result_viewer-adl-aggregated.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-19T09:59:36.815481Z",
     "start_time": "2022-05-19T09:57:25.686858Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3589999954254b9e9d689f0788eefa50",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/585 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "File test can not import!\n",
      "File 0-wardpaper-b can not import! 'accuracy'\n",
      "File 0-wardpaper-a can not import! 'accuracy'\n"
     ]
    }
   ],
   "source": [
    "import result_analyse.resultloader\n",
    "import general.utils as utils\n",
    "utils.reload()\n",
    "allrestable=result_analyse.resultloader.get_all_runs_table()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-08-19 10:24:59.552561: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "select columns to filter\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dec61246d1ea4c09b510c0d54eb7526f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "interactive(children=(Dropdown(description='split', options=('', 'train', 'test'), value=''), Dropdown(descrip…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from ipywidgets import interact, interactive, fixed, interact_manual\n",
    "import constants\n",
    "print('select columns to filter')\n",
    "def res(**filter):\n",
    "\tglobal allfilters\n",
    "\tallfilters=filter\n",
    "\tglobal filter_restable\n",
    "\tfilter_restable=allrestable.copy()\n",
    "\tfor k in filter:\n",
    "\t\tif filter[k]:\n",
    "\t\t\tfilter_restable=filter_restable.loc[filter_restable[k].astype(str)==filter[k]]\n",
    "\tprint(f'total data are={len(filter_restable)}, the first three rows are:')\n",
    "\tdisplay(filter_restable.iloc[0:3])\n",
    "\t\n",
    "\t\n",
    "params={}\n",
    "ignore_columns=['fold','runname','accuracy', 'precision', 'recall', 'f1',\n",
    "        'dataset_path', 'run_date',\n",
    "        'classifier_metric',\n",
    "       'classifier_metric_params', 'event_metric', 'event_metric_params','type']\n",
    "       \n",
    "for c in allrestable.columns.drop(ignore_columns):\n",
    "\t\n",
    "\t\n",
    "\tup=list(allrestable[c].dropna().astype(str).unique())\n",
    "\t\n",
    "\tif(len(up)<=1):continue\n",
    "\t# if(par=='name'):continue\n",
    "\tup.insert(0,'')\n",
    "\tparams[c]=up\n",
    "\t\n",
    "w=interact(res,**params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Aggregated Table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-27T11:52:40.875477Z",
     "start_time": "2020-10-27T11:52:40.808016Z"
    }
   },
   "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></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>classic_acc</th>\n",
       "      <th>classic_prc</th>\n",
       "      <th>classic_tpr</th>\n",
       "      <th>classic_f1</th>\n",
       "      <th>count</th>\n",
       "      <th>event_acc</th>\n",
       "      <th>event_prc</th>\n",
       "      <th>event_tpr</th>\n",
       "      <th>event_f1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"29\" valign=\"top\">test</th>\n",
       "      <th rowspan=\"29\" valign=\"top\">Home2</th>\n",
       "      <th rowspan=\"29\" valign=\"top\">FixedEventWindow</th>\n",
       "      <th>{'size': 10.0, 'shift': 1.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>51±0</td>\n",
       "      <td>35±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>19±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 2.0, 'shift': 1.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>33±0</td>\n",
       "      <td>51±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40±0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>52±0</td>\n",
       "      <td>28±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 1.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>66±0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>25±0</td>\n",
       "      <td>24±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>24±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 10.0}</th>\n",
       "      <td>52±0</td>\n",
       "      <td>42±0</td>\n",
       "      <td>55±0</td>\n",
       "      <td>43±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>24±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 15.0}</th>\n",
       "      <td>54±13</td>\n",
       "      <td>45±6</td>\n",
       "      <td>54±2</td>\n",
       "      <td>41±3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>30±2</td>\n",
       "      <td>29±7</td>\n",
       "      <td>43±4</td>\n",
       "      <td>24±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 2.0}</th>\n",
       "      <td>54±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>69±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>51±0</td>\n",
       "      <td>26±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 20.0}</th>\n",
       "      <td>21±0</td>\n",
       "      <td>11±0</td>\n",
       "      <td>38±0</td>\n",
       "      <td>10±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>18±0</td>\n",
       "      <td>10±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>12±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 5.0}</th>\n",
       "      <td>53±2</td>\n",
       "      <td>48±0</td>\n",
       "      <td>64±2</td>\n",
       "      <td>47±0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>31±2</td>\n",
       "      <td>50±0</td>\n",
       "      <td>27±3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 3.0, 'shift': 1.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>60±0</td>\n",
       "      <td>40±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40±0</td>\n",
       "      <td>33±0</td>\n",
       "      <td>56±0</td>\n",
       "      <td>34±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 3.0, 'shift': 3.0}</th>\n",
       "      <td>50±0</td>\n",
       "      <td>20±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20±0</td>\n",
       "      <td>19±0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>20±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 1.0}</th>\n",
       "      <td>53±0</td>\n",
       "      <td>48±0</td>\n",
       "      <td>62±0</td>\n",
       "      <td>48±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20±0</td>\n",
       "      <td>30±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>25±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 10.0}</th>\n",
       "      <td>31±0</td>\n",
       "      <td>35±0</td>\n",
       "      <td>70±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>21±0</td>\n",
       "      <td>25±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>26±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 15.0}</th>\n",
       "      <td>45±12</td>\n",
       "      <td>42±2</td>\n",
       "      <td>54±1</td>\n",
       "      <td>39±2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>22±1</td>\n",
       "      <td>25±0</td>\n",
       "      <td>43±2</td>\n",
       "      <td>21±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 2.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>42±0</td>\n",
       "      <td>70±0</td>\n",
       "      <td>50±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>21±0</td>\n",
