[15w] -5 pvalue 10thr.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "692ce7f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = 'all'\n",
    "\n",
    "import numpy as np\n",
    "from tqdm import tqdm\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a834ff9",
   "metadata": {},
   "source": [
    "## reg results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95c29b25",
   "metadata": {},
   "source": [
    "## 10thr\n",
    "reg_results = [[0, 'DecisionTree', 22.794256645140226, -0.0011155847315544865],\n",
    " [1, 'DecisionTree', 18.96271388027769, -0.0014687302722780071],\n",
    " [2, 'DecisionTree', 18.124000488338556, 0.0011064808393549397],\n",
    " [3, 'DecisionTree', 21.51762591563431, -0.001251284154066612],\n",
    " [4, 'DecisionTree', 19.81580289630115, 0.005772318106706087],\n",
    " [5, 'DecisionTree', 27.269565673037228, 0.00019830218510685782],\n",
    " [6, 'DecisionTree', 19.032642691370285, -0.0021424921225838833],\n",
    " [7, 'random_forest', 22.780907625720165, -0.0005293005815447671],\n",
    " [8, 'random_forest', 18.93809401437832, -0.00016849308063027202],\n",
    " [9, 'random_forest', 18.100895413712855, 0.00237990330026161],\n",
    " [10, 'random_forest', 21.430550372951856, 0.0028004871366440476],\n",
    " [11, 'random_forest', 19.851966790214593, 0.003957849896563226],\n",
    " [12, 'random_forest', 27.33704284646527, -0.002275657003169007],\n",
    " [13, 'random_forest', 19.01724333665828, -0.0013316563412153481],\n",
    " [14, 'xgboost_linear', 22.759476320043927, 0.0004119546849744582],\n",
    " [15, 'xgboost_linear', 18.950053650763202, -0.0008001116316724843],\n",
    " [16, 'xgboost_linear', 18.217737328199465, -0.004059769399046509],\n",
    " [17, 'xgboost_linear', 21.547575993560862, -0.0026449115970779147],\n",
    " [18, 'xgboost_linear', 19.950277990636852, -0.0009747646137581167],\n",
    " [19, 'xgboost_linear', 27.231096491384108, 0.0016087226366644503],\n",
    " [20, 'xgboost_linear', 18.983430803998893, 0.00044870470972901977],\n",
    " [21, 'gbm', 22.66350550393702, 0.004626958542295334],\n",
    " [22, 'gbm', 18.839675493713678, 0.005029237136024434],\n",
    " [23, 'gbm', 18.067964670981414, 0.004194861621231949],\n",
    " [24, 'gbm', 21.42896031286685, 0.0028744753037511783],\n",
    " [25, 'gbm', 19.88597288795631, 0.002251645817681558],\n",
    " [26, 'gbm', 27.186348184555268, 0.0032493587097960352],\n",
    " [27, 'gbm', 18.918224488273914, 0.0038820702597682777],\n",
    " [28, 'dummy', 22.78603871796385, -0.0007546563144305019],\n",
    " [29, 'dummy', 18.935811983865147, -4.7973295598247034e-05],\n",
    " [30, 'dummy', 18.151718651034788, -0.00042118922985090457],\n",
    " [31, 'dummy', 21.4990648745393, -0.0003876077287734425],\n",
    " [32, 'dummy', 19.93132568806165, -2.3861743231767818e-05],\n",
    " [33, 'dummy', 27.322843262729894, -0.001755048494140432],\n",
    " [34, 'dummy', 18.99269042730624, -3.885038852091682e-05],\n",
    " [35, 'linear', 23.077135649022804, -0.013539529227264424],\n",
    " [36, 'linear', 19.09897137410644, -0.008664832064262828],\n",
    " [37, 'linear', 18.280346149303995, -0.00751041737733571],\n",
    " [38, 'linear', 22.425477405919477, -0.043495139216569445],\n",
    " [39, 'linear', 20.046425022850755, -0.005798795285601344],\n",
    " [40, 'linear', 27.87582797086563, -0.022029483983461162],\n",
    " [41, 'linear', 19.309977548570078, -0.016745248526595313],\n",
    " [42, 'ridge', 23.306065148729115, -0.02359403082951994],\n",
    " [43, 'ridge', 19.06005737390709, -0.006609685600760473],\n",
    " [44, 'ridge', 18.237283396521025, -0.005137039341961858],\n",
    " [45, 'ridge', 22.232039999332216, -0.034494171707035504],\n",
    " [46, 'ridge', 20.025810793077458, -0.0047645077531315305],\n",
    " [47, 'ridge', 27.817106926808627, -0.019876556421299707],\n",
    " [48, 'ridge', 19.261822189369003, -0.014209682002209734],\n",
    " [49, 'lasso', 22.771354279694453, -0.00010972104703443897],\n",
    " [50, 'lasso', 18.93238326741051, 0.00013310586511483535],\n",
    " [51, 'lasso', 18.146462805806493, -0.00013151644267406226],\n",
    " [52, 'lasso', 21.45077175674837, 0.001859552176020851],\n",
    " [53, 'lasso', 19.91729404759222, 0.0006801544124640513],\n",
    " [54, 'lasso', 27.356152534703206, -0.002976287844674852],\n",
    " [55, 'lasso', 19.04984034423305, -0.0030480152797209836],\n",
    " [56, 'elasticnet', 22.770731941986828, -8.238819789774254e-05],\n",
    " [57, 'elasticnet', 18.93048798894362, 0.0002332002994733795],\n",
    " [58, 'elasticnet', 18.14750007677335, -0.00018868501575330932],\n",
    " [59, 'elasticnet', 21.449773483456372, 0.0019060035094284888],\n",
    " [60, 'elasticnet', 19.91649399489795, 0.0007202958359422418],\n",
    " [61, 'elasticnet', 27.364896630258233, -0.003296878267220915],\n",
    " [62, 'elasticnet', 19.04955444453375, -0.003032961551145519],\n",
    " [63, 'ardr_linear', 22.78603871796386, -0.0007546563144307239],\n",
    " [64, 'ardr_linear', 18.93581198386515, -4.797329559846908e-05],\n",
    " [65, 'ardr_linear', 18.151718651034784, -0.00042118922985090457],\n",
    " [66, 'ardr_linear', 21.4990648745393, -0.0003876077287734425],\n",
    " [67, 'ardr_linear', 19.931325688061655, -2.3861743232211907e-05],\n",
    " [68, 'ardr_linear', 27.322843262729894, -0.0017550484941408762],\n",
    " [69, 'ardr_linear', 18.992690427306236, -3.8850388520472734e-05],\n",
    " [70, 'bayesian_ridge', 39.818135870151224, -0.7487982606822892],\n",
    " [71, 'bayesian_ridge', 236.85940524581002, -11.509142379207521],\n",
    " [72, 'bayesian_ridge', 18.22978282334477, -0.004723649707561339],\n",
    " [73, 'bayesian_ridge', 22.190173980930584, -0.032546075534557684],\n",
    " [74, 'bayesian_ridge', 19.976560771010536, -0.002293463025393905],\n",
    " [75, 'bayesian_ridge', 27.42650592924302, -0.005555699785126267],\n",
    " [76, 'bayesian_ridge', 352.461131627053, -17.558446273214344]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b06a7cb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "nm_dic = {i:'' for i in ['linear', 'ridge', 'lasso', 'elasticnet',\n",
    "              'ardr_linear', 'bayesian_ridge',\n",
    "              'DecisionTree', 'random_forest',\n",
    "              'xgboost_linear', 'gbm', 'dummy']}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "2340df3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "nm_list = ['Linear', 'Ridge', 'Lasso', 'ElasticNet',\n",
    "              'ARDR_Linear', 'Bayesian_Ridge',\n",
    "              'DecisionTree', 'Random_Forest',\n",
    "              'XGBoost_Linear', 'LGBM', 'dummy']\n",
    "for idx,i in enumerate(nm_dic):\n",
    "    nm_dic[i] = nm_list[idx]\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f3a10a83",
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[0, 'DecisionTree', 22.794256645140226, -0.0011155847315544865],\n",
       " [1, 'DecisionTree', 18.96271388027769, -0.0014687302722780071],\n",
       " [2, 'DecisionTree', 18.124000488338556, 0.0011064808393549397],\n",
       " [3, 'DecisionTree', 21.51762591563431, -0.001251284154066612],\n",
       " [4, 'DecisionTree', 19.81580289630115, 0.005772318106706087],\n",
       " [5, 'DecisionTree', 27.269565673037228, 0.00019830218510685782],\n",
       " [6, 'DecisionTree', 19.032642691370285, -0.0021424921225838833],\n",
       " [7, 'Random_Forest', 22.780907625720165, -0.0005293005815447671],\n",
       " [8, 'Random_Forest', 18.93809401437832, -0.00016849308063027202],\n",
       " [9, 'Random_Forest', 18.100895413712855, 0.00237990330026161],\n",
       " [10, 'Random_Forest', 21.430550372951856, 0.0028004871366440476],\n",
       " [11, 'Random_Forest', 19.851966790214593, 0.003957849896563226],\n",
       " [12, 'Random_Forest', 27.33704284646527, -0.002275657003169007],\n",
       " [13, 'Random_Forest', 19.01724333665828, -0.0013316563412153481],\n",
       " [14, 'XGBoost_Linear', 22.759476320043927, 0.0004119546849744582],\n",
       " [15, 'XGBoost_Linear', 18.950053650763202, -0.0008001116316724843],\n",
       " [16, 'XGBoost_Linear', 18.217737328199465, -0.004059769399046509],\n",
       " [17, 'XGBoost_Linear', 21.547575993560862, -0.0026449115970779147],\n",
       " [18, 'XGBoost_Linear', 19.950277990636852, -0.0009747646137581167],\n",
