Shapelet-analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset_name = \"Chinatown\"\n",
    "coef = 0.1\n",
    "model = \"dt\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>att1</th>\n",
       "      <th>att2</th>\n",
       "      <th>att3</th>\n",
       "      <th>att4</th>\n",
       "      <th>att5</th>\n",
       "      <th>att6</th>\n",
       "      <th>att7</th>\n",
       "      <th>att8</th>\n",
       "      <th>att9</th>\n",
       "      <th>att10</th>\n",
       "      <th>...</th>\n",
       "      <th>att16</th>\n",
       "      <th>att17</th>\n",
       "      <th>att18</th>\n",
       "      <th>att19</th>\n",
       "      <th>att20</th>\n",
       "      <th>att21</th>\n",
       "      <th>att22</th>\n",
       "      <th>att23</th>\n",
       "      <th>att24</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>573</td>\n",
       "      <td>375</td>\n",
       "      <td>301</td>\n",
       "      <td>212</td>\n",
       "      <td>55</td>\n",
       "      <td>34</td>\n",
       "      <td>25</td>\n",
       "      <td>33</td>\n",
       "      <td>113</td>\n",
       "      <td>143</td>\n",
       "      <td>...</td>\n",
       "      <td>1081</td>\n",
       "      <td>866</td>\n",
       "      <td>1096</td>\n",
       "      <td>1039</td>\n",
       "      <td>975</td>\n",
       "      <td>746</td>\n",
       "      <td>581</td>\n",
       "      <td>409</td>\n",
       "      <td>182</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>394</td>\n",
       "      <td>264</td>\n",
       "      <td>140</td>\n",
       "      <td>144</td>\n",
       "      <td>104</td>\n",
       "      <td>28</td>\n",
       "      <td>28</td>\n",
       "      <td>25</td>\n",
       "      <td>70</td>\n",
       "      <td>153</td>\n",
       "      <td>...</td>\n",
       "      <td>1240</td>\n",
       "      <td>1109</td>\n",
       "      <td>1137</td>\n",
       "      <td>1290</td>\n",
       "      <td>1137</td>\n",
       "      <td>791</td>\n",
       "      <td>638</td>\n",
       "      <td>597</td>\n",
       "      <td>316</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>603</td>\n",
       "      <td>348</td>\n",
       "      <td>176</td>\n",
       "      <td>177</td>\n",
       "      <td>47</td>\n",
       "      <td>30</td>\n",
       "      <td>40</td>\n",
       "      <td>42</td>\n",
       "      <td>101</td>\n",
       "      <td>180</td>\n",
       "      <td>...</td>\n",
       "      <td>1243</td>\n",
       "      <td>1141</td>\n",
       "      <td>1178</td>\n",
       "      <td>1256</td>\n",
       "      <td>1114</td>\n",
       "      <td>814</td>\n",
       "      <td>635</td>\n",
       "      <td>304</td>\n",
       "      <td>168</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>428</td>\n",
       "      <td>309</td>\n",
       "      <td>199</td>\n",
       "      <td>117</td>\n",
       "      <td>82</td>\n",
       "      <td>43</td>\n",
       "      <td>24</td>\n",
       "      <td>64</td>\n",
       "      <td>152</td>\n",
       "      <td>183</td>\n",
       "      <td>...</td>\n",
       "      <td>1460</td>\n",
       "      <td>1365</td>\n",
       "      <td>1520</td>\n",
       "      <td>1700</td>\n",
       "      <td>1797</td>\n",
       "      <td>1596</td>\n",
       "      <td>1139</td>\n",
       "      <td>910</td>\n",
       "      <td>640</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>372</td>\n",
       "      <td>310</td>\n",
       "      <td>203</td>\n",
       "      <td>133</td>\n",
       "      <td>65</td>\n",
       "      <td>39</td>\n",
       "      <td>27</td>\n",
       "      <td>36</td>\n",
       "      <td>107</td>\n",
       "      <td>139</td>\n",
       "      <td>...</td>\n",
       "      <td>971</td>\n",
       "      <td>1104</td>\n",
       "      <td>844</td>\n",
       "      <td>1023</td>\n",
       "      <td>1019</td>\n",
       "      <td>862</td>\n",
       "      <td>643</td>\n",
       "      <td>591</td>\n",
       "      <td>452</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>448</td>\n",
       "      <td>344</td>\n",
       "      <td>183</td>\n",
       "      <td>146</td>\n",
       "      <td>71</td>\n",