       "      <td>25±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>24±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 20.0}</th>\n",
       "      <td>27±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>54±0</td>\n",
       "      <td>33±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>19±0</td>\n",
       "      <td>20±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>17±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 25.0}</th>\n",
       "      <td>37±16</td>\n",
       "      <td>38±2</td>\n",
       "      <td>58±3</td>\n",
       "      <td>41±4</td>\n",
       "      <td>2.0</td>\n",
       "      <td>22±2</td>\n",
       "      <td>29±2</td>\n",
       "      <td>43±0</td>\n",
       "      <td>24±3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 30.0}</th>\n",
       "      <td>50±0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>42±0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>24±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 5.0}</th>\n",
       "      <td>54±0</td>\n",
       "      <td>47±2</td>\n",
       "      <td>60±4</td>\n",
       "      <td>46±5</td>\n",
       "      <td>2.0</td>\n",
       "      <td>21±0</td>\n",
       "      <td>23±3</td>\n",
       "      <td>44±0</td>\n",
       "      <td>21±1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 4.0, 'shift': 1.0}</th>\n",
       "      <td>63±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>49±0</td>\n",
       "      <td>35±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>42±0</td>\n",
       "      <td>27±0</td>\n",
       "      <td>51±0</td>\n",
       "      <td>30±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 4.0, 'shift': 3.0}</th>\n",
       "      <td>47±0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>49±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>46±0</td>\n",
       "      <td>30±0</td>\n",
       "      <td>49±0</td>\n",
       "      <td>28±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 40.0, 'shift': 25.0}</th>\n",
       "      <td>51±0</td>\n",
       "      <td>43±0</td>\n",
       "      <td>56±0</td>\n",
       "      <td>43±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>30±0</td>\n",
       "      <td>26±0</td>\n",
       "      <td>43±0</td>\n",
       "      <td>26±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 40.0, 'shift': 35.0}</th>\n",
       "      <td>54±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>46±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>12±0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>16±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 40.0, 'shift': 5.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>62±0</td>\n",
       "      <td>49±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>21±0</td>\n",
       "      <td>25±0</td>\n",
       "      <td>40±0</td>\n",
       "      <td>22±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 5.0, 'shift': 1.0}</th>\n",
       "      <td>64±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>51±0</td>\n",
       "      <td>38±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>55±0</td>\n",
       "      <td>36±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 5.0, 'shift': 3.0}</th>\n",
       "      <td>57±0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>25±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>17±0</td>\n",
       "      <td>40±0</td>\n",
       "      <td>19±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 5.0, 'shift': 5.0}</th>\n",
       "      <td>52±0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>24±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>49±0</td>\n",
       "      <td>19±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>23±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 6.0, 'shift': 1.0}</th>\n",
       "      <td>66±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>42±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>46±0</td>\n",
       "      <td>33±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 6.0, 'shift': 3.0}</th>\n",
       "      <td>52±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>51±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>51±0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>56±0</td>\n",
       "      <td>40±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 6.0, 'shift': 5.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>30±0</td>\n",
       "      <td>55±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>21±0</td>\n",
       "      <td>48±0</td>\n",
       "      <td>23±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"29\" valign=\"top\">train</th>\n",
       "      <th rowspan=\"29\" valign=\"top\">Home2</th>\n",
       "      <th rowspan=\"29\" valign=\"top\">FixedEventWindow</th>\n",
       "      <th>{'size': 10.0, 'shift': 1.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>44±0</td>\n",
       "      <td>54±0</td>\n",
       "      <td>38±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>33±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>22±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 2.0, 'shift': 1.0}</th>\n",
       "      <td>54±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40±0</td>\n",
       "      <td>33±0</td>\n",
       "      <td>52±0</td>\n",
       "      <td>28±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 1.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>72±0</td>\n",
       "      <td>50±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>44±0</td>\n",
       "      <td>26±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 10.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>52±0</td>\n",
       "      <td>61±0</td>\n",
       "      <td>53±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>26±0</td>\n",
       "      <td>55±0</td>\n",
       "      <td>28±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 15.0}</th>\n",
       "      <td>59±11</td>\n",
       "      <td>56±4</td>\n",
       "      <td>63±1</td>\n",
       "      <td>53±2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>36±2</td>\n",
       "      <td>37±1</td>\n",
       "      <td>45±3</td>\n",
       "      <td>30±2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 2.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>68±0</td>\n",
       "      <td>46±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>51±0</td>\n",
       "      <td>31±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 20.0}</th>\n",
       "      <td>19±0</td>\n",
       "      <td>16±0</td>\n",
       "      <td>35±0</td>\n",
       "      <td>10±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17±0</td>\n",
       "      <td>14±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>12±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 20.0, 'shift': 5.0}</th>\n",
       "      <td>60±1</td>\n",
       "      <td>53±9</td>\n",
       "      <td>69±7</td>\n",
       "      <td>54±9</td>\n",
       "      <td>2.0</td>\n",
       "      <td>32±1</td>\n",
       "      <td>35±7</td>\n",
       "      <td>49±9</td>\n",
       "      <td>31±5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 3.0, 'shift': 1.0}</th>\n",
       "      <td>51±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>53±0</td>\n",