       " [19, 'XGBoost_Linear', 27.231096491384108, 0.0016087226366644503],\n",
       " [20, 'XGBoost_Linear', 18.983430803998893, 0.00044870470972901977],\n",
       " [21, 'LGBM', 22.66350550393702, 0.004626958542295334],\n",
       " [22, 'LGBM', 18.839675493713678, 0.005029237136024434],\n",
       " [23, 'LGBM', 18.067964670981414, 0.004194861621231949],\n",
       " [24, 'LGBM', 21.42896031286685, 0.0028744753037511783],\n",
       " [25, 'LGBM', 19.88597288795631, 0.002251645817681558],\n",
       " [26, 'LGBM', 27.186348184555268, 0.0032493587097960352],\n",
       " [27, 'LGBM', 18.918224488273914, 0.0038820702597682777],\n",
       " [28, 'dummy', 22.78603871796385, -0.0007546563144305019],\n",
       " [29, 'dummy', 18.935811983865147, -4.7973295598247034e-05],\n",
       " [30, 'dummy', 18.151718651034788, -0.00042118922985090457],\n",
       " [31, 'dummy', 21.4990648745393, -0.0003876077287734425],\n",
       " [32, 'dummy', 19.93132568806165, -2.3861743231767818e-05],\n",
       " [33, 'dummy', 27.322843262729894, -0.001755048494140432],\n",
       " [34, 'dummy', 18.99269042730624, -3.885038852091682e-05],\n",
       " [35, 'Linear', 23.077135649022804, -0.013539529227264424],\n",
       " [36, 'Linear', 19.09897137410644, -0.008664832064262828],\n",
       " [37, 'Linear', 18.280346149303995, -0.00751041737733571],\n",
       " [38, 'Linear', 22.425477405919477, -0.043495139216569445],\n",
       " [39, 'Linear', 20.046425022850755, -0.005798795285601344],\n",
       " [40, 'Linear', 27.87582797086563, -0.022029483983461162],\n",
       " [41, 'Linear', 19.309977548570078, -0.016745248526595313],\n",
       " [42, 'Ridge', 23.306065148729115, -0.02359403082951994],\n",
       " [43, 'Ridge', 19.06005737390709, -0.006609685600760473],\n",
       " [44, 'Ridge', 18.237283396521025, -0.005137039341961858],\n",
       " [45, 'Ridge', 22.232039999332216, -0.034494171707035504],\n",
       " [46, 'Ridge', 20.025810793077458, -0.0047645077531315305],\n",
       " [47, 'Ridge', 27.817106926808627, -0.019876556421299707],\n",
       " [48, 'Ridge', 19.261822189369003, -0.014209682002209734],\n",
       " [49, 'Lasso', 22.771354279694453, -0.00010972104703443897],\n",
       " [50, 'Lasso', 18.93238326741051, 0.00013310586511483535],\n",
       " [51, 'Lasso', 18.146462805806493, -0.00013151644267406226],\n",
       " [52, 'Lasso', 21.45077175674837, 0.001859552176020851],\n",
       " [53, 'Lasso', 19.91729404759222, 0.0006801544124640513],\n",
       " [54, 'Lasso', 27.356152534703206, -0.002976287844674852],\n",
       " [55, 'Lasso', 19.04984034423305, -0.0030480152797209836],\n",
       " [56, 'ElasticNet', 22.770731941986828, -8.238819789774254e-05],\n",
       " [57, 'ElasticNet', 18.93048798894362, 0.0002332002994733795],\n",
       " [58, 'ElasticNet', 18.14750007677335, -0.00018868501575330932],\n",
       " [59, 'ElasticNet', 21.449773483456372, 0.0019060035094284888],\n",
       " [60, 'ElasticNet', 19.91649399489795, 0.0007202958359422418],\n",
       " [61, 'ElasticNet', 27.364896630258233, -0.003296878267220915],\n",
       " [62, 'ElasticNet', 19.04955444453375, -0.003032961551145519],\n",
       " [63, 'ARDR_Linear', 22.78603871796386, -0.0007546563144307239],\n",
       " [64, 'ARDR_Linear', 18.93581198386515, -4.797329559846908e-05],\n",
       " [65, 'ARDR_Linear', 18.151718651034784, -0.00042118922985090457],\n",
       " [66, 'ARDR_Linear', 21.4990648745393, -0.0003876077287734425],\n",
       " [67, 'ARDR_Linear', 19.931325688061655, -2.3861743232211907e-05],\n",
       " [68, 'ARDR_Linear', 27.322843262729894, -0.0017550484941408762],\n",
       " [69, 'ARDR_Linear', 18.992690427306236, -3.8850388520472734e-05],\n",
       " [70, 'Bayesian_Ridge', 39.818135870151224, -0.7487982606822892],\n",
       " [71, 'Bayesian_Ridge', 236.85940524581002, -11.509142379207521],\n",
       " [72, 'Bayesian_Ridge', 18.22978282334477, -0.004723649707561339],\n",
       " [73, 'Bayesian_Ridge', 22.190173980930584, -0.032546075534557684],\n",
       " [74, 'Bayesian_Ridge', 19.976560771010536, -0.002293463025393905],\n",
       " [75, 'Bayesian_Ridge', 27.42650592924302, -0.005555699785126267],\n",
       " [76, 'Bayesian_Ridge', 352.461131627053, -17.558446273214344]]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for i in range(len(reg_results)):\n",
    "#     reg_results[i][1] = (reg_results[i][1][0]).upper()+reg_results[i][1][1:]\n",
    "\n",
    "    if reg_results[i][1] in nm_dic:\n",
    "        reg_results[i][1] = nm_dic[reg_results[i][1]]\n",
    "    \n",
    "reg_results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53067bcc",
   "metadata": {},
   "source": [
    "## plot을 위한 준비"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "6b591239",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('LGBM', 20.998664506040637),\n",
       " ('Random_Forest', 21.065242914300192),\n",
       " ('DecisionTree', 21.073801170014207),\n",
       " ('dummy', 21.08849908650012),\n",
       " ('ARDR_Linear', 21.088499086500125),\n",
       " ('Lasso', 21.089179862312616),\n",
       " ('ElasticNet', 21.08991979440716),\n",
       " ('XGBoost_Linear', 21.091378368369615),\n",
       " ('Ridge', 21.420026546820647),\n",
       " ('Linear', 21.444880160091316),\n",
       " ('Bayesian_Ridge', 102.42309946393473)]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "[('Bayesian_Ridge', -4.265929400165256),\n",
       " ('Linear', -0.01682620652587003),\n",
       " ('Ridge', -0.015526524807988393),\n",
       " ('XGBoost_Linear', -0.0008585964585981566),\n",
       " ('ElasticNet', -0.0005344876267390537),\n",
       " ('Lasso', -0.0005132468800720856),\n",
       " ('ARDR_Linear', -0.0004898838849353002),\n",
       " ('dummy', -0.0004898838849351732),\n",
       " ('DecisionTree', 0.00015700140724069937),\n",
       " ('Random_Forest', 0.000690447618129927),\n",
       " ('LGBM', 0.003729801055792681)]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_list = ['Linear', 'Ridge', 'Lasso', 'ElasticNet',\n",
    "              'ARDR_Linear', 'Bayesian_Ridge',\n",
    "              'DecisionTree', 'Random_Forest',\n",
    "              'XGBoost_Linear', 'LGBM', 'dummy']\n",
    "# ['linear', 'ridge', 'lasso', 'elasticnet', 'ardr_linear', 'bayesian_ridge', 'DecisionTree', 'random_forest', 'xgboost_linear', 'lgbm', 'dummy']\n",
    "mse_dic = {i:[] for i in model_list}\n",
    "r2_dic = {i:[] for i in model_list}\n",
    "for i in reg_results:\n",
    "    mse_dic[i[1]].append(i[2])\n",
    "    r2_dic[i[1]].append(i[3])\n",
    "\n",
    "\n",
    "# import math\n",
    "# 리스트에 mean 함수가 내장이 안돼있다니...\n",
    "mean_mse_score = {i:[] for i in model_list}\n",
    "mean_r2_score = {i:[] for i in model_list}\n",
    "mean_cv_score = {i:[] for i in model_list}\n",
    "for i in model_list:\n",
    "    mean_mse_score[i] = sum(mse_dic[i])/len(mse_dic[i])\n",
    "    mean_r2_score[i] = sum(r2_dic[i])/len(r2_dic[i])\n",
    "\n",
    "# 정렬\n",
    "mse_score_lst = sorted(mean_mse_score.items(), key = lambda t : t[1]) #mse 작은 순으로 줄 세워짐.\n",
    "r2_score_lst = sorted(mean_r2_score.items(), key = lambda t : t[1]) #r2 작은 순으로 줄 세워짐.\n",
    "\n",
    "mse_score_lst\n",
    "r2_score_lst"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "aee9251a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 다시 dic\n",
    "mse_score_dic = {}\n",
    "r2_score_dic = {}\n",
    "\n",
    "for i in mse_score_lst:\n",
    "    mse_score_dic[i[0]] = i[1]\n",
    "r2_score_dic = {i[0]:i[1] for i in r2_score_lst}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "f949f15a",
   "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>key</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>20.998665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Random_Forest</td>\n",
       "      <td>21.065243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>21.073801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>dummy</td>\n",
       "      <td>21.088499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ARDR_Linear</td>\n",
       "      <td>21.088499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Lasso</td>\n",
       "      <td>21.089180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ElasticNet</td>\n",
       "      <td>21.089920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>XGBoost_Linear</td>\n",