       "      <td>14</td>\n",
       "      <td>30</td>\n",
       "      <td>41</td>\n",
       "      <td>108</td>\n",
       "      <td>137</td>\n",
       "      <td>...</td>\n",
       "      <td>1062</td>\n",
       "      <td>1093</td>\n",
       "      <td>1030</td>\n",
       "      <td>1069</td>\n",
       "      <td>1151</td>\n",
       "      <td>898</td>\n",
       "      <td>754</td>\n",
       "      <td>467</td>\n",
       "      <td>362</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>621</td>\n",
       "      <td>322</td>\n",
       "      <td>221</td>\n",
       "      <td>150</td>\n",
       "      <td>65</td>\n",
       "      <td>40</td>\n",
       "      <td>42</td>\n",
       "      <td>84</td>\n",
       "      <td>148</td>\n",
       "      <td>190</td>\n",
       "      <td>...</td>\n",
       "      <td>1279</td>\n",
       "      <td>1318</td>\n",
       "      <td>1336</td>\n",
       "      <td>1440</td>\n",
       "      <td>1479</td>\n",
       "      <td>1417</td>\n",
       "      <td>1347</td>\n",
       "      <td>1003</td>\n",
       "      <td>803</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>597</td>\n",
       "      <td>409</td>\n",
       "      <td>142</td>\n",
       "      <td>93</td>\n",
       "      <td>48</td>\n",
       "      <td>30</td>\n",
       "      <td>34</td>\n",
       "      <td>87</td>\n",
       "      <td>132</td>\n",
       "      <td>157</td>\n",
       "      <td>...</td>\n",
       "      <td>1706</td>\n",
       "      <td>1520</td>\n",
       "      <td>1562</td>\n",
       "      <td>1608</td>\n",
       "      <td>1766</td>\n",
       "      <td>1533</td>\n",
       "      <td>1441</td>\n",
       "      <td>1252</td>\n",
       "      <td>964</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>525</td>\n",
       "      <td>431</td>\n",
       "      <td>248</td>\n",
       "      <td>240</td>\n",
       "      <td>91</td>\n",
       "      <td>64</td>\n",
       "      <td>29</td>\n",
       "      <td>117</td>\n",
       "      <td>200</td>\n",
       "      <td>236</td>\n",
       "      <td>...</td>\n",
       "      <td>1398</td>\n",
       "      <td>1465</td>\n",
       "      <td>1459</td>\n",
       "      <td>1631</td>\n",
       "      <td>1891</td>\n",
       "      <td>1847</td>\n",
       "      <td>1731</td>\n",
       "      <td>1375</td>\n",
       "      <td>1188</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>587</td>\n",
       "      <td>382</td>\n",
       "      <td>165</td>\n",
       "      <td>192</td>\n",
       "      <td>130</td>\n",
       "      <td>44</td>\n",
       "      <td>21</td>\n",
       "      <td>35</td>\n",
       "      <td>73</td>\n",
       "      <td>132</td>\n",
       "      <td>...</td>\n",
       "      <td>1069</td>\n",
       "      <td>1070</td>\n",
       "      <td>1164</td>\n",
       "      <td>1547</td>\n",
       "      <td>1575</td>\n",
       "      <td>1503</td>\n",
       "      <td>1139</td>\n",
       "      <td>1066</td>\n",
       "      <td>776</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>144</td>\n",
       "      <td>73</td>\n",
       "      <td>21</td>\n",
       "      <td>16</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>26</td>\n",
       "      <td>100</td>\n",
       "      <td>177</td>\n",
       "      <td>220</td>\n",
       "      <td>...</td>\n",
       "      <td>1367</td>\n",
       "      <td>1319</td>\n",
       "      <td>1205</td>\n",
       "      <td>1250</td>\n",
       "      <td>1269</td>\n",
       "      <td>981</td>\n",
       "      <td>806</td>\n",
       "      <td>550</td>\n",
       "      <td>320</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>141</td>\n",
       "      <td>63</td>\n",
       "      <td>51</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "      <td>14</td>\n",
       "      <td>28</td>\n",
       "      <td>103</td>\n",
       "      <td>172</td>\n",
       "      <td>205</td>\n",
       "      <td>...</td>\n",
       "      <td>872</td>\n",
       "      <td>847</td>\n",
       "      <td>1227</td>\n",
       "      <td>1561</td>\n",
       "      <td>1611</td>\n",
       "      <td>1386</td>\n",
       "      <td>1007</td>\n",
       "      <td>810</td>\n",