       "      <td>38±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>52±0</td>\n",
       "      <td>34±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 3.0, 'shift': 3.0}</th>\n",
       "      <td>46±0</td>\n",
       "      <td>22±0</td>\n",
       "      <td>30±0</td>\n",
       "      <td>22±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20±0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>27±0</td>\n",
       "      <td>18±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 1.0}</th>\n",
       "      <td>65±0</td>\n",
       "      <td>60±0</td>\n",
       "      <td>76±0</td>\n",
       "      <td>62±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>28±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 10.0}</th>\n",
       "      <td>37±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>78±0</td>\n",
       "      <td>42±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>19±0</td>\n",
       "      <td>25±0</td>\n",
       "      <td>38±0</td>\n",
       "      <td>23±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 15.0}</th>\n",
       "      <td>55±14</td>\n",
       "      <td>60±0</td>\n",
       "      <td>67±4</td>\n",
       "      <td>53±4</td>\n",
       "      <td>2.0</td>\n",
       "      <td>24±4</td>\n",
       "      <td>33±3</td>\n",
       "      <td>49±3</td>\n",
       "      <td>25±1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 2.0}</th>\n",
       "      <td>69±0</td>\n",
       "      <td>56±0</td>\n",
       "      <td>85±0</td>\n",
       "      <td>64±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>53±0</td>\n",
       "      <td>30±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 20.0}</th>\n",
       "      <td>27±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>60±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>15±0</td>\n",
       "      <td>18±0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>14±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 25.0}</th>\n",
       "      <td>43±12</td>\n",
       "      <td>47±3</td>\n",
       "      <td>69±2</td>\n",
       "      <td>49±0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>22±1</td>\n",
       "      <td>30±5</td>\n",
       "      <td>38±2</td>\n",
       "      <td>24±4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 30.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>33±0</td>\n",
       "      <td>33±0</td>\n",
       "      <td>35±0</td>\n",
       "      <td>23±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 30.0, 'shift': 5.0}</th>\n",
       "      <td>62±0</td>\n",
       "      <td>61±5</td>\n",
       "      <td>68±2</td>\n",
       "      <td>56±3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>22±0</td>\n",
       "      <td>27±4</td>\n",
       "      <td>43±1</td>\n",
       "      <td>24±2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 4.0, 'shift': 1.0}</th>\n",
       "      <td>60±0</td>\n",
       "      <td>35±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40±0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>46±0</td>\n",
       "      <td>32±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 4.0, 'shift': 3.0}</th>\n",
       "      <td>52±0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>35±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>46±0</td>\n",
       "      <td>33±0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>30±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 40.0, 'shift': 25.0}</th>\n",
       "      <td>61±0</td>\n",
       "      <td>56±0</td>\n",
       "      <td>70±0</td>\n",
       "      <td>59±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>39±0</td>\n",
       "      <td>27±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 40.0, 'shift': 35.0}</th>\n",
       "      <td>52±0</td>\n",
       "      <td>44±0</td>\n",
       "      <td>55±0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>15±0</td>\n",
       "      <td>26±0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>16±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 40.0, 'shift': 5.0}</th>\n",
       "      <td>70±0</td>\n",
       "      <td>63±0</td>\n",
       "      <td>83±0</td>\n",
       "      <td>70±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>23±0</td>\n",
       "      <td>27±0</td>\n",
       "      <td>49±0</td>\n",
       "      <td>25±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 5.0, 'shift': 1.0}</th>\n",
       "      <td>62±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>46±0</td>\n",
       "      <td>38±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>43±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>48±0</td>\n",
       "      <td>35±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 5.0, 'shift': 3.0}</th>\n",
       "      <td>56±0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>27±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>19±0</td>\n",
       "      <td>37±0</td>\n",
       "      <td>20±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 5.0, 'shift': 5.0}</th>\n",
       "      <td>49±0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>32±0</td>\n",
       "      <td>22±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50±0</td>\n",
       "      <td>34±0</td>\n",
       "      <td>31±0</td>\n",
       "      <td>23±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 6.0, 'shift': 1.0}</th>\n",
       "      <td>62±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>40±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>38±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>43±0</td>\n",
       "      <td>31±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 6.0, 'shift': 3.0}</th>\n",
       "      <td>55±0</td>\n",
       "      <td>48±0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>36±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>50±0</td>\n",
       "      <td>47±0</td>\n",
       "      <td>50±0</td>\n",
       "      <td>38±0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>{'size': 6.0, 'shift': 5.0}</th>\n",
       "      <td>52±0</td>\n",
       "      <td>41±0</td>\n",
       "      <td>53±0</td>\n",
       "      <td>38±0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>35±0</td>\n",
       "      <td>28±0</td>\n",
       "      <td>45±0</td>\n",
       "      <td>28±0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                           classic_acc  \\\n",
       "test  Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}         56±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}          56±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}         56±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}        52±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}       54±13   \n",
       "                             {'size': 20.0, 'shift': 2.0}         54±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}        21±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}         53±2   \n",
       "                             {'size': 3.0, 'shift': 1.0}          56±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}          50±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}         53±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}        31±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}       45±12   \n",
