBI0pY0yef3x8ym\nm3Rz7ezO7Oz7+XjkQXZ2ZufzzYbuO9/5fr9j7o6IiIjI5qqKuwARERFJB4UKERERiYRChYiIiERC\noUJEREQioVAhIiIikVCoEBERkUgoVIiIiEgkFCpEREQkEgoVIiIiEgmFChEREYmEQoWIAGBmO5vZ\ndWb2nJmtM7MWM7vPzM4xs5FFOuceZnaJmY0vxusP4vx14fmnxnF+kbSpibsAEYmfmR0B/A+wHrgR\neALIAPsDVwDvAc4swqnfA1wC3A2sLMLrD6Q+PL8D98ZwfpFUUagQqXBmNgFoAlYA09x9dd7T15rZ\nV4EjinV6gg/0we1sNtLd10d8fhGJiC5/iMiFQANweq9AAYC7P+/u3wUws2oz+6qZPWtm681shZl9\nw8wy+ceY2QtmdpuZfcTM/hReTnnOzE7J22cmQe8IwD1m1mVmnblLEXmvcaiZ/dnM1gFzwudOM7Pf\nm9mrYR1PmtkmPSlm9kEz+62ZvWZma83seTP7UfjcjsBqglBzaXj+LjO7ePN/pCKVyXTrc5HKZmYv\nAuvdfddB7HsD8GmCMHAPMAWYCfzS3Y/N228FwaWUUcCPgFXALOADwPvcfZmZ7QScC5wNfANYHh7+\nO3d/LXyNDcBWwHXAC8DT7n6vmf2J4BLNY0AH8HHgMOBz7n5tWMM24WuuBn4AvAlMAI5x9z3NrB44\nGfge8IvwC+Bxd39ikD8+EcmjUCFSwcxsC6AF+JW7HzPAvnsBjwLfd/cz87ZfAXyB4NLJH8JtK4Dx\nwAHu/kC4bTTwIvBdd78g3HYsQUA52N3v7XW+3Gsc5u539XpuhLu/3WvbYmCXXDgys6MJgsIH3f2R\nPtq0NfAacKm7X9Zf+0VkYLr8IVLZtgz/+9Yg9j2c4FLBwl7bv0UwNqH3uIuncoECwN3XAE8DOw+h\nvhW9A0X4Wt2Bwsy2DMPBvcDOYVCCoGfCgKPMTOPHREpAoUKksv0r/O8W/e4V2BHoAp7N3+jurxJ8\ngO/Ya/9Cszn+CbxzCPWtKLQxHKtxl5m1hud+jeASCgSXXAh7TX4OXAysMbNfmdmpvcd/iEh0FCpE\nKpi7v0Uw3mHPoRw2yP06+9g+lBkX6zY52Gxn4C6CsRZzCXpQPsrGHpTuf9fc/VPAfsB3ge2AHwN/\nCcdTiEjEFCpE5P+AiWY2ZYD9/k7wb0aPAZ1mNgZ4R/j8UA1nUNfHCdbQ+Li7/8Ddf+PuSwgGhm56\nAveH3f2r7r4vcBJBgDpxM84vIn1QqBCRK4C1wA/DgNCDmU00s3OAOwh6Gc7rtcsXCD6cbx/GudvC\n13zHEI7J9YB0//tlZqOAU/N3MrNCr/lY+N8R4X/Xhv8dyvlFpA8avCRS4dz9eTObAfwUWGZm+Stq\nfhg4Hvixu19lZouAOWb2TuAPBFNKPw38IjfzY4geJQgJF4Yh4G3g9+Ggzr7cSTDV9P/M7DqC8SCf\nAV4Fts3bb6aZnQX8Engu3G82wWyXO8K2rzezp4ATzOwZ4A3gCXd/chhtEal4ChUigrv/Opwy+kXg\nKIIlud8GHicYt/DDcNfTCT6gTwU+AbxCMECy93RMp+9LC93b3f1VMzsD+FJ4jmrgYDYumb3Ja7j7\n38KpqF8H/jOs4RrgdYI1MXL+AOwDnACMJQgTfwJmuHv+pZrTCcZcLCAIUl8DFCpEhkHrVIiIiEgk\nEjemwsz+PVwqd8EA+x1vZsvC5X8fM7OPlapGERER2VSiQoWZ7UOwtv9jA+z3YeBmgqV33w/cCvzK\nzN5T9CJFRESkoMSECjPLAjcRDLh6c4DdzwEWu/sCd3/a3S8GmoHPF7lMERER6UNiQgXwX8Cvw/nm\nA9mPYPGbfL8Nt4uIiEgMEjH7w8xOJLiM8cFBHrItwfSxfL2nk4mIiEgJxR4qzGwH4NvAR919QxHP\nszXBrZFfoI+V90RERKSgkcAE4Lfu/npfO8UeKoDJwDZAs5nl7glQDUw1s88DI3zTea+vEMw7zzc2\n3N6Xw4D/jqBeERGRSnUSwUSJgpIQKu4C3tdr2w3AMuCbBQIFwIPAIcBVedv+X7i9Ly8A3DRvHnvs\ntNNway0bc7/1LRZ+4Qtxl1EyldTeSmorqL1pNpS2/mDlL/n5qt/z2w9dXeSqiqdY7+1L61Zz9J/P\n53t7XcQ+7yjOJMhlK1Zw8le/CuFnaV9iDxXu3gY8lb/NzNqA1919Wfh4EfCyu18U7vId4B4zO5/g\nfgONBD0es/s51XqAPQ4/nEmTJkXbiAQa9bOfMemkk+Iuo2Qqqb2V1FZQe9NsKG3d7cHVvH3PnWX9\nsynWe1vz6uPwMkz6xCeYtP0+kb8+AM3NEISKfocPJGn2R77evRPvIm8Qprs/CMwgWNPiUeAY4Gh3\nfwoREUmdhkwDre2taBXoTbW2twLBzyhusfdUFOLu0/p7