       "      <td>680</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>67</td>\n",
       "      <td>67</td>\n",
       "      <td>107</td>\n",
       "      <td>14</td>\n",
       "      <td>11</td>\n",
       "      <td>19</td>\n",
       "      <td>18</td>\n",
       "      <td>75</td>\n",
       "      <td>185</td>\n",
       "      <td>223</td>\n",
       "      <td>...</td>\n",
       "      <td>765</td>\n",
       "      <td>824</td>\n",
       "      <td>1032</td>\n",
       "      <td>1237</td>\n",
       "      <td>1084</td>\n",
       "      <td>759</td>\n",
       "      <td>543</td>\n",
       "      <td>377</td>\n",
       "      <td>168</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>69</td>\n",
       "      <td>54</td>\n",
       "      <td>35</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>19</td>\n",
       "      <td>76</td>\n",
       "      <td>212</td>\n",
       "      <td>238</td>\n",
       "      <td>...</td>\n",
       "      <td>931</td>\n",
       "      <td>929</td>\n",
       "      <td>1093</td>\n",
       "      <td>1275</td>\n",
       "      <td>1161</td>\n",
       "      <td>1083</td>\n",
       "      <td>726</td>\n",
       "      <td>441</td>\n",
       "      <td>241</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>142</td>\n",
       "      <td>104</td>\n",
       "      <td>55</td>\n",
       "      <td>33</td>\n",
       "      <td>51</td>\n",
       "      <td>13</td>\n",
       "      <td>36</td>\n",
       "      <td>77</td>\n",
       "      <td>185</td>\n",
       "      <td>222</td>\n",
       "      <td>...</td>\n",
       "      <td>942</td>\n",
       "      <td>1019</td>\n",
       "      <td>1250</td>\n",
       "      <td>1663</td>\n",
       "      <td>1781</td>\n",
       "      <td>1513</td>\n",
       "      <td>1188</td>\n",
       "      <td>1000</td>\n",
       "      <td>638</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>256</td>\n",
       "      <td>171</td>\n",
       "      <td>104</td>\n",
       "      <td>67</td>\n",
       "      <td>51</td>\n",
       "      <td>26</td>\n",
       "      <td>25</td>\n",
       "      <td>41</td>\n",
       "      <td>170</td>\n",
       "      <td>192</td>\n",
       "      <td>...</td>\n",
       "      <td>1221</td>\n",
       "      <td>1168</td>\n",
       "      <td>1284</td>\n",
       "      <td>1400</td>\n",
       "      <td>1321</td>\n",
       "      <td>1099</td>\n",
       "      <td>791</td>\n",
       "      <td>498</td>\n",
       "      <td>272</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>130</td>\n",
       "      <td>68</td>\n",
       "      <td>52</td>\n",
       "      <td>23</td>\n",
       "      <td>10</td>\n",
       "      <td>16</td>\n",
       "      <td>28</td>\n",
       "      <td>128</td>\n",
       "      <td>226</td>\n",
       "      <td>271</td>\n",
       "      <td>...</td>\n",
       "      <td>1124</td>\n",
       "      <td>920</td>\n",
       "      <td>1063</td>\n",
       "      <td>1160</td>\n",
       "      <td>1184</td>\n",
       "      <td>893</td>\n",
       "      <td>720</td>\n",
       "      <td>541</td>\n",
       "      <td>304</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>78</td>\n",
       "      <td>60</td>\n",
       "      <td>51</td>\n",
       "      <td>17</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>22</td>\n",
       "      <td>73</td>\n",
       "      <td>172</td>\n",
       "      <td>186</td>\n",
       "      <td>...</td>\n",
       "      <td>846</td>\n",
       "      <td>872</td>\n",
       "      <td>1051</td>\n",
       "      <td>1200</td>\n",
       "      <td>1276</td>\n",
       "      <td>1004</td>\n",
       "      <td>841</td>\n",
       "      <td>525</td>\n",
       "      <td>315</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>95</td>\n",
       "      <td>63</td>\n",
       "      <td>45</td>\n",
       "      <td>26</td>\n",
       "      <td>3</td>\n",
       "      <td>23</td>\n",
       "      <td>28</td>\n",
       "      <td>108</td>\n",
       "      <td>209</td>\n",
       "      <td>241</td>\n",
       "      <td>...</td>\n",
       "      <td>1102</td>\n",
       "      <td>1083</td>\n",