       "                             {'size': 30.0, 'shift': 2.0}         56±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}        27±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}       37±16   \n",
       "                             {'size': 30.0, 'shift': 30.0}        50±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}         54±0   \n",
       "                             {'size': 4.0, 'shift': 1.0}          63±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}          47±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}        51±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}        54±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}         56±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}          64±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}          57±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}          52±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}          66±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}          52±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}          56±0   \n",
       "train Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}         56±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}          54±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}         56±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}        56±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}       59±11   \n",
       "                             {'size': 20.0, 'shift': 2.0}         56±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}        19±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}         60±1   \n",
       "                             {'size': 3.0, 'shift': 1.0}          51±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}          46±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}         65±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}        37±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}       55±14   \n",
       "                             {'size': 30.0, 'shift': 2.0}         69±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}        27±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}       43±12   \n",
       "                             {'size': 30.0, 'shift': 30.0}        56±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}         62±0   \n",
       "                             {'size': 4.0, 'shift': 1.0}          60±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}          52±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}        61±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}        52±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}         70±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}          62±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}          56±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}          49±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}          62±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}          55±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}          52±0   \n",
       "\n",
       "                                                           classic_prc  \\\n",
       "test  Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}         36±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}          33±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}         39±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}        42±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}        45±6   \n",
       "                             {'size': 20.0, 'shift': 2.0}         37±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}        11±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}         48±0   \n",
       "                             {'size': 3.0, 'shift': 1.0}          36±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}          20±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}         48±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}        35±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}        42±2   \n",
       "                             {'size': 30.0, 'shift': 2.0}         42±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}        34±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}        38±2   \n",
       "                             {'size': 30.0, 'shift': 30.0}        31±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}         47±2   \n",
       "                             {'size': 4.0, 'shift': 1.0}          34±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}          32±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}        43±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}        36±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}         47±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}          34±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}          23±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}          23±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}          41±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}          37±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}          30±0   \n",
       "train Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}         44±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}          36±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}         45±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}        52±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}        56±4   \n",
       "                             {'size': 20.0, 'shift': 2.0}         41±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}        16±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}         53±9   \n",
       "                             {'size': 3.0, 'shift': 1.0}          37±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}          22±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}         60±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}        41±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}        60±0   \n",
       "                             {'size': 30.0, 'shift': 2.0}         56±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}        34±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}        47±3   \n",
       "                             {'size': 30.0, 'shift': 30.0}        41±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}         61±5   \n",
       "                             {'size': 4.0, 'shift': 1.0}          35±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}          39±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}        56±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}        44±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}         63±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}          37±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}          28±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}          32±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}          45±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}          48±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}          41±0   \n",
