├── questionare.xlsx
├── eegdata_72var.xlsx
├── README.md
├── main.py
├── extract_data.py
├── basic_model_kfold.ipynb
├── mymontage.xml
└── permutation_importance.ipynb


/questionare.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/manfred1201/EEGprocessing/HEAD/questionare.xlsx


--------------------------------------------------------------------------------
/eegdata_72var.xlsx:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/manfred1201/EEGprocessing/HEAD/eegdata_72var.xlsx


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # EEGprocessing
2 | 本文件为读取EEG数据文件并进行预处理，以及机器学习建模的代码文档
3 | 视频说明详见：https://www.bilibili.com/video/BV18q4y1m7C5?spm_id_from=333.999.0.0
4 | 
5 | 原始数据已上传至百度网盘：链接：https://pan.baidu.com/s/1zwNA6lMvgJbSZZMTIwtCfw 提取码：9e87 
6 | 


--------------------------------------------------------------------------------
/main.py:
--------------------------------------------------------------------------------
  1 | #!/usr/bin/env python
  2 | # coding: utf-8
  3 | import os
  4 | import numpy as np
  5 | import pandas as pd
  6 | import mne
  7 | import matplotlib.pyplot as plt
  8 | import xml.etree.ElementTree
  9 | from xml.etree.ElementTree import parse
 10 | xml = parse('rawdata\mymontage.xml')
 11 | root = xml.getroot()
 12 | mydic = dict()
 13 | for channel in root.findall('CHANNEL'):
 14 |     label = channel.find('LABEL')
 15 |     pos = channel.find('POSITION')
 16 |     if pos is None:
 17 |         continue
 18 |     lis = pos.text.split()
 19 |     a = np.array(lis).astype(np.float64) / 1000
 20 |     a[0] = -a[0]
 21 |     a[1] = -a[1]
 22 |     mydic[label.text] = a
 23 | mymontage = mne.channels.make_dig_montage(ch_pos=mydic,coord_frame='head')
 24 | 
 25 | 
 26 | #导入数据并剔除无用channels
 27 | raw = mne.io.read_raw_eeglab('rawdata/eeg/f008.set')
 28 | raw.set_channel_types({'VEOG':'eog'})
 29 | raw.pick(picks='all',exclude=['HEOG','EKG','EMG','Trigger'])
 30 | 
 31 | #电极定位
 32 | raw.set_montage(mymontage,on_missing='warn')
 33 | 
 34 | #提取社会奖赏实验的片段
 35 | events = mne.events_from_annotations(raw)
 36 | ind = np.where(events[0][:,2] == events[1]['14'])[0].min()
 37 | start_t = events[0][ind,0] - 5000
 38 | ind = np.where(events[0][:,2] == events[1]['18'])[0].max()
 39 | end_t = events[0][ind,0] + 8500
 40 | raw_cropped = raw.copy()
 41 | raw_cropped.crop(tmin=start_t/1000,tmax=end_t/1000)
 42 | 
 43 | 
 44 | #插值坏导
 45 | raw_cropped.plot(start=20,duration=1,n_channels=33,block=True,title='请检查并选中坏导')   #定义坏导
 46 | plt.show()
 47 | badflag = False
 48 | if raw_cropped.info['bads']:
 49 |     print('已选择坏导: ',raw_cropped.info['bads'], '开始进行插值')
 50 |     badflag = True
 51 | else:
 52 |     print('无坏导，跳过插值')
 53 | if badflag:
 54 |     raw_cropped.load_data()
 55 |     raw_cropped.interpolate_bads(exclude=['F11','F12','FT11','FT12'])
 56 |     raw_cropped.plot(start=20,duration=1,n_channels=33,block=True,title='坏导插值完成，如无误请关闭窗口')
 57 |     plt.show()
 58 | 
 59 | 
 60 | 
 61 | #重参考
 62 | raw_ref = raw_cropped.copy()
 63 | raw_ref.load_data()
 64 | raw_ref.set_eeg_reference(ref_channels=['M1','M2'])