HG67BbilZEWJiEhsspksjrOuYx31tfVx\nl5MouVCRzWRjriS5PRUiIiLdch+YuQ9Q2UihQoqusbEx7hJKqpLaW0ltBbU3zYbS1twHZlt7W7HK\nKbpivbe5n0kSQkUiL3/EZeXKlaxZsybuMiKx++6709zcHHcZJdNfe0ePHs348eNLXFHxVNKHDqi9\naTacUFHOPRXFem9b21upraolU50pyusPhUJFaOXKleyxxx6sXbs27lIkYvX19SxbtixVwUKk0qQh\nVBRLa3trInopQKGi25o1a1i7di033XQTe+yxR9zlSESWLVvGySefzJo1axQqRMpYQ20ws0GhYlOt\n7a2JmPkBChWb2GOPPSpiHQsRkXKinoq+JamnQgM1RUQk8RQq+qZQISIiMgSZ6gw1VTW0bSjf2R/F\n0rahTaFCRERksMyMbCarnooC1FMhIiIyRA21DQoVBbS2t3YPZI2bQoWUxB/+8Aeqqqq49957u7ed\neuqp7FQBd4wVkWiop6Iw9VRIyS1atIiqqqrur7q6OnbffXfOPvtsVq9eXZIazGyTx1VVQ/8VvPzy\ny7n11lujKktEyoRCRWFJChWaUlpBzIx58+YxYcIE1q9fz3333ce1117L4sWLeeKJJxg5cmRJ6/nh\nD39IV1fXkI+bP38+xx9/PEcffXQRqhKRpMpmshqoWUCSBmoqVFSY6dOnd6/DMWvWLLbaaisWLlzI\nrbfeygknnLDJ/mvXrqW+vjh3BKyurqa6uroory0i6aOeisKS1FOhyx8Vbtq0abg7K1as6L5Ecu+9\n93LWWWcxduxY3vWud3Xvu2rVKmbNmsW2227LyJEj2XPPPbn++us3ec2XX36ZT3ziE2SzWcaOHcv5\n55/P22+/jbv32K/QmAp35zvf+Q577bUXdXV1jBkzho997GPd9/Woqqpi7dq13HDDDd2XcmbNmlWE\nn4yIJE1DRgM1C0nSQE31VFS4Z599FoCtt966e9tZZ53FmDFjuOSSS2hrC7oaV69ezZQpU6iuruac\nc85h9OjRLF68mNNPP5233nqLc845B4D169czbdo0XnrpJc4991zGjRvHT37yE5YsWVJwTEXvbbNm\nzWLRokUcccQRzJ49m46ODv74xz/y0EMPMWnSJG666SZOP/10pkyZwpw5cwCYOHFi0X4+IpIc2dos\nK1tWxl1GonR0dbC+Y31ieioUKipMS0sLr7/+eveYinnz5tHQ0MCRRx7JnXfeCQR39fz973/f4wP/\noosuwt159NFHecc73gHAnDlzmDFjBpdeeilnnHEGI0aM4LrrruPZZ5/lf//3fznmmGMAmD17Nnvt\ntdeAtd19990sWrSI8847jwULFnRvnzt3bvf3M2bM4IwzzmDnnXdmxowZkfxMRKQ86PLHppJ023NQ\nqBi+tWth+fLinuPd74YIxzO4O4ccckj3YzNjwoQJNDU1MW7cuO5ts2fP3qQH4Re/+AUnnHACnZ2d\nvP76693bDz30UH7605/S3NzMfvvtx+LFixk3blx3oAAYOXIkc+bM4cILL+y3vltuuYWqqiouvvji\nKJorIimjULGp3M9DoaLcLV8OkycX9xxLl0KENzczM6655hp23XVXampqGDt2LLvvvvsm+02YMKHH\n49dee40333yT73//+1x33XUFXzc3LfXvf/87u+yyyyb7FDpPb88//zzbbbddd0+IiEi+bCbb/Ze5\nBHKzYRQqyt273x186Bf7HBHbZ599BrwLa11dXY/HuWmfJ598MjNnzix4zGAub4iIbA71VGxKPRVp\nUV8faS9Ckm2zzTZsscUWdHZ2Mm3atH733XHHHXnyySc32b58EJeKJk6cyJ133smbb77Zb29F70sz\nIlIZGjINrOtYR2dXJ9VVmo4OG0NFQyYZsz80pVQGVFVVxbHHHsstt9xSMDCsWbOm+/vDDz+cVatW\nccstt3RvW7t2LT/4wQ8GPM+xxx5LV1cXX/va1/rdr6GhgTfffHMILRCRNMj9Na4FsDZST4XEpvc6\nEUPZ55vf/Cb33HMPU6ZMYfbs2bznPe/hjTfeYOnSpSxZsqQ7WMyePZurr76aU045hb/85S/dU0ob\nGgZO0QcddBCnnHIKV111FX/729+YPn06XV1d/PGPf2TatGmcddZZAEyePJm77rqLhQsXst1227HT\nTjux7777DuEnISLlKPfB2dreypYjtoy5mmRQqJDYDOayQV/7jBkzhocffpjLLruMX/7yl1x77bVs\nvfXWvPe97+WKK67o3q+uro4lS5Zw9tlnc/XVV1NfX8/JJ5/M9OnTmT59+oDnu+GGG9h777350Y9+\nxAUXXMCoUaP44Ac/yIc//OHufRYsWMAZZ5zBV7/6VdatW8fMmTMVKkQqQHdPhQZrdsv9LJKy+JUN\n5q/XNDCzScDSpUuXFhyo2NzczOTJk+nreSlPel9F0qP5H81M/v5kmuc084FxH4i7nET47p++y4V3\nXcjaL68t6nly/5YCk929ua/9NKZCRETKQu6vcc0A2ai1vTUxgzRBoUJERMpE/pgKCSTpZmKgUCEi\nImVCoWJTChUiIiLDkOvmV6jYSKFCRERkGGqqahhZM1LrVORp29CmUCEiIjIcDbUN6qnI09rempjp\npKBQISIiZUT3/+hJlz9ERESGSaGiJ4UKERGRYVKo6EmhQkREZJiymawGaubRQE0REZFhUk9FT+qp\nkIpzww03UFVVxcqVK+MuRUTKXENGsz9y3F2zPyQeixYtoqqqqvurtraWHXbYgdNOO41Vq1YV9dxm\nNqg7pIqIDCRbq56KnPbOdjq6OhLVU6Fbn1cQM2PevHlMmDCB9evX89BDD3H99ddz//3388QTT5DJ\nZOIuUUSkX7r8sVHu56BQIbGZPn169y3AZ82axdZbb80VV1zBbbfdxnHHHRdzdSIi/VOo2CiJoUKX\nPyrcAQccgLvz3HPP9di+ePFipk6dSjabZcstt+TII4/kqaee6rHPX//6V0477TQmTpxIXV0d48aN\n4/TTT+eNN94oZRNEpIJkM1na2jX7A+ieBaNQkcfMzjSzx8ysJfx6wMymD3DMeWa23MzWmtlKM1tg\nZiNKVXOarFixAoB3vvOd3dt+8pOfcOSRR7LFFltwxRVXcPHFF7Ns2TIOOOCAHoMtf/e737FixQpm\nzZrF1VdfTWNjIz/96U854ogjSt4OEakMGqi5Ue7nkLvRWhIk4fLHi8CFwDOAAacCt5rZ+919We+d\nzWwGcHm434PAbsAioAv4t9KUXL5aWlp4/fXXu8dUXHbZZdTV1XHkkUcC0NbWxrnnnsucOXO49tpr\nu4+bOXMmu+22G/Pnz+d73/seAJ/73Oc4//zze7z+lClTmDFjBvfffz8f+chHStcwEakI2UyWDV0b\naO9sJ1Nd2ePAknj5I/ZQ4e6399r0FTP7LPAhYJNQAewH3OfuPwsfrzSzJmDfIpa5ibUb1rJ8zfKi\nnuPdo99NfW19ZK/n7hxyyCE9tu20007cfPPNbLfddkDQ+9DS0sKJJ57I66+/3r2fmTFlyhTuvvvu\n7m0jRmzsHHr77bdpbW1lypQpuDvNzc0KFSISudwHaGt7K1vVbRVzNfFSqBiAmVUBnwLqCXohCnkA\nOMnM9nH3P5vZzsDhBL0VJbN8zXImf39yUc+xdM5SJo2bFNnrmRnXXHMNu+66Ky0tLfz4xz/m3nvv\n7THr45lnnsHdOfjggwseP2rUqO7H//znP7n00kv52c9+xurVq3vs19LSElndIiI5ChUbKVT0wcz2\nJAgRI4G3gE+6e8FuAHdvMrPRwH0WLH5QDXzP3f+jZAUT9CIsnbO06OeI2j777NM9++Poo49m//33\nZ8aMGTz99NPU19fT1dWFmXHTTTcxduzYTY6vqdn4K3P88cfz0EMPccEFF7D33nuTzWbp6urisMMO\no6urK/LaRURyH6AarBn8DAyjrqYu7lK6JSJUAMuBvYFRwHHAjWY2tVCwMLODgIuAM4GHgV2Aq8zs\nH+7+9YFONHfu3B5/bQM0Njay++67D6ng+tr6SHsR4lBVVcXll1/OwQcfzNVXX80FF1zAxIkTcXe2\n2WYbpk2b1uexb775JkuWLGHevHl8+ctf7t7+7LPPlqJ0EalQudUjNVgz+Bk0ZBoiX1ywqamJpqam\nHtsG2/uciFDh7h3A8+HDR8xsX+Bc4LMFdr8MuNHdrw8fP2lmWeA6YMBQsXDhwu6/1PM1NzcPp/Sy\nd+CBB7Lvvvvy7W9/m/POO4/DDjuMLbfckvnz53PQQQf16JkAWLNmDaNHj6a6uhpgkx6JhQsXavVM\nESma/Msfla5Y9/1obGyksbGxx7bm5mYmTx74kn8iQkUBVUBfU0TrCWZ65OsCMDNzdy9mYeWsrx/N\nF7/4RY4//nhuuOGG7lkfn/70p5k0aRInnngi22yzDStXruT2229n//3356qrrmKLLbZg6tSpXHHF\nFbS3t7P99ttz55138sILL/R5HhGRzaVQsVHSbiYGCQgVZjYfWAysBLYATgIOBA4Nn78ReMndLwoP\n+TUw18weBf4E7ErQe3GbAkX/+upBOOaYY5g4cSJXXnkls2fPprGxke23355vfvObXHnllbz99tts\nv/32HHDAAZx22mndxzU1NXH22WdzzTXX4O4cdthhLF68mO222069FSJSFAoVGylUFDaGYObGOKAF\neBw41N2XhM/vAHTk7T+PoGdiHrA98BpwG/CVUhVcjmbOnMnMmTMLPmdmPPPMMz22TZ06lalTp/b7\nmuPGjePnP//5Jts7OzsHfW4RkaGoq63DMIUKoHWDQsUm3P0zAzw/rdfjXKCYV8y6REQkeaqsioZM\nQ/cS1ZWsrb0tcaEi9mW6RUREhqKhVkt1Qzj7ozY5S3SDQoWIiJQZ3ak0kMQxFQoVIiJSVhQqAgoV\nIiIim0mhIqBQISIispmymawGagJtGzRQU0REZLM0ZDRQEzRQU0REZLPp8gd0eVcip5TGvk5F0ixb\ntizuEiRCej9F0idbq1CxbsM6HFeoSKrRo0dTX1/PySefHHcpErH6+npGjx4ddxkiEhH1VGxcplyh\nIqHGjx/PsmXLWLNmTdylSMRGjx7N+PHj4y5DRCKiUKFQURbGjx+vDx8RkYTLZrK0tVf27I/c7Jek\nhQoN1BQRkbKSm/1RyTemzvVUNGQ0+0NERGTYspksjrOuY13cpcQmqZc/FCpERKSs5D5IK3lchUKF\niIhIBBQqFCpEREQikfsgreTBmm3tbdRW1ZKpzsRdSg8KFSIiUlZyS1NXek9F0gZpgkKFiIiUGV3+\nSOYdSkGhQkREyoxChUKFiIhIJBQqFCpEREQikanOUFNVU9mhYoNChYiIyGYzMxpqG7qXqq5Ebe1t\n3QNWk0ShQkREyk6l31RMlz9EREQiolChUCEiIhIJhQqFChERkUgoVChUiIiIRCKbyVb2QM0NbQoV\nIiIiUWjINFR8T4Vmf4iIiEQgW1u5lz86ujpY37FePRUiIiJRqOQxFbm7sypUiIiIRKCSQ0Wu3QoV\nIiIiEVCoUKgQERGJREOmofsyQKXJzXppyGigpoiIyGbLZrKs61hHZ1dn3KWUnHoqREREIpT7QK3E\ntSoUKkRERCKU+0CtxHEVChUiIiIRUqhAi1+JiIhEIfeBWomDNdva2xhZM5Lqquq4S9mEQoWIiJSd\nSu+pSOKlD1CoEBGRMqRQoVBRkJmdaWaPmVlL+PWAmU0f4JhRZvZfZrbKzNab2fKBjhERkfRQqEhm\nqKiJuwDgReBC4BnAgFOBW83s/e6+rPfOZlYL3AW8AhwDrAJ2BN4sVcEiIhKv3MJPChXJEnuocPfb\ne236ipl9FvgQsEmoAE4H3gF8yN1zq56sLGKJIiKSMDVVNYysGVmZoWJDckNF7Jc/8plZlZmdCNQD\nD/ax28fD564xs1fM7K9m9iUzS1RbRESkuBpqGypy8au29rZETieFBPRUAJjZngRBYSTwFvBJd1/e\nx+47A9OAm4CPAbsA1xK0ZV7xqxURkSSo1JuKtba3ssOWO8RdRkGJCBXAcmBvYBRwHHCjmU3tI1hU\nAa8Cc9zdgUfMbAfg31CoEBGpGJUcKpJ6+SMRocLdO4Dnw4ePmNm+wLnAZwvs/g+gPQwUOcuAbc2s\nJnytPs2dO5dRo0b12NbY2EhjY+Ow6xcRkdJTqCiOpqYmmpqaemxraWkZ1LGJCBUFVAEj+njufqB3\nAtgd+MdAgQJg4cKFTJo0aTPLExGRuClUFEehP7Sbm5uZPHnygMfGPrjRzOab2QFmtqOZ7WlmlwMH\nEoyZwMxuNLP5eYdcC2xlZleZ2a5mdgTwJeDq0lcvIiJxachU6EDNDRqo2Z8xwCJgHNACPA4c6u5L\nwud3ALp7INz9JTM7DFgIPAa8HH5/RSmLFhGReGUzWVa9tSruMkrK3TWmoj/u/pkBnp9WYNufgA8X\nrSgREUm8bG3lXf5o72yno6sjsaEi9ssfIiIiw1GJYypy7VWoEBERiZBCRfIoVIiISFlqyDRUbKjI\n3fskaRQqRESkLGUzWdraK2v2R262i3oqREREIpTNZNnQtYH2zva4SykZXf4QEREpgtwHayVdAlGo\nEBERKQKFiuRRqBARkbJUqaHCMOpq6uIupSCFChERKUu5paorabBmW3sbDZkGzCzuUgpSqBARkbJU\nqT0VSb30AQoVIiJSphQqkkehQkREypJCRfIoVIiISFmqq63DsMoKFRsUKkRERCJXZVXU19ZXVqho\nb+0eoJpEChUiIlK2spls99LVlaCtvU09FSIiIsVQaXcq1ZgKERGRIlGoSBaFChERKVsKFcmiUCEi\nImVLoSJZFCpERKRsNWQaKmug5oY2zf4QEREphkrqqejyLs3+EBERKZZsbeWEinUb1uG4QoWIiEgx\nVFJPRa6dChUiIiJFoFCRLAoVIiJSthoyDRUXKhoyGqgpIiISuWwmS1t7G+4edylFl5vlop4KERGR\nIshmsjjOuo51cZdSdLr8ISIiUkS5D9hKuASiUCEiIlJEChXJolAhIiJlK7e6ZKWEipqqGjLVmbhL\n6ZNChYiIlK3cX+1t7elfqjvpq2nCMEOFmU03s/3zHn/OzB41s5vN7J3RlSciItK3Srv8kcpQAfwn\nsCWAmb0P+BZwB7ATsCCa0kRERPqnUJEsNcM8bifgqfD7Y4H/c/eLzGwSQbgQEREpOoWKZBluT0U7\nUB9+/1HgzvD7Nwh7MERERIotU52hpqqmMkLFhuSHiuH2VNwHLDCz+4F9gRPC7bsBL0VRmIiIyEDM\njIbayliqu7W9tXu2S1INt6fi80AHcBzwWXd/Odz+MeA3URQmIiIyGNlMtnsJ6zQrh9kfw+qpcPeV\nwJEFts/d7IpERESGoFLuVNra3sqEd0yIu4x+DXdK6aRw1kfu8dFm9iszm29myV2VQ0REUqeSQkXS\neyqGe/njOoLxE5jZzsBPgbXA8cAV0ZQmIiIyMIWK5BhuqNgNeDT8/njgXnefAZxKMMVURESkJBoy\nGqiZFMMNFZZ37EfZuDbFi8DoIb2Q2Zlm9piZtYRfD5jZ9EEee6KZdZnZL4ZyThERSY+KGai5IfkD\nNYcbKv4CfMXMTgEOBG4Pt+8EvDrE13oRuBCYBEwGlgC3mtke/R1kZhMIVva8d4jnExGRFMnWpv/y\nR0dXB+s71qc2VJxHEAKuBr7h7s+G248DHhjKC7n77e7+G3d/zt2fdfevAK3Ah/o6xsyqgJuAi4EV\nw2mAiIikQyWMqcjdMC3poWK4U0ofB95X4KkvAp3DLSYMC58iWK3zwX52vQR41d2vN7Opwz2fiIiU\nv0oIFbn2pTJU9MXd1w/nODPbkyBEjATeAj7p7sv72Hd/4DRg7+HWKSIi6VEJAzVz7WvIJHug5rBC\nhZlVA3MJehXGAz3WpnD3rYb4kssJQsIogksoN5rZ1N7BwsyywI3AbHf/53Bqnzt3LqNGjeqxrbGx\nkcbGxuG8nIiIxEw9FdFqamqiqampx7aWlpZBHTvcnopLgM8Q3PL868A3gAnAJ4DLhvpi7t4BPB8+\nfMTM9gXOBT7ba9eJwI7Ar83Mwm1VAGbWDuzu7v2OsVi4cCGTJk0aaokiIpJQ2UyW9R3r6ezqpLqq\nOu5yiiI3u6UUoaLQH9rNzc1Mnjx5wGOHO1DzJILegm8R3AOkyd0/QxAo+hxgOQRVwIgC25cRjOV4\nP0HPxt7AbQQzRvYmmEkiIiIVJPdBm+ZppWkfU7Et8Nfw+1aCyxYA/wfMG8oLmdl8YDGwEtiCILAc\nCBwaPn8j8JK7X+Tu7cBTvY5/E3B3Xza8poiISDnLfdC2trey5YgtY66mONIeKl4CxhEEgecIAkAz\nsA/w9hBfawywKHy9FuBx4FB3XxI+vwNBb4iIiMgm8kNFWnUP1Ez4iprDDRW/BA4B/gR8F7jJzE4n\nGLS5cCjvH45dAAAVNElEQVQvFF426e/5aQM8f9pQziciIumS+6BNe6gYWTMy8WNGhrtOxb/nff8z\nM1sJ7Ac84+6/jqo4ERGRgXSPqWhP75iKtvbkL9ENEa1T4e4P0v9iVSIiIkVRKZc/UhUqzOyowe7r\n7rcNrxwREZGhUahIjqH0VPxqkPs5kOyLPiIikhq5VSYVKuI36FDh7sNd00JERKRoaqpqGFE9It2h\nYkNr4md+wBAXvzKzaWb2lJltMhHYzEaZ2ZNmdkB05YmIiAws7Ut1l0tPxVB7H84DfuDu/+r9hLu3\nANcB50dRmIiIyGBlM9lUr6hZLrM/hhoq9gZ+08/zdwIDLw4uIiISIfVUJMNQQ8VYYEM/z3cA2wy/\nHBERkaFTqEiGoYaKl4E9+3l+L+Afwy9HRERk6BoyDakPFakbqAncAcwzs5G9nzCzOuBrBDcVExER\nKRn1VCTDUFfU/DpwDPA3M7saeDrc/m7gcwTrU3wjuvJEREQGls1kWfXWqrjLKAp3p21DeQzUHFKo\ncPdXzezDwLXA5YDlngJ+C3zO3V+NtkQREZH+ZWvT21PR3tlOR1dH+kIFgLv/HTjczN4J7EIQLJ5x\n939GXZyIiMhgpPnyR65dqQwVOWGI+HOEtYiIiAyLQkUyaOltEREpe2me/ZFrV+4eJ0mmUCEiImVP\nPRXJoFAhIiJlL5vJ0tHVQXtne9ylRC63/LhChYiISAnkPnDT2FuhngoREZESUqhIBoUKEREpe7kl\nrNMaKgyjrqYu7lIGpFAhIiJlL+09FQ2ZBsxs4J1jplAhIiJlLxcq2trbYq4kem3t5bFENyhUiIhI\nCqS9p0KhQkREpEQUKpJBoUJERMpeXW0wiDGVoWJDa/dA1KRTqBARkbJXZVU01KZzqW71VIiIiJRY\nWpfqVqgQEREpsWwm272kdZpo9oeIiEiJqacifgoVIiKSCgoV8VOoEBGRVGjIpHegpmZ/iIiIlJB6\nKuKnUCEiIqmQxoGaXd7F2g1rFSpERERKKVubvp6KdRvW4bhChYiISCml8fJHrj0KFSIiIiWUxoGa\nufY0ZDRQU0REpGTUUxE/hQoREUmFbCZLW3sb7h53KZFRqBAREYlBNpPFcdZ1rIu7lMjkZrMoVIiI\niJRQ7oM3TZdA1FMxRGZ2ppk9ZmYt4dcDZja9n/0/Y2b3mtkb4dfvzGyfUtYsIiLJo1ARv9hDBfAi\ncCEwCZgMLAFuNbM9+tj/QOBm4CDgQ+Hxd5rZuOKXKiIiSZVbyjptoaKmqoZMdSbuUgalJu4C3P32\nXpu+YmafJQgMywrsf0r+YzP7DHAscAhwU7HqFBGRZEtrT0W59FJAAkJFPjOrAj4F1AMPDvKwBqAW\neKNYdYmISPLlPnzb2tOzVHdbe5tCxVCZ2Z4EIWIk8BbwSXdfPsjD/wN4GbirSOWJiEgZUE9F/BIR\nKoDlwN7AKOA44EYzmzpQsDCzfyfo2TjQ3dsHc6K5c+cyatSoHtsaGxtpbGwcVuEiIpIMChXRaGpq\noqmpqce2lpaWQR2biFDh7h3A8+HDR8xsX+Bc4LN9HWNm/wZcABzi7k8O9lwLFy5k0qRJm1OuiIgk\nUKY6Q7VVpytUbGjtHoBaKoX+0G5ubmby5MkDHpuIUFFAFTCiryfN7ALgS8Ch7v5IyaoSEZHEMrPU\nLdWtyx9DZGbzgcXASmAL4CSCaaOHhs/fCLzk7heFjy8EvgY0AivNbGz4Uq3unp7ROSIiMmRpDBXb\n1G8TdxmDloR1KsYAiwjGVdxFsFbFoe6+JHx+B2DbvP3PJJjt8XNgVd7XF0pVsIiIJFM2k+1e2joN\nNPtjiNz9MwM8P63X452KW5GIiJSrNPZUlFOoSEJPhYiISCQaMg2pCxWlHqi5ORQqREQkNdRTES+F\nChERSQ2FingpVIiISGpka9MzULOjq4O3O99WqBAREYlDmnoqcvcwUagQERGJQZpCRa4dChUiIiIx\nSNPsj1w7GjKa/SEiIlJy6qmIl0KFiIikRjaTZX3Hejq7OuMuZbMpVIiIiMQo9wGchhkguTYoVIiI\niMQg9wGchksg6qkQERGJUW5J6zSFCi3TLSIiEoO09VSMrBlJdVV13KUMmkKFiIikRtpCRTld+gCF\nChERSZHugZrtKRio2d6mUCEiIhIX9VTES6FCRERSI7f6ZFpCRTkN0gSFChERSZGaqhpGVI9IR6jY\noJ4KERGRWKVlqW5d/hAREYmZQkV8FCpERCRVsplsOpbp1uwPERGReKmnIj4KFSIikioNmYbUhArN\n/hAREYmReirio1AhIiKpkoZQ4e4KFSIiInHL1pb/QM32znY6vVOhQkREJE5p6KnI1a9QISIiEqM0\nDNTM1Z9bdrxcKFSIiEiqqKciPgoVIiKSKrlQ4e5xlzJsChUiIiIJkM1k6ejqoL2zPe5Shk2hQkRE\nJAFyH8TlPAMkV7tChYiISIxyq1CW87iK7oGaWlFTREQkPrm/7ss9VBhGXW1d3KUMiUKFiIikSlpC\nRUOmgSorr4/p8qpWRERkAGkJFeU2ngIUKkREJGW6B2q2l/FAzfY2hQoREZG4qaciPgoVIiKSKrnB\njWUdKja0lt3MD1CoEBGRlKmyKhpqy/v+H+qpGCYzO9PMHjOzlvDrATObPsAxx5vZMjNbFx77sVLV\nKyIiyVfu9/9QqBi+F4ELgUnAZGAJcKuZ7VFoZzP7MHAz8APg/cCtwK/M7D2lKVdERJJOoSIesYcK\nd7/d3X/j7s+5+7Pu/hWgFfhQH4ecAyx29wXu/rS7Xww0A58vVc0iIpJs2Uy2vJfp1uyPzWdmVWZ2\nIlAPPNjHbvsBd/Xa9ttwu4iICA2Z8h9TUY4DNWviLgDAzPYkCBEjgbeAT7r78j523xZ4tde2V8Pt\nIiIiuvwRk0SECmA5sDcwCjgOuNHMpvYTLERERPq0RWYL7njmDna5ape4SxmW1W2rFSqGy907gOfD\nh4+Y2b7AucBnC+z+CjC217ax4fYBzZ07l1GjRvXY1tjYSGNj45BqFhGR5Prih7/IrlvtGncZw1Zd\nVc2n3vupWM7d1NREU1NTj20tLS2DOtbcvRg1bRYz+z3wd3efVeC5nwJ17n503rb7gcfc/ax+XnMS\nsHTp0qVMmjSpGGWLiIikUnNzM5MnTwaY7O7Nfe0Xe0+Fmc0HFgMrgS2Ak4ADgUPD528EXnL3i8JD\nvgPcY2bnA7cDjQRTUWeXuHQRERHJE3uoAMYAi4BxQAvwOHCouy8Jn98B6Mjt7O4PmtkM4Bvh1zPA\n0e7+VEmrFhERkR5iDxXu/pkBnp9WYNstwC1FK0pERESGLFHrVIiIiEj5UqgQERGRSChUiIiISCQU\nKkRERCQSChUiIiISCYUKERERiYRChYiIiERCoUJEREQioVAhIiIikVCoEBERkUgoVIiIiEgkFCpE\nREQkEgoVIiIiEgmFChEREYmEQoWIiIhEQqFCREREIqFQISIiIpFQqBAREZFIKFSIiIhIJBQqRERE\nJBIKFSIiIhIJhQoRERGJhEKFiIiIREKhQkRERCKhUCEiIiKRUKgQERGRSChUiIiISCQUKkRERCQS\nChUiIiISCYUKERERiYRChYiIiERCoUJEREQioVAhIiIikVCoEBERkUgoVIiIiEgkFCpEREQkEgoV\nIiIiEgmFChEREYmEQoWIiIhEQqFCREREIqFQISIiIpGIPVSY2ZfM7GEz+5eZvWpmvzSz3QZx3Hlm\nttzM1prZSjNbYGYjSlFzOWhqaoq7hJKqpPZWUltB7U2zSmorVEZ7Yw8VwAHAd4EpwEeBWuBOM6vr\n6wAzmwFcDlwCvBuYBZwAfKPo1ZaJSvjlzVdJ7a2ktoLam2aV1FaojPbWxF2Aux+e/9jMTgVWA5OB\n+/o4bD/gPnf/Wfh4pZk1AfsWq04RERHpXxJ6Knp7B+DAG/3s8wAw2cz2ATCznYHDgduLX56IiIgU\nEntPRT4zM+DbBL0QT/W1n7s3mdlo4L7wmGrge+7+HyUqVURERHpJVKgArgHeA3ykv53M7CDgIuBM\n4GFgF+AqM/uHu3+9j8NGAixbtiyyYpOspaWF5ubmuMsomUpqbyW1FdTeNKuktkJ5tzfvs3Nkf/uZ\nuxe/mkEws6uBjwMHuPvKAfa9F3jQ3S/M23YScJ27Z/s4Zgbw3xGWLCIiUmlOcveb+3oyET0VYaA4\nGjhwoEARqge6em3rCl/LvHBS+i1wEvACsH741YqIiFSckcAEgs/SPsXeU2Fm1wCNwFHA3/KeanH3\n9eE+i4CX3f2i8PElwFzgDOBPwK4El07+7O4zSli+iIiIhJIQKroIZnv0dpq73xjuswR4wd1nhY+r\ngC8DpwDbA68BtwFfcfd/laRwERER6SH2UCEiIiLpkMR1KkRERKQMKVSIiIhIJCoiVJjZ58xshZmt\nM7OHcitxpo2ZXWJmXb2++lxErJyY2QFmdpuZvRy266gC+1xmZqvCm8z9zsx2iaPWKAzUXjO7vsB7\nfUdc9W6OwdxU0MxGmNl/mdkaM3vLzH5uZmPiqnlzDLK99/R6bzvDQe1lx8zONLPHzKwl/HrAzKbn\nPZ+m93agtqbmfe1L6kOFmZ0AfIvg5mMfAB4DfhuuyJlGTwBjgW3Dr/3jLScyDcCjwFkUGNhrZhcC\nnwfmENwDpo3gfc6UssgI9dve0GJ6vteNpSktcoO5qeC3gSOAY4GpwHbALSWuMyqDaa8D32fj+zsO\nuKDEdUblReBCYBLBPZ2WALea2R7h82l6bwdqa5re18LcPdVfwEPAd/IeG/AScEHctRWhrZcAzXHX\nUYJ2dgFH9dq2Cpib93hLYB3wqbjrLVJ7rwd+EXdtRWrv6LDN++e9l28Dn8zbZ/dwn33jrjfq9obb\n7gYWxF1bEdv8OnBa2t/b/LZWwvvq7unuqTCzWoK0+PvcNg/e2bsI7nSaRruGXebPmdlNZvauuAsq\nNjPbiSD157/P/yJYwySt7zPAQWH3+XIzu8bMtoq7oIj0vqngZIKF+vLf36eBlaTj/e3rJoonmdlr\nZvZXM5vfqyejLJlZlZmdSLCA4YOk+L3t1dYH8p5K3fuaLxErahbRaIKbjb3aa/urBGk4bR4CTgWe\nJuhWuxS418z2dPe2GOsqtm0J/lEu9D5vW/pySmIxQRfxCmAicDlwh5ntFwbnsmRW8KaC2wLtvuka\nNGX//vbRXghuKfB3gh64vYArgN2A40peZATMbE+CEDESeIugZ2K5mX2AlL23fbT16fDpVL2vhaQ9\nVFQUd89fPvUJM3uY4Bf4UwTd5ZIS7v4/eQ+fNLO/As8BBxF0sZar3E0F0zIWaCAFb6Lo7j/Me/ik\nmb0C3GVmO7n7ilIWGJHlwN7AKIIP0BvNbGq8JRVNwba6+/IUvq+bSPXlD2AN0EkwKCbfWOCV0pdT\nWu7eQrD0ednOghikVwjGylTk+wwQ/oO0hjJ+ry24B9DhwEHuvirvqVeAjJlt2euQsn5/e7X3HwPs\n/ieC3/GyfH/dvcPdn3f3R9z9ywQD5s8lhe9tP20tpKzf10JSHSrcfQOwFDgkty3sbjyEnte4UsnM\nsgRd4wP9g1XWwg/UV+j5Pm9JMLo+9e8zgJntAGxNmb7XtvGmggf7pjcVXAp00PP93R0YT9DNXHYG\naG8hHyC4xFeW728BVcAIUvjeFpBrayFpe18r4vLHAuAGM1sKPExwI7J64IY4iyoGM/tP4NcElzy2\nB75G8D9sU5x1RcHMGgjSvIWbdjazvYE33P1FguvSXzGzZwnuRDuPYJbPrTGUu9n6a2/4dQnBmIpX\nwv3+g6BXqt87CCaR9bypYJuZ5XqcWtx9vbv/y8x+BCwws38SXKe+Crjf3R+Op+rhG6i9ZrYzMAO4\ng2DmwN4E/479wd2fiKPmzWFm8wnGAK0EtiC4W/SBwKEpfG/7bGva3tc+xT39pBRfBHP9XyCYYvgg\n8MG4aypSO5sIPkjXEfxS3wzsFHddEbXtQIJpZp29vn6ct8+lBAOg1hJ8uO4Sd93FaC/BALDfEASK\n9cDzwLXANnHXPcy2FmpnJ/DpvH1GEKztsIbgg+d/gTFx116M9gI7APcQ3ChxLcHA68uBbNy1D7O9\nPwx/R9eFv7N3AtNS