       "\n",
       "                                                           classic_tpr  \\\n",
       "test  Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}         51±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}          51±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}         66±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}        55±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}        54±2   \n",
       "                             {'size': 20.0, 'shift': 2.0}         69±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}        38±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}         64±2   \n",
       "                             {'size': 3.0, 'shift': 1.0}          60±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}          34±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}         62±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}        70±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}        54±1   \n",
       "                             {'size': 30.0, 'shift': 2.0}         70±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}        54±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}        58±3   \n",
       "                             {'size': 30.0, 'shift': 30.0}        42±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}         60±4   \n",
       "                             {'size': 4.0, 'shift': 1.0}          49±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}          49±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}        56±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}        46±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}         62±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}          51±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}          39±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}          36±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}          42±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}          51±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}          55±0   \n",
       "train Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}         54±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}          47±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}         72±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}        61±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}        63±1   \n",
       "                             {'size': 20.0, 'shift': 2.0}         68±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}        35±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}         69±7   \n",
       "                             {'size': 3.0, 'shift': 1.0}          53±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}          30±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}         76±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}        78±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}        67±4   \n",
       "                             {'size': 30.0, 'shift': 2.0}         85±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}        60±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}        69±2   \n",
       "                             {'size': 30.0, 'shift': 30.0}        45±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}         68±2   \n",
       "                             {'size': 4.0, 'shift': 1.0}          45±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}          45±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}        70±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}        55±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}         83±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}          46±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}          36±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}          32±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}          40±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}          47±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}          53±0   \n",
       "\n",
       "                                                           classic_f1 count  \\\n",
       "test  Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}        35±0   1.0   \n",
       "                             {'size': 2.0, 'shift': 1.0}         34±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 1.0}        47±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 10.0}       43±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 15.0}       41±3   2.0   \n",
       "                             {'size': 20.0, 'shift': 2.0}        45±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 20.0}       10±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 5.0}        47±0   2.0   \n",
       "                             {'size': 3.0, 'shift': 1.0}         40±0   1.0   \n",
       "                             {'size': 3.0, 'shift': 3.0}         23±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 1.0}        48±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 10.0}       37±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 15.0}       39±2   2.0   \n",
       "                             {'size': 30.0, 'shift': 2.0}        50±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 20.0}       33±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 25.0}       41±4   2.0   \n",
       "                             {'size': 30.0, 'shift': 30.0}       28±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 5.0}        46±5   2.0   \n",
       "                             {'size': 4.0, 'shift': 1.0}         35±0   1.0   \n",
       "                             {'size': 4.0, 'shift': 3.0}         34±0   1.0   \n",
       "                             {'size': 40.0, 'shift': 25.0}       43±0   1.0   \n",
       "                             {'size': 40.0, 'shift': 35.0}       37±0   1.0   \n",
       "                             {'size': 40.0, 'shift': 5.0}        49±0   1.0   \n",
       "                             {'size': 5.0, 'shift': 1.0}         38±0   1.0   \n",
       "                             {'size': 5.0, 'shift': 3.0}         25±0   1.0   \n",
       "                             {'size': 5.0, 'shift': 5.0}         24±0   1.0   \n",
       "                             {'size': 6.0, 'shift': 1.0}         37±0   1.0   \n",
       "                             {'size': 6.0, 'shift': 3.0}         37±0   1.0   \n",
       "                             {'size': 6.0, 'shift': 5.0}         36±0   1.0   \n",
       "train Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}        38±0   1.0   \n",
       "                             {'size': 2.0, 'shift': 1.0}         34±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 1.0}        50±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 10.0}       53±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 15.0}       53±2   2.0   \n",