 65 | raw_ref.plot(start=20,duration=1,block=True,title='重参考完成，无误请关闭窗口')
 66 | plt.show()
 67 | 
 68 | 
 69 | #滤波
 70 | raw_filter = raw_ref.copy()
 71 | raw_filter.filter(l_freq=1,h_freq=30)
 72 | raw_filter.notch_filter(freqs=50)
 73 | raw_filter.plot_psd(fmax=60)
 74 | plt.show(block=False)
 75 | raw_filter.plot(start=20,duration=1,block=True,title='滤波完成，准备ICA，无误请关闭窗口')
 76 | 
 77 | #ICA
 78 | ica = mne.preprocessing.ICA(n_components=10, method='picard', max_iter=800)
 79 | ica.fit(raw_filter)
 80 | raw_filter.load_data()
 81 | ica.plot_components()
 82 | ica.plot_sources(raw_filter, show_scrollbars=False, title='请选择需要去除的成分')
 83 | plt.show(block=True)
 84 | print(ica)
 85 | raw_recons = raw_filter.copy()
 86 | raw_recons = ica.apply(raw_recons)
 87 | raw_filter.plot(start=20,duration=1,n_channels=33,title='ICA处理前, 确认请关闭')
 88 | raw_recons.plot(start=20,duration=1,n_channels=33,title='ICA处理后, 确认请关闭')
 89 | plt.show(block=True)
 90 | 
 91 | #提取epochs
 92 | events = mne.events_from_annotations(raw_recons)
 93 | event_dic = {'pos' : events[1]['20'], 'neg' : events[1]['22']}
 94 | reject_criteria = dict(eeg=100e-6)  # 100 µV
 95 | epochs = mne.Epochs(raw_recons, events[0], event_id=event_dic, preload=True, tmax=1, tmin=-0.2, reject=reject_criteria)
 96 | epochs.plot(events=events[0],block=True,title='请目视挑选出坏EPOCHES')
 97 | plt.show()
 98 | 
 99 | #对两个condition的epochs做平均
100 | #epochs.equalize_event_counts(['pos','neg']) 是否需要将两个condition一一对应？？
101 | pos_epochs = epochs['pos']
102 | neg_epochs = epochs['neg']
103 | 
104 | pos_evoked = pos_epochs.average()
105 | neg_evoked = neg_epochs.average()
106 | neg_evoked.plot()
107 | plt.show(block=True)
108 | pos_evoked.save('pos_ave.fif')
109 | neg_evoked.save('neg_ave.fif')
110 | #最终可视化差异
111 | mne.viz.plot_compare_evokeds(dict(positive=pos_evoked,negative=neg_evoked),combine='mean',picks = np.arange(32), invert_y=True, legend=True, ci=True)
112 | plt.show()


--------------------------------------------------------------------------------
/extract_data.py:
--------------------------------------------------------------------------------
  1 | import os
  2 | import numpy as np
  3 | import pandas as pd
  4 | import mne
  5 | import time
  6 | from mne.viz.utils import _convert_psds
  7 | import matplotlib.pyplot as plt
  8 | import xml.etree.ElementTree
  9 | from xml.etree.ElementTree import parse
 10 | 
 11 | def find_close(arr, e):
 12 |     low = 0
 13 |     high = len(arr) - 1
 14 |     idx = -1
 15 | 
 16 |     while low <= high:
 17 |         mid = int((low + high) / 2)
 18 |         if e == arr[mid] or mid == low:
 19 |             idx = mid
 20 |             break
 21 |         elif e > arr[mid]:
 22 |             low = mid
 23 |         elif e < arr[mid]:
 24 |             high = mid
 25 |     return idx, idx+1
 26 | 
 27 | def auc(x,y,x1,x2):
 28 |     if x1 not in x:
 29 |         bd1, bd2 = find_close(x, x1)
 30 |         x1y = (y[bd2]-y[bd1])/(x[bd2]-x[bd1])*(x1-x[bd1])+y[bd1]
 31 |         x = np.insert(x,bd2,x1)
 32 |         y = np.insert(y,bd2,x1y)
 33 |     if x2 not in x:
 34 |         bd1, bd2 = find_close(x, x2)