       "      <td>21.091378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Ridge</td>\n",
       "      <td>21.420027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Linear</td>\n",
       "      <td>21.444880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>102.423099</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               key       value\n",
       "0             LGBM   20.998665\n",
       "1    Random_Forest   21.065243\n",
       "2     DecisionTree   21.073801\n",
       "3            dummy   21.088499\n",
       "4      ARDR_Linear   21.088499\n",
       "5            Lasso   21.089180\n",
       "6       ElasticNet   21.089920\n",
       "7   XGBoost_Linear   21.091378\n",
       "8            Ridge   21.420027\n",
       "9           Linear   21.444880\n",
       "10  Bayesian_Ridge  102.423099"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>key</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>-4.265929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Linear</td>\n",
       "      <td>-0.016826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Ridge</td>\n",
       "      <td>-0.015527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>XGBoost_Linear</td>\n",
       "      <td>-0.000859</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ElasticNet</td>\n",
       "      <td>-0.000534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Lasso</td>\n",
       "      <td>-0.000513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ARDR_Linear</td>\n",
       "      <td>-0.000490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>dummy</td>\n",
       "      <td>-0.000490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>0.000157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Random_Forest</td>\n",
       "      <td>0.000690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>0.003730</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               key     value\n",
       "0   Bayesian_Ridge -4.265929\n",
       "1           Linear -0.016826\n",
       "2            Ridge -0.015527\n",
       "3   XGBoost_Linear -0.000859\n",
       "4       ElasticNet -0.000534\n",
       "5            Lasso -0.000513\n",
       "6      ARDR_Linear -0.000490\n",
       "7            dummy -0.000490\n",
       "8     DecisionTree  0.000157\n",
       "9    Random_Forest  0.000690\n",
       "10            LGBM  0.003730"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mse_df = pd.DataFrame(mse_score_dic.items(), columns=['key', 'value'])\n",
    "mse_df\n",
    "\n",
    "r2_df = pd.DataFrame(r2_score_dic.items(), columns=['key', 'value'])\n",
    "r2_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83a847ef",
   "metadata": {},
   "source": [
    "## 테스트 데이터셋 정확도 플랏 (유의성 기호 추가)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "20bc35e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "from statannot import add_stat_annotation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "46573304",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dum_idx = list(mse_df.iloc[:,0].values).index('dummy')\n",
    "dum_idx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "8755481b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'dummy'"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dummy랑 비교하라고 하셨던 것 같음.\n",
    "mse_df.iloc[dum_idx,0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "86eb6c85",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df_cla_results = pd.DataFrame(reg_results,\n",
    "             columns= ['iter', 'method', 'mse', 'r2_score'])\n",
    "df_cla_results['iter'] = [i for i in range(7)]*len(model_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "d170a020",
   "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>iter</th>\n",
       "      <th>method</th>\n",
       "      <th>mse</th>\n",
       "      <th>r2_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>22.794257</td>\n",
       "      <td>-0.001116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>18.962714</td>\n",
       "      <td>-0.001469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>18.124000</td>\n",
       "      <td>0.001106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>21.517626</td>\n",
       "      <td>-0.001251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>19.815803</td>\n",
       "      <td>0.005772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72</th>\n",
       "      <td>2</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>18.229783</td>\n",
       "      <td>-0.004724</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>3</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>22.190174</td>\n",
       "      <td>-0.032546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74</th>\n",
       "      <td>4</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>19.976561</td>\n",
       "      <td>-0.002293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>5</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>27.426506</td>\n",
       "      <td>-0.005556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>6</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>352.461132</td>\n",
       "      <td>-17.558446</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>77 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    iter          method         mse   r2_score\n",
       "0      0    DecisionTree   22.794257  -0.001116\n",
       "1      1    DecisionTree   18.962714  -0.001469\n",
       "2      2    DecisionTree   18.124000   0.001106\n",
       "3      3    DecisionTree   21.517626  -0.001251\n",
       "4      4    DecisionTree   19.815803   0.005772\n",
       "..   ...             ...         ...        ...\n",
       "72     2  Bayesian_Ridge   18.229783  -0.004724\n",
       "73     3  Bayesian_Ridge   22.190174  -0.032546\n",
       "74     4  Bayesian_Ridge   19.976561  -0.002293\n",
       "75     5  Bayesian_Ridge   27.426506  -0.005556\n",
       "76     6  Bayesian_Ridge  352.461132 -17.558446\n",
       "\n",
       "[77 rows x 4 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_cla_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "5f4d857c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_acc_test = df_cla_results.pivot_table(index='method', columns='iter')[\"mse\"].reset_index()\n",
    "# df_acc_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "968a3e0c",
   "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>iter</th>\n",
       "      <th>method</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>22.663506</td>\n",
       "      <td>18.839675</td>\n",
       "      <td>18.067965</td>\n",
       "      <td>21.428960</td>\n",
       "      <td>19.885973</td>\n",
       "      <td>27.186348</td>\n",
       "      <td>18.918224</td>\n",
       "      <td>20.998665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Random_Forest</td>\n",
       "      <td>22.780908</td>\n",
       "      <td>18.938094</td>\n",
       "      <td>18.100895</td>\n",
       "      <td>21.430550</td>\n",
       "      <td>19.851967</td>\n",
       "      <td>27.337043</td>\n",
       "      <td>19.017243</td>\n",
       "      <td>21.065243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>22.794257</td>\n",
       "      <td>18.962714</td>\n",
       "      <td>18.124000</td>\n",
       "      <td>21.517626</td>\n",
       "      <td>19.815803</td>\n",
       "      <td>27.269566</td>\n",
       "      <td>19.032643</td>\n",
       "      <td>21.073801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>dummy</td>\n",
       "      <td>22.786039</td>\n",
       "      <td>18.935812</td>\n",
       "      <td>18.151719</td>\n",
       "      <td>21.499065</td>\n",
       "      <td>19.931326</td>\n",
       "      <td>27.322843</td>\n",
       "      <td>18.992690</td>\n",
       "      <td>21.088499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ARDR_Linear</td>\n",
       "      <td>22.786039</td>\n",
       "      <td>18.935812</td>\n",
       "      <td>18.151719</td>\n",
       "      <td>21.499065</td>\n",
       "      <td>19.931326</td>\n",
       "      <td>27.322843</td>\n",
       "      <td>18.992690</td>\n",
       "      <td>21.088499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Lasso</td>\n",
       "      <td>22.771354</td>\n",
       "      <td>18.932383</td>\n",