       "      <td>1107</td>\n",
       "      <td>1109</td>\n",
       "      <td>1193</td>\n",
       "      <td>900</td>\n",
       "      <td>660</td>\n",
       "      <td>442</td>\n",
       "      <td>226</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>82</td>\n",
       "      <td>93</td>\n",
       "      <td>51</td>\n",
       "      <td>28</td>\n",
       "      <td>39</td>\n",
       "      <td>17</td>\n",
       "      <td>30</td>\n",
       "      <td>138</td>\n",
       "      <td>225</td>\n",
       "      <td>261</td>\n",
       "      <td>...</td>\n",
       "      <td>858</td>\n",
       "      <td>830</td>\n",
       "      <td>928</td>\n",
       "      <td>954</td>\n",
       "      <td>871</td>\n",
       "      <td>730</td>\n",
       "      <td>506</td>\n",
       "      <td>262</td>\n",
       "      <td>150</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    att1  att2  att3  att4  att5  att6  att7  att8  att9  att10  ...  att16  \\\n",
       "0    573   375   301   212    55    34    25    33   113    143  ...   1081   \n",
       "1    394   264   140   144   104    28    28    25    70    153  ...   1240   \n",
       "2    603   348   176   177    47    30    40    42   101    180  ...   1243   \n",
       "3    428   309   199   117    82    43    24    64   152    183  ...   1460   \n",
       "4    372   310   203   133    65    39    27    36   107    139  ...    971   \n",
       "5    448   344   183   146    71    14    30    41   108    137  ...   1062   \n",
       "6    621   322   221   150    65    40    42    84   148    190  ...   1279   \n",
       "7    597   409   142    93    48    30    34    87   132    157  ...   1706   \n",
       "8    525   431   248   240    91    64    29   117   200    236  ...   1398   \n",
       "9    587   382   165   192   130    44    21    35    73    132  ...   1069   \n",
       "10   144    73    21    16    10    12    26   100   177    220  ...   1367   \n",
       "11   141    63    51    14    16    14    28   103   172    205  ...    872   \n",
       "12    67    67   107    14    11    19    18    75   185    223  ...    765   \n",
       "13    69    54    35     6     7    15    19    76   212    238  ...    931   \n",
       "14   142   104    55    33    51    13    36    77   185    222  ...    942   \n",
       "15   256   171   104    67    51    26    25    41   170    192  ...   1221   \n",
       "16   130    68    52    23    10    16    28   128   226    271  ...   1124   \n",
       "17    78    60    51    17    10    12    22    73   172    186  ...    846   \n",
       "18    95    63    45    26     3    23    28   108   209    241  ...   1102   \n",
       "19    82    93    51    28    39    17    30   138   225    261  ...    858   \n",
       "\n",
       "    att17  att18  att19  att20  att21  att22  att23  att24  target  \n",
       "0     866   1096   1039    975    746    581    409    182       1  \n",
       "1    1109   1137   1290   1137    791    638    597    316       1  \n",
       "2    1141   1178   1256   1114    814    635    304    168       1  \n",
       "3    1365   1520   1700   1797   1596   1139    910    640       1  \n",
       "4    1104    844   1023   1019    862    643    591    452       1  \n",
       "5    1093   1030   1069   1151    898    754    467    362       1  \n",
       "6    1318   1336   1440   1479   1417   1347   1003    803       1  \n",
       "7    1520   1562   1608   1766   1533   1441   1252    964       1  \n",
       "8    1465   1459   1631   1891   1847   1731   1375   1188       1  \n",
       "9    1070   1164   1547   1575   1503   1139   1066    776       1  \n",
       "10   1319   1205   1250   1269    981    806    550    320       2  \n",
       "11    847   1227   1561   1611   1386   1007    810    680       2  \n",
       "12    824   1032   1237   1084    759    543    377    168       2  \n",
       "13    929   1093   1275   1161   1083    726    441    241       2  \n",
       "14   1019   1250   1663   1781   1513   1188   1000    638       2  \n",
       "15   1168   1284   1400   1321   1099    791    498    272       2  \n",
       "16    920   1063   1160   1184    893    720    541    304       2  \n",