 35 |         x2y = (y[bd2]-y[bd1])/(x[bd2]-x[bd1])*(x2-x[bd1])+y[bd1]
 36 |         x = np.insert(x,bd2,x2)
 37 |         y = np.insert(y,bd2,x2y)
 38 |     idx1 = int(np.where(x==x1)[0])
 39 |     idx2 = int(np.where(x==x2)[0])
 40 |     ans = np.trapz(y=y[idx1:idx2+1],x=x[idx1:idx2+1])
 41 |     return ans
 42 | 
 43 | picks_front = ['Fp1','Fp2','Fz','F3','F4','F7','F8','F11','F12']
 44 | picks_top = ['Cz','FC3','FCz','FC4','C3','C4','CP3','CPz','CP4']
 45 | picks_left = ['FT11','T7','M1','C3','FC3','CP3']
 46 | picks_right = ['FT12','T8','M2','C4','FC4','CP4']
 47 | picks_back = ['Oz','O1','O2','Pz','P3','P4','P7','P8']
 48 | picks_list = list([picks_front, picks_top, picks_left, picks_right, picks_back])
 49 | #被试数量
 50 | path = 'data\\'
 51 | filename = os.listdir(path)
 52 | filelist=[]
 53 | for i in filename:
 54 |     if i[-3:] == 'fif':
 55 |         filelist.append(os.path.join(path,i))
 56 | print('总计被试数量: ', len(filelist))
 57 | all_data = np.empty((1,62),dtype='<U32')
 58 | 
 59 | for sub in filelist:
 60 |     epochs = mne.read_epochs(sub)
 61 |     epochs = epochs.drop_channels(['VEOG'])
 62 |     epochs = epochs.apply_baseline(baseline=(None,0))
 63 |     epochs_res = epochs.resample(200)
 64 |     epochs_res = epochs_res.filter(l_freq=None,h_freq=30)
 65 | 
 66 |     #试次数量
 67 |     arr = epochs_res.get_data()
 68 |     trials = arr.shape[0]
 69 |     epo_data = np.empty((1,62),dtype='<U32')
 70 |     for i in range(trials):
 71 |         condition = list(epochs_res[i].event_id.keys())[0]
 72 |         reslist = list()
 73 |         result = np.empty((1,62),dtype='<U32')
 74 |         j = 0
 75 |         for pick in picks_list:
 76 |             data = epochs_res[i].get_data(units='uV', picks=pick)
 77 |             data = np.squeeze(data)
 78 |             data = np.average(data,axis=0)
 79 |             psd, freqs = mne.time_frequency.psd_multitaper(epochs[i], picks=pick, fmin=0 , fmax=100, proj=False, n_jobs=1, bandwidth=None, adaptive=False, low_bias=True, normalization='length')
 80 |             psd = 10 * np.log10(np.mean(np.mean(psd, axis=0),axis=0) * 1e12)
 81 |             #计算不同频段的PSD
 82 |             #Delta波1-3Hz
 83 |             Delta = auc(freqs,psd,1,3)
 84 |             #Theta波4-7Hz
 85 |             Theta = auc(freqs,psd,4,7)
 86 |             #Alpha波8-11Hz
 87 |             Alpha = auc(freqs,psd,8,11)
 88 |             #Beta1波12-20Hz
 89 |             Beta1 = auc(freqs,psd,12,20)
 90 |             #Beta2波21-29Hz
 91 |             Beta2 = auc(freqs,psd,21,29)
 92 |             #Gamma1波30-65Hz
 93 |             Gamma1 = auc(freqs,psd,30,65)
 94 |             #Gamma2波66-100Hz
 95 |             Gamma2 = auc(freqs,psd,66,100)
 96 |             reslist[j:j+12] = np.array([Delta, Theta, Alpha, Beta1, Beta2, Gamma1, Gamma2, np.mean(data[216:220]),np.mean(data[220:226]), np.mean(data[250:260]), np.mean(data[255:265]), np.mean(data[275:285])])
 97 |             j += 12
 98 | 
 99 |         result[0][:2] = [sub[-12:-8], condition]
100 |         result[0][2:] = reslist
101 |         epo_data = np.concatenate((epo_data,result),axis=0)