       "      <td>18.146463</td>\n",
       "      <td>21.450772</td>\n",
       "      <td>19.917294</td>\n",
       "      <td>27.356153</td>\n",
       "      <td>19.049840</td>\n",
       "      <td>21.089180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ElasticNet</td>\n",
       "      <td>22.770732</td>\n",
       "      <td>18.930488</td>\n",
       "      <td>18.147500</td>\n",
       "      <td>21.449773</td>\n",
       "      <td>19.916494</td>\n",
       "      <td>27.364897</td>\n",
       "      <td>19.049554</td>\n",
       "      <td>21.089920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>XGBoost_Linear</td>\n",
       "      <td>22.759476</td>\n",
       "      <td>18.950054</td>\n",
       "      <td>18.217737</td>\n",
       "      <td>21.547576</td>\n",
       "      <td>19.950278</td>\n",
       "      <td>27.231096</td>\n",
       "      <td>18.983431</td>\n",
       "      <td>21.091378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Ridge</td>\n",
       "      <td>23.306065</td>\n",
       "      <td>19.060057</td>\n",
       "      <td>18.237283</td>\n",
       "      <td>22.232040</td>\n",
       "      <td>20.025811</td>\n",
       "      <td>27.817107</td>\n",
       "      <td>19.261822</td>\n",
       "      <td>21.420027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Linear</td>\n",
       "      <td>23.077136</td>\n",
       "      <td>19.098971</td>\n",
       "      <td>18.280346</td>\n",
       "      <td>22.425477</td>\n",
       "      <td>20.046425</td>\n",
       "      <td>27.875828</td>\n",
       "      <td>19.309978</td>\n",
       "      <td>21.444880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>39.818136</td>\n",
       "      <td>236.859405</td>\n",
       "      <td>18.229783</td>\n",
       "      <td>22.190174</td>\n",
       "      <td>19.976561</td>\n",
       "      <td>27.426506</td>\n",
       "      <td>352.461132</td>\n",
       "      <td>102.423099</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "iter          method          0           1          2          3          4  \\\n",
       "4               LGBM  22.663506   18.839675  18.067965  21.428960  19.885973   \n",
       "7      Random_Forest  22.780908   18.938094  18.100895  21.430550  19.851967   \n",
       "2       DecisionTree  22.794257   18.962714  18.124000  21.517626  19.815803   \n",
       "10             dummy  22.786039   18.935812  18.151719  21.499065  19.931326   \n",
       "0        ARDR_Linear  22.786039   18.935812  18.151719  21.499065  19.931326   \n",
       "5              Lasso  22.771354   18.932383  18.146463  21.450772  19.917294   \n",
       "3         ElasticNet  22.770732   18.930488  18.147500  21.449773  19.916494   \n",
       "9     XGBoost_Linear  22.759476   18.950054  18.217737  21.547576  19.950278   \n",
       "8              Ridge  23.306065   19.060057  18.237283  22.232040  20.025811   \n",
       "6             Linear  23.077136   19.098971  18.280346  22.425477  20.046425   \n",
       "1     Bayesian_Ridge  39.818136  236.859405  18.229783  22.190174  19.976561   \n",
       "\n",
       "iter          5           6        mean  \n",
       "4     27.186348   18.918224   20.998665  \n",
       "7     27.337043   19.017243   21.065243  \n",
       "2     27.269566   19.032643   21.073801  \n",
       "10    27.322843   18.992690   21.088499  \n",
       "0     27.322843   18.992690   21.088499  \n",
       "5     27.356153   19.049840   21.089180  \n",
       "3     27.364897   19.049554   21.089920  \n",
       "9     27.231096   18.983431   21.091378  \n",
       "8     27.817107   19.261822   21.420027  \n",
       "6     27.875828   19.309978   21.444880  \n",
       "1     27.426506  352.461132  102.423099  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_acc_test['mean']=np.mean(df_acc_test, axis=1)\n",
    "df_acc_test=df_acc_test.sort_values('mean')\n",
    "df_acc_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "111e348d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LGBM vs. dummy WilcoxonResult(statistic=0.0, pvalue=0.015625)\n",
      "Random_Forest vs. dummy WilcoxonResult(statistic=8.0, pvalue=0.375)\n",
      "DecisionTree vs. dummy WilcoxonResult(statistic=11.0, pvalue=0.6875)\n",
      "ARDR_Linear vs. dummy WilcoxonResult(statistic=5.0, pvalue=0.4795001221869535)\n",
      "Lasso vs. dummy WilcoxonResult(statistic=12.0, pvalue=0.8125)\n",
      "ElasticNet vs. dummy WilcoxonResult(statistic=12.0, pvalue=0.8125)\n",
      "XGBoost_Linear vs. dummy WilcoxonResult(statistic=12.0, pvalue=0.8125)\n",
      "Ridge vs. dummy WilcoxonResult(statistic=0.0, pvalue=0.015625)\n",
      "Linear vs. dummy WilcoxonResult(statistic=0.0, pvalue=0.015625)\n",
      "Bayesian_Ridge vs. dummy WilcoxonResult(statistic=0.0, pvalue=0.015625)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from scipy import stats\n",
    "\n",
    "# 비모수적 검정(parametric test)\n",
    "for i in range(len(model_list)):\n",
    "    if i == dum_idx:\n",
    "        continue\n",
    "    print(df_acc_test.iloc[i,0], 'vs.', df_acc_test.iloc[dum_idx,0],\n",
    "          stats.wilcoxon(df_acc_test.iloc[i,1:-1], df_acc_test.iloc[dum_idx,1:-1])) # Wilcoxon signed-rank test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "4cb5f307",
   "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>iter</th>\n",
       "      <th>method</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>22.663506</td>\n",
       "      <td>18.839675</td>\n",
       "      <td>18.067965</td>\n",
       "      <td>21.428960</td>\n",
       "      <td>19.885973</td>\n",
       "      <td>27.186348</td>\n",
       "      <td>18.918224</td>\n",
       "      <td>20.998665</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Random_Forest</td>\n",
       "      <td>22.780908</td>\n",
       "      <td>18.938094</td>\n",
       "      <td>18.100895</td>\n",
       "      <td>21.430550</td>\n",
       "      <td>19.851967</td>\n",
       "      <td>27.337043</td>\n",
       "      <td>19.017243</td>\n",
       "      <td>21.065243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>22.794257</td>\n",
       "      <td>18.962714</td>\n",
       "      <td>18.124000</td>\n",
       "      <td>21.517626</td>\n",
       "      <td>19.815803</td>\n",
       "      <td>27.269566</td>\n",
       "      <td>19.032643</td>\n",
       "      <td>21.073801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>dummy</td>\n",
       "      <td>22.786039</td>\n",
       "      <td>18.935812</td>\n",
       "      <td>18.151719</td>\n",
       "      <td>21.499065</td>\n",
       "      <td>19.931326</td>\n",
       "      <td>27.322843</td>\n",
       "      <td>18.992690</td>\n",
       "      <td>21.088499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ARDR_Linear</td>\n",
       "      <td>22.786039</td>\n",
       "      <td>18.935812</td>\n",
       "      <td>18.151719</td>\n",
       "      <td>21.499065</td>\n",
       "      <td>19.931326</td>\n",
       "      <td>27.322843</td>\n",
       "      <td>18.992690</td>\n",
       "      <td>21.088499</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Lasso</td>\n",
       "      <td>22.771354</td>\n",
       "      <td>18.932383</td>\n",
       "      <td>18.146463</td>\n",
       "      <td>21.450772</td>\n",
       "      <td>19.917294</td>\n",
       "      <td>27.356153</td>\n",
       "      <td>19.049840</td>\n",
       "      <td>21.089180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ElasticNet</td>\n",
       "      <td>22.770732</td>\n",
       "      <td>18.930488</td>\n",
       "      <td>18.147500</td>\n",
       "      <td>21.449773</td>\n",
       "      <td>19.916494</td>\n",
       "      <td>27.364897</td>\n",
       "      <td>19.049554</td>\n",
       "      <td>21.089920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>XGBoost_Linear</td>\n",
       "      <td>22.759476</td>\n",
       "      <td>18.950054</td>\n",
       "      <td>18.217737</td>\n",
       "      <td>21.547576</td>\n",
       "      <td>19.950278</td>\n",
       "      <td>27.231096</td>\n",
       "      <td>18.983431</td>\n",
       "      <td>21.091378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Ridge</td>\n",
       "      <td>23.306065</td>\n",
       "      <td>19.060057</td>\n",