       "17    872   1051   1200   1276   1004    841    525    315       2  \n",
       "18   1083   1107   1109   1193    900    660    442    226       2  \n",
       "19    830    928    954    871    730    506    262    150       2  \n",
       "\n",
       "[20 rows x 25 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset = pd.read_csv(f\"dataset/{dataset_name}/{dataset_name}_TRAIN.csv\")\n",
    "dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Shapelet:\n",
    "    def __init__(self, information_gain, series_id, start_pos, content):\n",
    "        self.information_gain = information_gain\n",
    "        self.series_id = series_id\n",
    "        self.start_pos = start_pos\n",
    "        self.content = content\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_shapelets(file):\n",
    "    shapelets = []\n",
    "    with open(file, encoding=\"utf8\") as f:\n",
    "        i = 0\n",
    "        line = f.readline()\n",
    "        while line:\n",
    "            if i == 0:\n",
    "                i += 1\n",
    "                line = f.readline().strip().strip(\",\")\n",
    "                continue\n",
    "            parts = line.split(\",\")\n",
    "            content = f.readline().strip().strip(\",\").split(\",\")\n",
    "            content = list(map(float, content))\n",
    "            shp = Shapelet(float(parts[0]), int(parts[1]), int(parts[2]), content)\n",
    "            shapelets.append(shp)\n",
    "            line = f.readline().strip().strip(\",\")\n",
    "    return shapelets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "st_shapelets = get_shapelets(f\"result_coef/{coef}/{model}/{dataset_name}/ED_Shapelets.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def recover(shapelet, dataset):\n",
    "    ts = dataset.loc[shapelet.series_id]\n",
    "    origin = ts[shapelet.start_pos : shapelet.start_pos + len(shapelet.content)]\n",
    "    mean = origin.mean()\n",
    "    std = origin.std()\n",
    "    print(\"Mean:\", mean, \"STD:\", std)\n",
    "    return [x*std + mean for x in shapelet.content]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_shapelet(shapelet, dataset, name):\n",
    "    origin = recover(shp, dataset)\n",
    "    plt.plot(range(shp.start_pos, shp.start_pos + len(shp.content)), origin, label=name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean: 207.0909090909091 STD: 296.6315743630319\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f0449d52da0>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 6))\n",
    "pos = 0\n",
    "shp = st_shapelets[pos]\n",
    "plot_shapelet(shp, dataset, \"shapelet\")\n",
    "plt.plot(dataset.loc[shp.series_id][:-1], label=\"st\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean: 207.0909090909091 STD: 296.6315743630319\n",
      "Series ID: 17 Start Pos: 2 Content: 11\n",
      "SHAPELET : [-0.9060988801189972, -0.6938137975185251, -0.647875364776118, -0.8149611688215105, -0.6561169339648202, -0.40912856226931965, 0.63067659075371, -0.04206459284245878, 0.30759229547283823, 0.543641185195891, 2.688149228889309]\n",
      "RECOVERED: [-61.68662824736916, 1.2838300181951467, 14.910619646345623, -34.6523054613522, 12.465910002679038, 85.73045954807706, 394.1694991200917, 194.61322269111062, 298.332495958956, 368.3520497441508, 1004.4808469791151]\n",
      "ORIGIN   : [  51   17   10   12   22   73  172  186  318  414 1003]\n"
     ]
    }
   ],
   "source": [
    "origin = recover(shp, dataset)\n",
    "origin_truth = dataset.loc[shp.series_id][shp.start_pos : shp.start_pos+len(shp.content)].values\n",
    "print(\"Series ID:\", shp.series_id, \"Start Pos:\", shp.start_pos, \"Content:\", len(shp.content))\n",
    "print(\"SHAPELET :\", shp.content)\n",
    "print(\"RECOVERED:\", origin)\n",
    "print(\"ORIGIN   :\", origin_truth)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.55189515, -0.67210993, -0.69686004, -0.68978858, -0.65443129,\n",
       "       -0.47410911, -0.12407194, -0.07457174,  0.39214448,  0.73157446,\n",
       "        2.81411882])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(origin_truth - np.mean(origin_truth)) / np.std(origin_truth)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}