102 |     epo_data = epo_data[1:]
103 |     all_data = np.concatenate((all_data,epo_data),axis=0)
104 |     print('当前被试:%s'%(sub[-12:-8]))
105 | 
106 | df = pd.DataFrame(all_data,index=None)
107 | df.to_excel('eegdata_72var.xlsx')
108 | 
109 | 


--------------------------------------------------------------------------------
/basic_model_kfold.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 12,
  6 |    "metadata": {},
  7 |    "outputs": [],
  8 |    "source": [
  9 |     "import pandas as pd\n",
 10 |     "import numpy as np\n",
 11 |     "from sklearn.preprocessing import StandardScaler\n",
 12 |     "from sklearn.model_selection import train_test_split, cross_val_score\n",
 13 |     "import tensorflow.compat.v1 as tf\n",
 14 |     "from tensorflow import keras\n",
 15 |     "from tensorflow.keras import layers\n",
 16 |     "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint\n",
 17 |     "from ctypes import *\n",
 18 |     "import matplotlib.pyplot as plt\n",
 19 |     "import joblib\n",
 20 |     "\n",
 21 |     "from math import pi, sin, cos, asin, pow, radians, sqrt, fabs\n",
 22 |     "from datetime import datetime\n",
 23 |     "\n",
 24 |     "from xgboost import XGBRegressor\n",
 25 |     "from sklearn.linear_model import LinearRegression\n",
 26 |     "from sklearn.tree import DecisionTreeRegressor\n",
 27 |     "from sklearn.neighbors import KNeighborsRegressor\n",
 28 |     "from sklearn.ensemble import RandomForestRegressor\n",
 29 |     "from sklearn.naive_bayes import GaussianNB\n",
 30 |     "from sklearn.svm import SVR\n",
 31 |     "from lightgbm import LGBMRegressor\n",
 32 |     "from sklearn.neural_network import MLPRegressor\n",
 33 |     "from sklearn.ensemble import GradientBoostingRegressor\n",
 34 |     "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
 35 |     "\n",
 36 |     "# Other machine learning tools\n",
 37 |     "from imblearn.over_sampling import SMOTE\n",
 38 |     "from sklearn.model_selection import KFold\n",
 39 |     "from sklearn.metrics import accuracy_score, r2_score\n",
 40 |     "from sklearn.metrics import roc_auc_score\n",
 41 |     "from sklearn.feature_selection import RFECV\n",
 42 |     "from sklearn.preprocessing import StandardScaler\n",
 43 |     "import scipy.stats as sps \n",
 44 |     "from sklearn.decomposition import PCA\n",
 45 |     "\n",
 46 |     "import warnings\n",
 47 |     "warnings.filterwarnings('ignore')"
 48 |    ]
 49 |   },
 50 |   {
 51 |    "cell_type": "code",
 52 |    "execution_count": 13,
 53 |    "metadata": {},
 54 |    "outputs": [],
 55 |    "source": [
 56 |     "eeg_df = pd.read_excel('eegdata.xlsx')\n",
 57 |     "ques_df = pd.read_excel('questionare.xlsx')\n",
 58 |     "ques_df = ques_df.loc[:,['sub','group','gender','age','外向性','宜人性','尽责性','神经质','开放性']]"
 59 |    ]
 60 |   },
 61 |   {
 62 |    "cell_type": "code",
 63 |    "execution_count": 14,
 64 |    "metadata": {},
 65 |    "outputs": [],
 66 |    "source": [
 67 |     "df_initial = pd.merge(eeg_df,ques_df,on=['sub'])\n",
 68 |     "#将分类变量转为0与1\n",