       "      <td>18.237283</td>\n",
       "      <td>22.232040</td>\n",
       "      <td>20.025811</td>\n",
       "      <td>27.817107</td>\n",
       "      <td>19.261822</td>\n",
       "      <td>21.420027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Linear</td>\n",
       "      <td>23.077136</td>\n",
       "      <td>19.098971</td>\n",
       "      <td>18.280346</td>\n",
       "      <td>22.425477</td>\n",
       "      <td>20.046425</td>\n",
       "      <td>27.875828</td>\n",
       "      <td>19.309978</td>\n",
       "      <td>21.444880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>39.818136</td>\n",
       "      <td>236.859405</td>\n",
       "      <td>18.229783</td>\n",
       "      <td>22.190174</td>\n",
       "      <td>19.976561</td>\n",
       "      <td>27.426506</td>\n",
       "      <td>352.461132</td>\n",
       "      <td>102.423099</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "iter          method          0           1          2          3          4  \\\n",
       "4               LGBM  22.663506   18.839675  18.067965  21.428960  19.885973   \n",
       "7      Random_Forest  22.780908   18.938094  18.100895  21.430550  19.851967   \n",
       "2       DecisionTree  22.794257   18.962714  18.124000  21.517626  19.815803   \n",
       "10             dummy  22.786039   18.935812  18.151719  21.499065  19.931326   \n",
       "0        ARDR_Linear  22.786039   18.935812  18.151719  21.499065  19.931326   \n",
       "5              Lasso  22.771354   18.932383  18.146463  21.450772  19.917294   \n",
       "3         ElasticNet  22.770732   18.930488  18.147500  21.449773  19.916494   \n",
       "9     XGBoost_Linear  22.759476   18.950054  18.217737  21.547576  19.950278   \n",
       "8              Ridge  23.306065   19.060057  18.237283  22.232040  20.025811   \n",
       "6             Linear  23.077136   19.098971  18.280346  22.425477  20.046425   \n",
       "1     Bayesian_Ridge  39.818136  236.859405  18.229783  22.190174  19.976561   \n",
       "\n",
       "iter          5           6        mean  \n",
       "4     27.186348   18.918224   20.998665  \n",
       "7     27.337043   19.017243   21.065243  \n",
       "2     27.269566   19.032643   21.073801  \n",
       "10    27.322843   18.992690   21.088499  \n",
       "0     27.322843   18.992690   21.088499  \n",
       "5     27.356153   19.049840   21.089180  \n",
       "3     27.364897   19.049554   21.089920  \n",
       "9     27.231096   18.983431   21.091378  \n",
       "8     27.817107   19.261822   21.420027  \n",
       "6     27.875828   19.309978   21.444880  \n",
       "1     27.426506  352.461132  102.423099  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_acc_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "5ac58ed2",
   "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>iter</th>\n",
       "      <th>method</th>\n",
       "      <th>mse</th>\n",
       "      <th>r2_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>22.794257</td>\n",
       "      <td>-0.001116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>18.962714</td>\n",
       "      <td>-0.001469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>18.124000</td>\n",
       "      <td>0.001106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>21.517626</td>\n",
       "      <td>-0.001251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>19.815803</td>\n",
       "      <td>0.005772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72</th>\n",
       "      <td>2</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>18.229783</td>\n",
       "      <td>-0.004724</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>3</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>22.190174</td>\n",
       "      <td>-0.032546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74</th>\n",
       "      <td>4</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>19.976561</td>\n",
       "      <td>-0.002293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>5</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>27.426506</td>\n",
       "      <td>-0.005556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>6</td>\n",
       "      <td>Bayesian_Ridge</td>\n",
       "      <td>352.461132</td>\n",
       "      <td>-17.558446</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>77 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    iter          method         mse   r2_score\n",
       "0      0    DecisionTree   22.794257  -0.001116\n",
       "1      1    DecisionTree   18.962714  -0.001469\n",
       "2      2    DecisionTree   18.124000   0.001106\n",
       "3      3    DecisionTree   21.517626  -0.001251\n",
       "4      4    DecisionTree   19.815803   0.005772\n",
       "..   ...             ...         ...        ...\n",
       "72     2  Bayesian_Ridge   18.229783  -0.004724\n",
       "73     3  Bayesian_Ridge   22.190174  -0.032546\n",
       "74     4  Bayesian_Ridge   19.976561  -0.002293\n",
       "75     5  Bayesian_Ridge   27.426506  -0.005556\n",
       "76     6  Bayesian_Ridge  352.461132 -17.558446\n",
       "\n",
       "[77 rows x 4 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_cla_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "ea8b8379",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('LGBM', 'dummy'),\n",
       " ('Random_Forest', 'dummy'),\n",
       " ('DecisionTree', 'dummy'),\n",
       " ('ARDR_Linear', 'dummy'),\n",
       " ('Lasso', 'dummy'),\n",
       " ('ElasticNet', 'dummy'),\n",
       " ('XGBoost_Linear', 'dummy'),\n",
       " ('Ridge', 'dummy'),\n",
       " ('Linear', 'dummy'),\n",
       " ('Bayesian_Ridge', 'dummy')]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[(df_acc_test.iloc[i,0], df_acc_test.iloc[dum_idx,0]) for i in range(len(df_acc_test)) if i != dum_idx]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "bcdbcdb9",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "15e1d72b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# colors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "947db660",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[10, 9, 8, 7, 6, 5, 4, 3, 2, 1]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(range(10,0,-1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "99ebf18e",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "p-value annotation legend:\n",
      "ns: 5.00e-02 < p <= 1.00e+00\n",
      "*: 1.00e-02 < p <= 5.00e-02\n",
      "**: 1.00e-03 < p <= 1.00e-02\n",
      "***: 1.00e-04 < p <= 1.00e-03\n",
      "****: p <= 1.00e-04\n",
      "\n",
      "Using zero_method  pratt\n",
      "dummy v.s. DecisionTree: Wilcoxon test (paired samples), P_val=6.875e-01 stat=1.100e+01\n",
      "Using zero_method  pratt\n",
      "ARDR_Linear v.s. dummy: Wilcoxon test (paired samples), P_val=5.393e-01 stat=9.000e+00\n",
      "Using zero_method  pratt\n",
      "dummy v.s. Random_Forest: Wilcoxon test (paired samples), P_val=3.750e-01 stat=8.000e+00\n",
      "Using zero_method  pratt\n",
      "Lasso v.s. dummy: Wilcoxon test (paired samples), P_val=8.125e-01 stat=1.200e+01\n",
      "Using zero_method  pratt\n",
      "dummy v.s. LGBM: Wilcoxon test (paired samples), P_val=1.562e-02 stat=0.000e+00\n",
      "Using zero_method  pratt\n",
      "ElasticNet v.s. dummy: Wilcoxon test (paired samples), P_val=8.125e-01 stat=1.200e+01\n",
      "Using zero_method  pratt\n",
      "XGBoost_Linear v.s. dummy: Wilcoxon test (paired samples), P_val=8.125e-01 stat=1.200e+01\n",
      "Using zero_method  pratt\n",
      "Ridge v.s. dummy: Wilcoxon test (paired samples), P_val=1.562e-02 stat=0.000e+00\n",
      "Using zero_method  pratt\n",
      "Linear v.s. dummy: Wilcoxon test (paired samples), P_val=1.562e-02 stat=0.000e+00\n",
      "Using zero_method  pratt\n",
      "Bayesian_Ridge v.s. dummy: Wilcoxon test (paired samples), P_val=1.562e-02 stat=0.000e+00\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x1200 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dum_idx = list(mse_df.iloc[:,0].values).index('dummy')\n",
    "# sns.set_palette(color_palette('husl'))\n",
    "\n",
    "# 파랑 \n",
    "# cubehelix_palette = sns.cubehelix_palette(start = .5, rot = -.5, as_cmap = True)\n",
    "# pal = sns.cubehelix_palette(n_colors=len(df_acc_test), start = .5, rot = -.5)\n",