       "                             {'size': 20.0, 'shift': 2.0}        46±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 20.0}       10±0   1.0   \n",
       "                             {'size': 20.0, 'shift': 5.0}        54±9   2.0   \n",
       "                             {'size': 3.0, 'shift': 1.0}         38±0   1.0   \n",
       "                             {'size': 3.0, 'shift': 3.0}         22±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 1.0}        62±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 10.0}       42±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 15.0}       53±4   2.0   \n",
       "                             {'size': 30.0, 'shift': 2.0}        64±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 20.0}       34±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 25.0}       49±0   2.0   \n",
       "                             {'size': 30.0, 'shift': 30.0}       37±0   1.0   \n",
       "                             {'size': 30.0, 'shift': 5.0}        56±3   2.0   \n",
       "                             {'size': 4.0, 'shift': 1.0}         34±0   1.0   \n",
       "                             {'size': 4.0, 'shift': 3.0}         35±0   1.0   \n",
       "                             {'size': 40.0, 'shift': 25.0}       59±0   1.0   \n",
       "                             {'size': 40.0, 'shift': 35.0}       47±0   1.0   \n",
       "                             {'size': 40.0, 'shift': 5.0}        70±0   1.0   \n",
       "                             {'size': 5.0, 'shift': 1.0}         38±0   1.0   \n",
       "                             {'size': 5.0, 'shift': 3.0}         27±0   1.0   \n",
       "                             {'size': 5.0, 'shift': 5.0}         22±0   1.0   \n",
       "                             {'size': 6.0, 'shift': 1.0}         36±0   1.0   \n",
       "                             {'size': 6.0, 'shift': 3.0}         36±0   1.0   \n",
       "                             {'size': 6.0, 'shift': 5.0}         38±0   1.0   \n",
       "\n",
       "                                                           event_acc  \\\n",
       "test  Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}       28±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}        40±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}       25±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}      34±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}      30±2   \n",
       "                             {'size': 20.0, 'shift': 2.0}       28±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}      18±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}       32±0   \n",
       "                             {'size': 3.0, 'shift': 1.0}        40±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}        20±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}       20±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}      21±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}      22±1   \n",
       "                             {'size': 30.0, 'shift': 2.0}       21±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}      19±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}      22±2   \n",
       "                             {'size': 30.0, 'shift': 30.0}      39±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}       21±0   \n",
       "                             {'size': 4.0, 'shift': 1.0}        42±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}        46±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}      30±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}      12±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}       21±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}        47±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}        32±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}        49±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}        39±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}        51±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}        32±0   \n",
       "train Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}       31±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}        40±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}       23±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}      37±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}      36±2   \n",
       "                             {'size': 20.0, 'shift': 2.0}       28±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}      17±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}       32±1   \n",
       "                             {'size': 3.0, 'shift': 1.0}        39±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}        20±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}       20±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}      19±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}      24±4   \n",
       "                             {'size': 30.0, 'shift': 2.0}       23±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}      15±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}      22±1   \n",
       "                             {'size': 30.0, 'shift': 30.0}      33±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}       22±0   \n",
       "                             {'size': 4.0, 'shift': 1.0}        40±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}        46±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}      28±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}      15±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}       23±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}        43±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}        31±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}        50±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}        38±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}        50±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}        35±0   \n",
       "\n",
       "                                                           event_prc  \\\n",
       "test  Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}       23±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}        31±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}       24±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}      23±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}      29±7   \n",
       "                             {'size': 20.0, 'shift': 2.0}       23±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}      10±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}       31±2   \n",
       "                             {'size': 3.0, 'shift': 1.0}        33±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}        19±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}       30±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}      25±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}      25±0   \n",
       "                             {'size': 30.0, 'shift': 2.0}       25±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}      20±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}      29±2   \n",