 69 |     "df_initial.replace('male',1,inplace=True)\n",
 70 |     "df_initial.replace('female',0,inplace=True)\n",
 71 |     "df_initial['group'].replace(1,0,inplace=True)  #group为0的是被社会拒斥的     \n",
 72 |     "df_initial['group'].replace(2,1,inplace=True)  #group为1的是未被社会拒斥的\n",
 73 |     "df_initial['con'].replace('neg',0,inplace=True)\n",
 74 |     "df_initial['con'].replace('pos',1,inplace=True)\n",
 75 |     "\n",
 76 |     "\n",
 77 |     "df_initial.drop(columns=['sub','group'],inplace=True)\n",
 78 |     "x = df_initial.drop(columns=['外向性','宜人性','尽责性','神经质','开放性'])\n",
 79 |     "x = StandardScaler().fit_transform(x)\n",
 80 |     "y = df_initial.loc[:,['外向性','宜人性','尽责性','神经质','开放性']]"
 81 |    ]
 82 |   },
 83 |   {
 84 |    "cell_type": "code",
 85 |    "execution_count": 15,
 86 |    "metadata": {},
 87 |    "outputs": [],
 88 |    "source": [
 89 |     "models = {\n",
 90 |     "    'Decision Tree': DecisionTreeRegressor(),\n",
 91 |     "    'K Nearest Neighbors': KNeighborsRegressor(n_neighbors=1),\n",
 92 |     "    'Random Forest': RandomForestRegressor(n_estimators=2,max_depth=3),\n",
 93 |     "    'Gaussian Naive Bayes': GaussianNB(var_smoothing=1),\n",
 94 |     "    'Light GBM': LGBMRegressor(learning_rate=0.01, max_depth=3, n_estimators=2500),\n",
 95 |     "    'XGBoost': XGBRegressor(n_estimators=2500,max_depth=3),\n",
 96 |     "    'Gradient Boosting': GradientBoostingRegressor(n_estimators=100,max_depth=5),\n",
 97 |     "    \"Neural Network\": MLPRegressor(hidden_layer_sizes=(50,50)),\n",
 98 |     "    #'Support Vector Machine': SVR()\n",
 99 |     "}"
100 |    ]
101 |   },
102 |   {
103 |    "cell_type": "code",
104 |    "execution_count": 16,
105 |    "metadata": {},
106 |    "outputs": [],
107 |    "source": [
108 |     "x = pd.DataFrame(x)"
109 |    ]
110 |   },
111 |   {
112 |    "cell_type": "code",
113 |    "execution_count": 20,
114 |    "metadata": {},
115 |    "outputs": [
116 |     {
117 |      "name": "stdout",
118 |      "output_type": "stream",
119 |      "text": [
120 |       "Decision Tree processed\n",
121 |       "K Nearest Neighbors processed\n",
122 |       "Random Forest processed\n",
123 |       "Gaussian Naive Bayes processed\n",
124 |       "Light GBM processed\n",
125 |       "XGBoost processed\n",
126 |       "Gradient Boosting processed\n",
127 |       "Neural Network processed\n",
128 |       "                           mse\n",
129 |       "Decision Tree        -3.069889\n",
130 |       "K Nearest Neighbors  -5.081464\n",
131 |       "Random Forest        -1.232696\n",
132 |       "Gaussian Naive Bayes -3.060711\n",
133 |       "Light GBM            -0.848961\n",
134 |       "XGBoost              -1.046759\n",
135 |       "Gradient Boosting    -1.113528\n",
136 |       "Neural Network       -1.115467\n",
137 |       "Decision Tree processed\n",
138 |       "K Nearest Neighbors processed\n",
139 |       "Random Forest processed\n",
140 |       "Gaussian Naive Bayes processed\n",
141 |       "Light GBM processed\n",
142 |       "XGBoost processed\n",
143 |       "Gradient Boosting processed\n",