    "# https://greeksharifa.github.io/machine_learning/2019/12/05/Seaborn-Module/\n",
    "\n",
    "# 주황 \"YlOrBr\"\n",
    "# colors = ['#fff7bd']*(len(df_acc_test)-1) + ['#f48000']\n",
    "# pal = sns.set_palette(sns.color_palette(colors))\n",
    "\n",
    "_ = plt.figure(figsize=(15,12), dpi = 100)\n",
    "_ = ax = sns.barplot(x = 'method', y = 'mse', data = df_cla_results, ci = None, capsize = None,\n",
    "                order=df_cla_results.groupby('method').mean().sort_values('mse', ascending = False).index,\n",
    "                palette = \"YlOrBr\") # palette = \"YlOrBr\", 'husl', pal\n",
    "\n",
    "\n",
    "_ = ax.set(ylim = (20.95, 21.6))\n",
    "_ = ax.set_xticklabels(ax.get_xticklabels(), rotation = 35, fontsize = 20)\n",
    "_ = plt.setp(ax.get_yticklabels(), fontsize =18)\n",
    "_ = ax.set_xlabel(\"\", fontsize = 28)\n",
    "_ = ax.set_ylabel(\"\", fontsize = 25)\n",
    "\n",
    "# dummy_mean = acc_df[acc_df.key=='dummy']['value'].values\n",
    "# _ = ax.plot([-1,len(acc_df)], [dummy_mean, dummy_mean], color = 'red', linewidth = 2.3)\n",
    "# _ = ax.annotate(text = f'dummy\\n: {dummy_mean[0]:.4f}', xy = (2, dummy_mean), xytext = (0, 0.53),\n",
    "#                arrowprops = dict(facecolor = 'red', shrink = 0), fontsize = 18)\n",
    "\n",
    "\n",
    "_ = test_results = add_stat_annotation(ax, data = df_cla_results, x = 'method', y = 'mse',\n",
    "                                   order=df_cla_results.groupby('method').mean().sort_values('mse', ascending = False).index,\n",
    "                                   box_pairs = [(df_acc_test.iloc[i,0], df_acc_test.iloc[dum_idx,0]) for i in range(len(df_acc_test)) if i != dum_idx],\n",
    "                                   test = 'Wilcoxon', text_format='star', comparisons_correction = None,\n",
    "                                   loc = 'outside', verbose = 2, linewidth=2, fontsize=15)\n",
    "### 기타 조정\n",
    "val = mse_df['value'].sort_values(ascending = False).reset_index(drop = True)\n",
    "cnt = 0\n",
    "for i in range(len(mse_df)):\n",
    "    tmp_fval = f'{val[i]:.3f}'\n",
    "    if tmp_fval.index('.') == 3:\n",
    "        _ = ax.text(cnt-0.32, 20.965, tmp_fval[:-1], fontsize=15)\n",
    "    else:\n",
    "        _ = ax.text(cnt-0.32, 20.965, tmp_fval, fontsize=15)\n",
    "    cnt += 1\n",
    "_ = plt.tight_layout()\n",
    "plt.savefig('[15w] -4 mse_reg_pvalue_plot.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "25d1103d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     102.423099\n",
       "1      21.444880\n",
       "2      21.420027\n",
       "3      21.091378\n",
       "4      21.089920\n",
       "5      21.089180\n",
       "6      21.088499\n",
       "7      21.088499\n",
       "8      21.073801\n",
       "9      21.065243\n",
       "10     20.998665\n",
       "Name: value, dtype: float64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mse_df['value'].sort_values(ascending = False).reset_index(drop = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2b8c628",
   "metadata": {},
   "source": [
    "# cla results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5d8a726a",
   "metadata": {},
   "source": [
    "## 10thr\n",
    "cla_results = [[0, 'XGBoost', 0.5533303429845786, 0.5274938602009643],\n",
    " [1, 'XGBoost', 0.5587648512640324, 0.528969220742626],\n",
    " [2, 'XGBoost', 0.5608311819609684, 0.5267130683977462],\n",
    " [3, 'XGBoost', 0.5525027920342322, 0.5209610157365271],\n",
    " [4, 'XGBoost', 0.5596736996872831, 0.52162189127671],\n",
    " [5, 'XGBoost', 0.5382265473507089, 0.51353989896573],\n",
    " [6, 'XGBoost', 0.5546100115373264, 0.5134707222923052],\n",
    " [7, 'dummy', 0.5022901438526222, 0.498624936882785],\n",
    " [8, 'dummy', 0.5004764315278564, 0.4967955180971402],\n",
    " [9, 'dummy', 0.5074560849235435, 0.5029978207328746],\n",
    " [10, 'dummy', 0.505364470266404, 0.5009225744971062],\n",
    " [11, 'dummy', 0.5049818265951175, 0.4999652479913408],\n",
    " [12, 'dummy', 0.5033045841262985, 0.5000358456090245],\n",
    " [13, 'dummy', 0.5045702122331435, 0.5002466642274119],\n",
    " [14, 'Logistic_Regression', 0.5388124103685078, 0.5006451548065087],\n",
    " [15, 'Logistic_Regression', 0.5422460768841378, 0.4992190478424038],\n",
    " [16, 'Logistic_Regression', 0.5529598671264285, 0.49972436498512524],\n",
    " [17, 'Logistic_Regression', 0.5486315895144998, 0.4997338524721153],\n",
    " [18, 'Logistic_Regression', 0.5520610208453058, 0.5000000566565985],\n",
    " [19, 'Logistic_Regression', 0.5322787458327147, 0.49968927327232476],\n",
    " [20, 'Logistic_Regression', 0.5521326166477359, 0.4998208245751821],\n",
    " [21, 'SVM', 0.45822461730906516, 0.4999856282727155],\n",
    " [22, 'SVM', 0.45512610547002946, 0.500040258196378],\n",
    " [23, 'SVM', 0.4466339296997707, 0.5000869632262043],\n",
    " [24, 'SVM', 0.45129311967474367, 0.500225094814467],\n",
    " [25, 'SVM', 0.44796136938658226, 0.5001228231705257],\n",
    " [26, 'SVM', 0.46762677320519913, 0.5002860894098915],\n",
    " [27, 'SVM', 0.44784502419261074, 0.5001912401255232],\n",
    " [28, 'Gaussian_NB', 0.4585847226872706, 0.5001861170154632],\n",
    " [29, 'Gaussian_NB', 0.45533454426346665, 0.4999844005243526],\n",
    " [30, 'Gaussian_NB', 0.44699499918759367, 0.5000782161269514],\n",
    " [31, 'Gaussian_NB', 0.4513872331881894, 0.5001233864776655],\n",
    " [32, 'Gaussian_NB', 0.44796883279721167, 0.5001437873510062],\n",
    " [33, 'Gaussian_NB', 0.467739250541016, 0.5003228724449605],\n",
    " [34, 'Gaussian_NB', 0.44788974251191743, 0.500238357284859],\n",
    " [35, 'MLP', 0.5070852312571468, 0.4981315384745352],\n",
    " [36, 'MLP', 0.5270523776910937, 0.4990287907340827],\n",
    " [37, 'MLP', 0.5287501579679009, 0.5006602983405445],\n",
    " [38, 'MLP', 0.5422193221317338, 0.5009435972809749],\n",
    " [39, 'MLP', 0.5345817131512759, 0.49794460684853803],\n",
    " [40, 'MLP', 0.5199872225746512, 0.5030644735501303],\n",
    " [41, 'MLP', 0.5424779315094221, 0.5025564917597585],\n",
    " [42, 'DecisionTree', 0.5477202802504312, 0.5177220448987767],\n",
    " [43, 'DecisionTree', 0.5436307060119704, 0.5164708209157598],\n",
    " [44, 'DecisionTree', 0.5499088299543247, 0.5111833363647642],\n",
    " [45, 'DecisionTree', 0.5475963408665973, 0.5227744579502975],\n",
    " [46, 'DecisionTree', 0.5525536059468455, 0.5023766400076397],\n",
    " [47, 'DecisionTree', 0.5365933764346484, 0.5119752568734229],\n",
    " [48, 'DecisionTree', 0.5529464900591177, 0.5114457071075491],\n",
    " [49, 'RandomForest', 0.5455091068754857, 0.5059471462013954],\n",
    " [50, 'RandomForest', 0.5504347437691689, 0.5089746337276014],\n",
    " [51, 'RandomForest', 0.5559567438753588, 0.5083075846348538],\n",
    " [52, 'RandomForest', 0.5557904907706015, 0.5117567260547392],\n",
    " [53, 'RandomForest', 0.5540089710195766, 0.5052001863636698],\n",
    " [54, 'RandomForest', 0.5336059783953533, 0.5023594017403971],\n",
    " [55, 'RandomForest', 0.5510280741608607, 0.4998035303951165],\n",
    " [56, 'GradientBoosting', 0.541781700329149, 0.5],\n",
    " [57, 'GradientBoosting', 0.544978113926689, 0.5],\n",
    " [58, 'GradientBoosting', 0.5536097922045098, 0.5],\n",
    " [59, 'GradientBoosting', 0.5492903841086195, 0.5],\n",
    " [60, 'GradientBoosting', 0.5522998499854463, 0.5],\n",
    " [61, 'GradientBoosting', 0.5328006406709048, 0.5],\n",
    " [62, 'GradientBoosting', 0.5519984616898158, 0.4999027123208462]]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ef5a99a",
   "metadata": {},
   "source": [
    "## plot을 위한 준비"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "3d396d29",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('SVM', 0.4535301341340001),\n",
       " ('Gaussian_NB', 0.4536999035966664),\n",
       " ('dummy', 0.5040633933607123),\n",
       " ('MLP', 0.5288791366118893),\n",
       " ('Logistic_Regression', 0.5455889038884757),\n",
       " ('GradientBoosting', 0.5466798489878764),\n",
       " ('DecisionTree', 0.5472785185034194),\n",
       " ('RandomForest', 0.5494763012666294),\n",
       " ('XGBoost', 0.5539913466884472)]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "[('Logistic_Regression', 0.4998332249443226),\n",