       "                             {'size': 30.0, 'shift': 30.0}      28±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}       23±3   \n",
       "                             {'size': 4.0, 'shift': 1.0}        27±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}        30±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}      26±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}      28±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}       25±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}        31±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}        17±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}        19±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}        34±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}        39±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}        21±0   \n",
       "train Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}       33±0   \n",
       "                             {'size': 2.0, 'shift': 1.0}        33±0   \n",
       "                             {'size': 20.0, 'shift': 1.0}       28±0   \n",
       "                             {'size': 20.0, 'shift': 10.0}      26±0   \n",
       "                             {'size': 20.0, 'shift': 15.0}      37±1   \n",
       "                             {'size': 20.0, 'shift': 2.0}       32±0   \n",
       "                             {'size': 20.0, 'shift': 20.0}      14±0   \n",
       "                             {'size': 20.0, 'shift': 5.0}       35±7   \n",
       "                             {'size': 3.0, 'shift': 1.0}        36±0   \n",
       "                             {'size': 3.0, 'shift': 3.0}        23±0   \n",
       "                             {'size': 30.0, 'shift': 1.0}       34±0   \n",
       "                             {'size': 30.0, 'shift': 10.0}      25±0   \n",
       "                             {'size': 30.0, 'shift': 15.0}      33±3   \n",
       "                             {'size': 30.0, 'shift': 2.0}       31±0   \n",
       "                             {'size': 30.0, 'shift': 20.0}      18±0   \n",
       "                             {'size': 30.0, 'shift': 25.0}      30±5   \n",
       "                             {'size': 30.0, 'shift': 30.0}      33±0   \n",
       "                             {'size': 30.0, 'shift': 5.0}       27±4   \n",
       "                             {'size': 4.0, 'shift': 1.0}        31±0   \n",
       "                             {'size': 4.0, 'shift': 3.0}        33±0   \n",
       "                             {'size': 40.0, 'shift': 25.0}      28±0   \n",
       "                             {'size': 40.0, 'shift': 35.0}      26±0   \n",
       "                             {'size': 40.0, 'shift': 5.0}       27±0   \n",
       "                             {'size': 5.0, 'shift': 1.0}        36±0   \n",
       "                             {'size': 5.0, 'shift': 3.0}        19±0   \n",
       "                             {'size': 5.0, 'shift': 5.0}        34±0   \n",
       "                             {'size': 6.0, 'shift': 1.0}        36±0   \n",
       "                             {'size': 6.0, 'shift': 3.0}        47±0   \n",
       "                             {'size': 6.0, 'shift': 5.0}        28±0   \n",
       "\n",
       "                                                           event_tpr event_f1  \n",
       "test  Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}       41±0     19±0  \n",
       "                             {'size': 2.0, 'shift': 1.0}        52±0     28±0  \n",
       "                             {'size': 20.0, 'shift': 1.0}       45±0     24±0  \n",
       "                             {'size': 20.0, 'shift': 10.0}      47±0     24±0  \n",
       "                             {'size': 20.0, 'shift': 15.0}      43±4     24±0  \n",
       "                             {'size': 20.0, 'shift': 2.0}       51±0     26±0  \n",
       "                             {'size': 20.0, 'shift': 20.0}      37±0     12±0  \n",
       "                             {'size': 20.0, 'shift': 5.0}       50±0     27±3  \n",
       "                             {'size': 3.0, 'shift': 1.0}        56±0     34±0  \n",
       "                             {'size': 3.0, 'shift': 3.0}        32±0     20±0  \n",
       "                             {'size': 30.0, 'shift': 1.0}       41±0     25±0  \n",
       "                             {'size': 30.0, 'shift': 10.0}      41±0     26±0  \n",
       "                             {'size': 30.0, 'shift': 15.0}      43±2     21±0  \n",
       "                             {'size': 30.0, 'shift': 2.0}       45±0     24±0  \n",
       "                             {'size': 30.0, 'shift': 20.0}      37±0     17±0  \n",
       "                             {'size': 30.0, 'shift': 25.0}      43±0     24±3  \n",
       "                             {'size': 30.0, 'shift': 30.0}      39±0     24±0  \n",
       "                             {'size': 30.0, 'shift': 5.0}       44±0     21±1  \n",
       "                             {'size': 4.0, 'shift': 1.0}        51±0     30±0  \n",
       "                             {'size': 4.0, 'shift': 3.0}        49±0     28±0  \n",
       "                             {'size': 40.0, 'shift': 25.0}      43±0     26±0  \n",
       "                             {'size': 40.0, 'shift': 35.0}      34±0     16±0  \n",
       "                             {'size': 40.0, 'shift': 5.0}       40±0     22±0  \n",
       "                             {'size': 5.0, 'shift': 1.0}        55±0     36±0  \n",
       "                             {'size': 5.0, 'shift': 3.0}        40±0     19±0  \n",
       "                             {'size': 5.0, 'shift': 5.0}        36±0     23±0  \n",
       "                             {'size': 6.0, 'shift': 1.0}        46±0     33±0  \n",
       "                             {'size': 6.0, 'shift': 3.0}        56±0     40±0  \n",
       "                             {'size': 6.0, 'shift': 5.0}        48±0     23±0  \n",
       "train Home2 FixedEventWindow {'size': 10.0, 'shift': 1.0}       37±0     22±0  \n",
       "                             {'size': 2.0, 'shift': 1.0}        52±0     28±0  \n",
       "                             {'size': 20.0, 'shift': 1.0}       44±0     26±0  \n",
       "                             {'size': 20.0, 'shift': 10.0}      55±0     28±0  \n",
       "                             {'size': 20.0, 'shift': 15.0}      45±3     30±2  \n",
       "                             {'size': 20.0, 'shift': 2.0}       51±0     31±0  \n",
       "                             {'size': 20.0, 'shift': 20.0}      34±0     12±0  \n",
       "                             {'size': 20.0, 'shift': 5.0}       49±9     31±5  \n",
       "                             {'size': 3.0, 'shift': 1.0}        52±0     34±0  \n",
       "                             {'size': 3.0, 'shift': 3.0}        27±0     18±0  \n",
       "                             {'size': 30.0, 'shift': 1.0}       41±0     28±0  \n",
       "                             {'size': 30.0, 'shift': 10.0}      38±0     23±0  \n",
       "                             {'size': 30.0, 'shift': 15.0}      49±3     25±1  \n",