+t722da0va99femGYiIiIhKJVI+pEBERkdJRqBAREZFIKFSIiIhIJBQqRERE\nJBIKFSIiIhIJhQoRERGJhEKFiIiIREKhQkRERCKhUCEiIiKRUKgQkX6Z2YFm1lngTpKJVG71iqSJ\nQoWI9GBmd5vZgrxN9wPj3P1fJTr/gWY2mDt39qWk9YrIRpVwl1IR2Qzu3gGsLuEpjwJuG+7BMdQr\nIiH1VIhINzO7nuAOqeeaWVd4GWFm+P2W4T4zzeyfZnaEmS03szYz+x8zqwufW2Fmb5jZd8zM8l47\nY2ZXmtlLZtZqZg+a2YEFyjiK8Jb1Ya/JVWa2MHzNV8zsdDOrN7Mfm9m/zOwZM5ued54D+6j3UDN7\nyszeMrPFebccF5GIKFSISL5zgQeBHwBjgXHAi0Dv2xnXA2cDnwIOAw4GfglMBz4GnAycARyXd8x/\nAVPCY95HcIvrxWY2MbeDmb0X2Aa4O++4TxPcLnof4Crge+Gx9wMfILi99I1mNjLvmEL1fgE4CTgA\nGA9cOfCPQ0SGQqFCRLqF4xDagbXu/pq7rwY6C+xaA5zp7o+7+33Az4GPALPcfbm730EQDA4GMLPx\nwKnA8e7+gLuvcPcFBMHgtLzXPQr4bXgJI+cxd5/v7s8B3wTWA6+5+4/CbZcBo4G9+mlaDXCGuz/i\n7o8CVwOHDOVnIyID05gKERmOte7+Qt7jV4EX3H1dr21jwu/3BKqBv+VfEgEywJq8x0cD3+11rsdz\n37h7l5m9Dvw1b9ur4UuOoW+96/3HAPuLyDAoVIjIcGzo9dj72JbrDc0CHcAkoKvXfq0AZrYt8H7g\n9mGcC/rveS30GlZoRxEZPoUKEemtnaBXIUqPhK851t3v72OfjwMPuPubEZ9bREpEoUJEensBmGJm\nOxL0IlSxmX/Vu/szZnYzwYDKfyMIGWOAaQRjJhazmVNJe1EvhEgMNFBTRHq7kmBw5lME6z2MZ9PZ\nFMNxKnBj+PrLgV8AHwRWmlk9wcDJ3qGi0HkHsy2KekVkiMxd/++JSLzM7JPAPHffM+5aRGT41FMh\nIknwFnBh3EWIyOZRT4WIiIhEQj0VIiIiEgmFChEREYmEQoWIiIhEQqFCREREIqFQISIiIpFQqBAR\nEZFIKFSIiIhIJBQqREREJBIKFSIiIhKJ/w9rn7QUSS64XAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7b3d0ffd10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#!/usr/bin/env python\n",
    "# -*- coding: utf-8 -*-  \n",
    "# 画出对比图，绿色表示实际值，红色表示预测值\n",
    "import numpy as np    \n",
    "import matplotlib.mlab as mlab    \n",
    "import matplotlib.pyplot as plt    \n",
    "# import ch\n",
    "%matplotlib inline\n",
    "\n",
    "# from pylab import *\n",
    "\n",
    "# mpl.rcParams['font.sans-serif'] = ['SimHei']\n",
    "# mpl.rcParams['axes.unicode_minus'] = False\n",
    "Y1=[]\n",
    "for i in range(60,100):\n",
    "    nn_index = sess.run(pred1, feed_dict={xtr1: Xtr3, xte1: Xte3[i,:]})  \n",
    "    Y1.append(np.argmax(train1_label[nn_index]))\n",
    "Y2=[]\n",
    "for i in range(60,100):\n",
    "    Y2.append(np.argmax(test1_label[i]))\n",
    "\n",
    "Y3 = pd.concat([Series(Y1),Series(Y2)],axis=1)\n",
    "# set_ch()\n",
    "Y3.columns = ['Predict','Real']\n",
    "X=range(60,100) \n",
    "fig = plt.figure()  \n",
    "Y3.plot(color=['red','green'],kind='line',use_index=False)  \n",
    "#plt.bar(X,Y2,0.4,color=\"red\")  \n",
    "plt.xlabel(\"time/min\")  \n",
    "plt.ylabel(\"Class\")  \n",
    "plt.title(\"Contrast\")  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}