       " ('dummy', 0.4999412297196691),\n",
       " ('GradientBoosting', 0.4999861017601209),\n",
       " ('SVM', 0.5001340138879579),\n",
       " ('Gaussian_NB', 0.5001538767464655),\n",
       " ('MLP', 0.5003328281412235),\n",
       " ('RandomForest', 0.5060498870168246),\n",
       " ('DecisionTree', 0.5134211805883158),\n",
       " ('XGBoost', 0.5218242396589441)]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_list = ['Logistic_Regression', 'SVM', 'Gaussian_NB',\n",
    "              'MLP',\n",
    "              'DecisionTree', 'RandomForest', 'GradientBoosting',\n",
    "              'XGBoost', 'dummy'] #  'LGBM', 'KNeighbors',\n",
    "acc_dic = {i:[] for i in model_list}\n",
    "rocauc_dic = {i:[] for i in model_list}\n",
    "f1_dic = {i:[] for i in model_list}\n",
    "jaccard_dic = {i:[] for i in model_list}\n",
    "dum_dic = {i:[] for i in model_list}\n",
    "for i in cla_results:\n",
    "    acc_dic[i[1]].append(i[2])\n",
    "    rocauc_dic[i[1]].append(i[3])\n",
    "\n",
    "\n",
    "# import math\n",
    "# 리스트에 mean 함수가 내장이 안돼있다니...\n",
    "mean_acc_score = {i:[] for i in model_list}\n",
    "mean_rocauc_score = {i:[] for i in model_list}\n",
    "for i in model_list:\n",
    "    mean_acc_score[i] = sum(acc_dic[i])/len(acc_dic[i])\n",
    "    mean_rocauc_score[i] = sum(rocauc_dic[i])/len(rocauc_dic[i])\n",
    "\n",
    "# 정렬\n",
    "acc_score_lst = sorted(mean_acc_score.items(), key = lambda t : t[1]) #acc 작은 순으로 줄 세워짐.\n",
    "rocauc_score_lst = sorted(mean_rocauc_score.items(), key = lambda t : t[1]) #rocauc 작은 순으로 줄 세워짐.\n",
    "\n",
    "acc_score_lst\n",
    "rocauc_score_lst"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "9d3239a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 다시 dic\n",
    "acc_score_dic = {}\n",
    "rocauc_score_dic = {}\n",
    "\n",
    "for i in acc_score_lst:\n",
    "    acc_score_dic[i[0]] = i[1]\n",
    "rocauc_score_dic = {i[0]:i[1] for i in rocauc_score_lst}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "a01cf4eb",
   "metadata": {
    "scrolled": true
   },
   "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>key</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SVM</td>\n",
       "      <td>0.453530</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Gaussian_NB</td>\n",
       "      <td>0.453700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>dummy</td>\n",
       "      <td>0.504063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>MLP</td>\n",
       "      <td>0.528879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Logistic_Regression</td>\n",
       "      <td>0.545589</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>GradientBoosting</td>\n",
       "      <td>0.546680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>0.547279</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>RandomForest</td>\n",
       "      <td>0.549476</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>XGBoost</td>\n",
       "      <td>0.553991</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   key     value\n",
       "0                  SVM  0.453530\n",
       "1          Gaussian_NB  0.453700\n",
       "2                dummy  0.504063\n",
       "3                  MLP  0.528879\n",
       "4  Logistic_Regression  0.545589\n",
       "5     GradientBoosting  0.546680\n",
       "6         DecisionTree  0.547279\n",
       "7         RandomForest  0.549476\n",
       "8              XGBoost  0.553991"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>key</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Logistic_Regression</td>\n",
       "      <td>0.499833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>dummy</td>\n",
       "      <td>0.499941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GradientBoosting</td>\n",
       "      <td>0.499986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SVM</td>\n",
       "      <td>0.500134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Gaussian_NB</td>\n",
       "      <td>0.500154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>MLP</td>\n",
       "      <td>0.500333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>RandomForest</td>\n",
       "      <td>0.506050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>0.513421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>XGBoost</td>\n",
       "      <td>0.521824</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   key     value\n",
       "0  Logistic_Regression  0.499833\n",
       "1                dummy  0.499941\n",
       "2     GradientBoosting  0.499986\n",
       "3                  SVM  0.500134\n",
       "4          Gaussian_NB  0.500154\n",
       "5                  MLP  0.500333\n",
       "6         RandomForest  0.506050\n",
       "7         DecisionTree  0.513421\n",
       "8              XGBoost  0.521824"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "acc_df = pd.DataFrame(acc_score_dic.items(), columns=['key', 'value'])\n",
    "acc_df\n",
    "\n",
    "rocauc_df = pd.DataFrame(rocauc_score_dic.items(), columns=['key', 'value'])\n",
    "rocauc_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "692e3709",
   "metadata": {},
   "source": [
    "## 테스트 데이터셋 정확도 플랏 (유의성 기호 추가)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "23f24e08",
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install statannot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "922db03b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "from statannot import add_stat_annotation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "9a17bf3a",
   "metadata": {},
   "outputs": [],
   "source": [
    "dum_idx = list(rocauc_df.iloc[:,0].values).index('dummy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "ef91a2a1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Gaussian_NB'"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dummy랑 비교하라고 하셨던 것 같음.\n",
    "acc_df.iloc[dum_idx,0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "5dbd7a01",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df_cla_results = pd.DataFrame(cla_results,\n",
    "             columns= ['iter', 'method', 'accuracy_test', 'roc_auc'])\n",
    "df_cla_results['iter'] = [i for i in range(7)]*9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "8bb73ba8",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_acc_test = df_cla_results.pivot_table(index='method', columns='iter')[\"roc_auc\"].reset_index()\n",
    "# df_acc_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "bad1a421",
   "metadata": {
    "scrolled": true
   },
   "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>iter</th>\n",
       "      <th>method</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Logistic_Regression</td>\n",
       "      <td>0.500645</td>\n",
       "      <td>0.499219</td>\n",
       "      <td>0.499724</td>\n",
       "      <td>0.499734</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.499689</td>\n",
       "      <td>0.499821</td>\n",
       "      <td>0.499833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>dummy</td>\n",
       "      <td>0.498625</td>\n",
       "      <td>0.496796</td>\n",
       "      <td>0.502998</td>\n",
       "      <td>0.500923</td>\n",
       "      <td>0.499965</td>\n",
       "      <td>0.500036</td>\n",
       "      <td>0.500247</td>\n",
       "      <td>0.499941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GradientBoosting</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.499903</td>\n",
       "      <td>0.499986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>SVM</td>\n",
       "      <td>0.499986</td>\n",
       "      <td>0.500040</td>\n",
       "      <td>0.500087</td>\n",
       "      <td>0.500225</td>\n",
       "      <td>0.500123</td>\n",
       "      <td>0.500286</td>\n",
       "      <td>0.500191</td>\n",
       "      <td>0.500134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Gaussian_NB</td>\n",
       "      <td>0.500186</td>\n",
       "      <td>0.499984</td>\n",
       "      <td>0.500078</td>\n",
       "      <td>0.500123</td>\n",
       "      <td>0.500144</td>\n",
       "      <td>0.500323</td>\n",
       "      <td>0.500238</td>\n",
       "      <td>0.500154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>MLP</td>\n",
       "      <td>0.498132</td>\n",