       "                             {'size': 30.0, 'shift': 2.0}       53±0     30±0  \n",
       "                             {'size': 30.0, 'shift': 20.0}      31±0     14±0  \n",
       "                             {'size': 30.0, 'shift': 25.0}      38±2     24±4  \n",
       "                             {'size': 30.0, 'shift': 30.0}      35±0     23±0  \n",
       "                             {'size': 30.0, 'shift': 5.0}       43±1     24±2  \n",
       "                             {'size': 4.0, 'shift': 1.0}        46±0     32±0  \n",
       "                             {'size': 4.0, 'shift': 3.0}        47±0     30±0  \n",
       "                             {'size': 40.0, 'shift': 25.0}      39±0     27±0  \n",
       "                             {'size': 40.0, 'shift': 35.0}      32±0     16±0  \n",
       "                             {'size': 40.0, 'shift': 5.0}       49±0     25±0  \n",
       "                             {'size': 5.0, 'shift': 1.0}        48±0     35±0  \n",
       "                             {'size': 5.0, 'shift': 3.0}        37±0     20±0  \n",
       "                             {'size': 5.0, 'shift': 5.0}        31±0     23±0  \n",
       "                             {'size': 6.0, 'shift': 1.0}        43±0     31±0  \n",
       "                             {'size': 6.0, 'shift': 3.0}        50±0     38±0  \n",
       "                             {'size': 6.0, 'shift': 5.0}        45±0     28±0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "restable=filter_restable.copy()\n",
    "\n",
    "# restable=restable.loc[restable.evalution==\"KFoldEval\"]\n",
    "# restable.strategy.loc[restable.strategy.str.contains(\"FastFinder\")]='FastFinder'\n",
    "# restable.strategy.loc[restable.segmentor==\"Probabilistic\"]='SimpleStrategy'\n",
    "# restable=restable.loc[restable.strategy =='SimpleStrategy']#FastFinder\n",
    "# restable=restable.loc[restable.strategy =='FastFinder']#FastFinder\n",
    "# print(len(restable),len(allrestable))\n",
    "# res2=restable.groupby(['run_date','dataset','featureExtractor','featureExtractor_params','strategy','segmentor','segmentor_params','type']).mean()\n",
    "restable['segmentor_params']=restable['segmentor_params'].astype(str)\n",
    "# res2=restable.groupby(['split','run_date','dataset','segmentor','segmentor_params','type']).mean()\n",
    "res2=restable\n",
    "# display(res2)\n",
    "#res2.groupby(['dataset','runname','type']).std()\n",
    "from IPython.display import display\n",
    "# 'segmentor_params'\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "def calccell(item):\n",
    "    if np.isnan(item['mean']):\n",
    "        return 'nan'\n",
    "    if np.isnan(item['std']):\n",
    "        item['std']=0\n",
    "    return ('%d±%d')%(item['mean'],item['std'])\n",
    "res3=res2.groupby(['split','dataset','segmentor','segmentor_params','type']).agg({'accuracy':['mean','std'],'precision':['mean','std'],'recall':['mean','std'],'f1':['mean','std','count']})*100#.round(2)\n",
    "\n",
    "alllist={}\n",
    "for k,v in res3.iterrows():\n",
    "    key=k[0:len(k)-1]\n",
    "    if not(key in alllist):\n",
    "        alllist[key]={}\n",
    "    \n",
    "    alllist[key][k[len(k)-1]+\"_acc\"]=calccell(v['accuracy'])\n",
    "    alllist[key][k[len(k)-1]+\"_prc\"]=calccell(v['precision'])\n",
    "    alllist[key][k[len(k)-1]+\"_tpr\"]=calccell(v['recall'])\n",
    "    alllist[key][k[len(k)-1]+\"_f1\"]=calccell(v['f1'])\n",
    "    alllist[key][\"count\"]=v['f1']['count']/100\n",
    "\n",
    "df2=pd.DataFrame(alllist).transpose()\n",
    "\n",
    "display(df2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# display best result base on window size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "test-TS-CS       40\n",
       "test-classic     50\n",
       "train-TS-CS      38\n",
       "train-classic    70\n",
       "dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# eval(df2.index.get_level_values(3)[0])\n",
    "plot_data={}\n",
    "for row in df2.index:\n",
    "\tslic=row[0]\n",
    "\tdataset=row[1]\n",
    "\tsegment=row[2]\n",
    "\tsegparm=eval(row[3])\n",
    "\tseg_size=segparm['size']\n",
    "\tseg_shift=segparm['shift']\n",
    "\t# keye=f'{slic}-{dataset}-TS-CS'\n",
    "\t# keyc=f'{slic}-{dataset}-classic'\n",
    "\tkeye=f'{slic}-TS-CS'\n",
    "\tkeyc=f'{slic}-classic'\n",
    "\tif not keye in plot_data:\n",
    "\t\tplot_data[keye]={}\n",
    "\t\tplot_data[keyc]={}\n",
    "\t\n",
    "\t# met='tpr'\n",
    "\t# met='acc'\n",
    "\tmet='f1'\n",
    "\tplot_data[keye][(seg_size,seg_shift)]=int(df2.loc[row][f'event_{met}'].split('±')[0])\n",
    "\tplot_data[keyc][(seg_size,seg_shift)]=int(df2.loc[row][f'classic_{met}'].split('±')[0])\n",
    "\t\n",
    "segmap={'FixedEventWindow':'EW','FixedSlidingWindow':'TW', 'Probablistic': 'DW', 'MetaSeg':'MetaSeg'}\n",
    "\n",
    "a=pd.DataFrame(plot_data)\n",
    "from scipy.interpolate import interp1d\n",
    "b=a.reset_index().rename({'level_0':'size','level_1':'shift'},axis=1)\n",
    "sizeagg=b.groupby('size').max().drop('shift',axis=1)\n",
    "shiftagg=b.groupby('shift').max().drop('size',axis=1)\n",
    "# sizeagg.loc[20.0]=[33,37,41,49]\n",
    "ax=sizeagg.plot()\n",
    "ax.set_ylabel(f\"macro average of {met} on {allfilters['dataset']}\")\n",
    "ax.set_xlabel(f\"size parameter of {segmap[allfilters['segmentor']]} with the best shift parameter\")\n",
    "\n",
    "# shiftagg.plot()\n",
    "pd.set_option('display.max_rows', None)\n",
    "\n",
    "a.max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# display the relationship between window size, shift size and measure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-10-27T11:52:19.854348Z",
     "start_time": "2020-10-27T11:52:19.644196Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x576 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "x=a.index.get_level_values(0)\n",
    "y=a.index.get_level_values(1)\n",
    "\n",
    "# ax3d = plt.axes(projection='3d')\n",
    "fig,axs=plt.subplots(len(a.columns)//2,2,sharex=True,sharey=True,figsize=(10,2*len(a.columns)))\n",
    "axs=axs.reshape(-1)\n",
    "for i,c in enumerate(a.columns):\n",
    "\t# ax3d.scatter3D(x,y,a[c].values)\n",
    "\t# s=(a[c].values-a[c].values.min())*10\n",
    "\t# ax2d.heatmap(x,y,s=s)\n",
    "\timport seaborn as sns\n",
    "\tdata=a[[c]].unstack().droplevel(0,axis=1)\n",
    "\t\n",
    "\n",
    "\tsns.heatmap(data,annot=True,cmap='rocket_r',ax=axs[i])\n",
    "\taxs[i].set_title(c)\n",
    "\taxs[i].set_xlabel('shift size')\n",
    "\taxs[i].set_ylabel('window size')\n",
    "\t\n",
    "\t# a[[c]].heatmap()\n",
    "# plt.show()\n",
    "# ax2d = plt.axes()\n",
    "# for c in a.columns:\n",
    "# \t# ax3d.scatter3D(x,y,a[c].values)\n",
    "# \ts=(a[c].values-a[c].values.min())*10\n",
    "\t\n",
    "# \tax2d.plot(x,y,s=s)\n",
    "# a[c].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.13 (conda)",
   "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": "3.8.13"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  },
  "vscode": {
   "interpreter": {
    "hash": "0f91b223a9be7f82bfd164eb4b10e5e1c532d3ceac636b749a2ce370dda12911"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}