       "      <td>0.499029</td>\n",
       "      <td>0.500660</td>\n",
       "      <td>0.500944</td>\n",
       "      <td>0.497945</td>\n",
       "      <td>0.503064</td>\n",
       "      <td>0.502556</td>\n",
       "      <td>0.500333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>RandomForest</td>\n",
       "      <td>0.505947</td>\n",
       "      <td>0.508975</td>\n",
       "      <td>0.508308</td>\n",
       "      <td>0.511757</td>\n",
       "      <td>0.505200</td>\n",
       "      <td>0.502359</td>\n",
       "      <td>0.499804</td>\n",
       "      <td>0.506050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DecisionTree</td>\n",
       "      <td>0.517722</td>\n",
       "      <td>0.516471</td>\n",
       "      <td>0.511183</td>\n",
       "      <td>0.522774</td>\n",
       "      <td>0.502377</td>\n",
       "      <td>0.511975</td>\n",
       "      <td>0.511446</td>\n",
       "      <td>0.513421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>XGBoost</td>\n",
       "      <td>0.527494</td>\n",
       "      <td>0.528969</td>\n",
       "      <td>0.526713</td>\n",
       "      <td>0.520961</td>\n",
       "      <td>0.521622</td>\n",
       "      <td>0.513540</td>\n",
       "      <td>0.513471</td>\n",
       "      <td>0.521824</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "iter               method         0         1         2         3         4  \\\n",
       "3     Logistic_Regression  0.500645  0.499219  0.499724  0.499734  0.500000   \n",
       "8                   dummy  0.498625  0.496796  0.502998  0.500923  0.499965   \n",
       "2        GradientBoosting  0.500000  0.500000  0.500000  0.500000  0.500000   \n",
       "6                     SVM  0.499986  0.500040  0.500087  0.500225  0.500123   \n",
       "1             Gaussian_NB  0.500186  0.499984  0.500078  0.500123  0.500144   \n",
       "4                     MLP  0.498132  0.499029  0.500660  0.500944  0.497945   \n",
       "5            RandomForest  0.505947  0.508975  0.508308  0.511757  0.505200   \n",
       "0            DecisionTree  0.517722  0.516471  0.511183  0.522774  0.502377   \n",
       "7                 XGBoost  0.527494  0.528969  0.526713  0.520961  0.521622   \n",
       "\n",
       "iter         5         6      mean  \n",
       "3     0.499689  0.499821  0.499833  \n",
       "8     0.500036  0.500247  0.499941  \n",
       "2     0.500000  0.499903  0.499986  \n",
       "6     0.500286  0.500191  0.500134  \n",
       "1     0.500323  0.500238  0.500154  \n",
       "4     0.503064  0.502556  0.500333  \n",
       "5     0.502359  0.499804  0.506050  \n",
       "0     0.511975  0.511446  0.513421  \n",
       "7     0.513540  0.513471  0.521824  "
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_acc_test['mean']=np.mean(df_acc_test, axis=1)\n",
    "df_acc_test=df_acc_test.sort_values('mean')\n",
    "df_acc_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "13f761f4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logistic_Regression vs. dummy WilcoxonResult(statistic=12.0, pvalue=0.8125)\n",
      "GradientBoosting vs. dummy WilcoxonResult(statistic=13.0, pvalue=0.9375)\n",
      "SVM vs. dummy WilcoxonResult(statistic=11.0, pvalue=0.6875)\n",
      "Gaussian_NB vs. dummy WilcoxonResult(statistic=11.0, pvalue=0.6875)\n",
      "MLP vs. dummy WilcoxonResult(statistic=11.0, pvalue=0.6875)\n",
      "RandomForest vs. dummy WilcoxonResult(statistic=1.0, pvalue=0.03125)\n",
      "DecisionTree vs. dummy WilcoxonResult(statistic=0.0, pvalue=0.015625)\n",
      "XGBoost vs. dummy WilcoxonResult(statistic=0.0, pvalue=0.015625)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from scipy import stats\n",
    "\n",
    "# 비모수적 검정(parametric test)\n",
    "for i in range(9):\n",
    "    if i == dum_idx:\n",
    "        continue\n",
    "    print(df_acc_test.iloc[i,0], 'vs.', df_acc_test.iloc[dum_idx,0],\n",
    "          stats.wilcoxon(df_acc_test.iloc[i,1:-1], df_acc_test.iloc[dum_idx,1:-1])) # Wilcoxon signed-rank test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "f2d8c944",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('Logistic_Regression', 'dummy'),\n",
       " ('GradientBoosting', 'dummy'),\n",
       " ('SVM', 'dummy'),\n",
       " ('Gaussian_NB', 'dummy'),\n",
       " ('MLP', 'dummy'),\n",
       " ('RandomForest', 'dummy'),\n",
       " ('DecisionTree', 'dummy'),\n",
       " ('XGBoost', 'dummy')]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[(df_acc_test.iloc[i,0], df_acc_test.iloc[dum_idx,0]) for i in range(len(df_acc_test)) if i != dum_idx]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "29876ec2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "13f9bf97",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "UsageError: %colors: you must specify a color scheme. See '%colors?'\n"
     ]
    }
   ],
   "source": [
    "colors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "e167fdce",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using zero_method  pratt\n",
      "Using zero_method  pratt\n",
      "Using zero_method  pratt\n",
      "Using zero_method  pratt\n",
      "Using zero_method  pratt\n",
      "Using zero_method  pratt\n",
      "Using zero_method  pratt\n",
      "Using zero_method  pratt\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x1200 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dum_idx = list(rocauc_df.iloc[:,0].values).index('dummy')\n",
    "# sns.set_palette(color_palette('husl'))\n",
    "\n",
    "# 파랑 \n",
    "# cubehelix_palette = sns.cubehelix_palette(start = .5, rot = -.5, as_cmap = True)\n",
    "# pal = sns.cubehelix_palette(n_colors=len(df_acc_test), start = .5, rot = -.5)\n",
    "# https://greeksharifa.github.io/machine_learning/2019/12/05/Seaborn-Module/\n",
    "\n",
    "# 주황 \"YlOrBr\"\n",
    "# colors = ['#fff7bd']*(len(df_acc_test)-1) + ['#f48000']\n",
    "# pal = sns.set_palette(sns.color_palette(colors))\n",
    "\n",
    "_ = plt.figure(figsize=(15,12), dpi = 100)\n",
    "_ = ax = sns.barplot(x = 'method', y = 'roc_auc', data = df_cla_results, capsize = .2, ci=None,\n",
    "                order=df_cla_results.groupby('method').mean().sort_values('roc_auc').index,\n",
    "                palette = \"YlOrBr\") # palette = \"YlOrBr\", 'husl', pal\n",
    "\n",
    "\n",
    "_ = ax.set(ylim = (0.49, 0.525))\n",
    "_ = ax.set_xticklabels(ax.get_xticklabels(), rotation = 35, fontsize = 20)\n",
    "_ = plt.setp(ax.get_yticklabels(), fontsize =18)\n",
    "_ = ax.set_xlabel(\"\", fontsize = 28)\n",
    "_ = ax.set_ylabel(\"\", fontsize = 25)\n",
    "\n",
    "# dummy_mean = acc_df[acc_df.key=='dummy']['value'].values\n",
    "# _ = ax.plot([-1,len(acc_df)], [dummy_mean, dummy_mean], color = 'red', linewidth = 2.3)\n",
    "# _ = ax.annotate(text = f'dummy\\n: {dummy_mean[0]:.4f}', xy = (2, dummy_mean), xytext = (0, 0.53),\n",
    "#                arrowprops = dict(facecolor = 'red', shrink = 0), fontsize = 18)\n",
    "\n",
    "\n",
    "_ = test_results = add_stat_annotation(ax, data = df_cla_results, x = 'method', y = 'roc_auc',\n",
    "                                   order=df_cla_results.groupby('method').mean().sort_values('roc_auc').index,\n",
    "                                   box_pairs = [(df_acc_test.iloc[i,0], df_acc_test.iloc[dum_idx,0]) for i in range(len(df_acc_test)) if i != dum_idx],\n",
    "                                   test = 'Wilcoxon', text_format='star', comparisons_correction = None,\n",
    "                                   loc = 'outside', verbose = 0, linewidth=2, fontsize=15)\n",
    "### 기타 조정\n",
    "val = acc_df['value']\n",
    "cnt = 0\n",
    "for i in range(len(acc_df)):\n",
    "    tmp_fval = f'{val[i]:.3f}'\n",
    "#     if (i == len(acc_df)-2) | (i == len(acc_df)-1):\n",
    "#         _ = ax.text(cnt-0.155, 0.432, tmp_fval, fontsize=15, color = 'white')\n",
    "#     else:\n",
    "    _ = ax.text(cnt-0.195, 0.4905, tmp_fval, fontsize=15)\n",
    "    cnt += 1\n",
    "_ = plt.tight_layout()\n",
    "plt.savefig('[15w] -4 5per_roc_auc_cla_pvalue_plot.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d44c4ba8",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.2"
  },
  "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": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}