144 |       "Neural Network processed\n",
145 |       "                           mse\n",
146 |       "Decision Tree        -1.293848\n",
147 |       "K Nearest Neighbors   -1.83079\n",
148 |       "Random Forest         -0.73194\n",
149 |       "Gaussian Naive Bayes -1.419298\n",
150 |       "Light GBM            -0.697939\n",
151 |       "XGBoost               -0.90065\n",
152 |       "Gradient Boosting    -0.800727\n",
153 |       "Neural Network       -1.372931\n",
154 |       "Decision Tree processed\n",
155 |       "K Nearest Neighbors processed\n",
156 |       "Random Forest processed\n",
157 |       "Gaussian Naive Bayes processed\n",
158 |       "Light GBM processed\n",
159 |       "XGBoost processed\n",
160 |       "Gradient Boosting processed\n",
161 |       "Neural Network processed\n",
162 |       "                           mse\n",
163 |       "Decision Tree        -1.455673\n",
164 |       "K Nearest Neighbors  -5.326974\n",
165 |       "Random Forest        -0.615393\n",
166 |       "Gaussian Naive Bayes -1.566519\n",
167 |       "Light GBM            -0.715032\n",
168 |       "XGBoost              -0.766682\n",
169 |       "Gradient Boosting    -0.368979\n",
170 |       "Neural Network       -0.907577\n",
171 |       "Decision Tree processed\n",
172 |       "K Nearest Neighbors processed\n",
173 |       "Random Forest processed\n",
174 |       "Gaussian Naive Bayes processed\n",
175 |       "Light GBM processed\n",
176 |       "XGBoost processed\n",
177 |       "Gradient Boosting processed\n",
178 |       "Neural Network processed\n",
179 |       "                           mse\n",
180 |       "Decision Tree        -1.212491\n",
181 |       "K Nearest Neighbors  -8.537275\n",
182 |       "Random Forest        -0.706081\n",
183 |       "Gaussian Naive Bayes -3.486398\n",
184 |       "Light GBM            -0.804913\n",
185 |       "XGBoost              -1.129595\n",
186 |       "Gradient Boosting     -1.07284\n",
187 |       "Neural Network       -0.689893\n",
188 |       "Decision Tree processed\n",
189 |       "K Nearest Neighbors processed\n",
190 |       "Random Forest processed\n",
191 |       "Gaussian Naive Bayes processed\n",
192 |       "Light GBM processed\n",
193 |       "XGBoost processed\n",
194 |       "Gradient Boosting processed\n",
195 |       "Neural Network processed\n",
196 |       "                           mse\n",
197 |       "Decision Tree        -0.414438\n",
198 |       "K Nearest Neighbors  -7.526344\n",
199 |       "Random Forest        -0.456317\n",
200 |       "Gaussian Naive Bayes -4.381091\n",
201 |       "Light GBM            -0.463175\n",
202 |       "XGBoost              -0.751612\n",
203 |       "Gradient Boosting    -0.273935\n",
204 |       "Neural Network        -0.86118\n"
205 |      ]
206 |     }
207 |    ],
208 |    "source": [
209 |     "for i in ['外向性','宜人性','尽责性','神经质','开放性']:\n",
210 |     "    y_temp = y.loc[:,[i]]\n",
211 |     "    holdout_summary = pd.DataFrame(columns = ['mse'])\n",
212 |     "    for name, model in models.items():\n",
213 |     "        scores = cross_val_score(model,x,y_temp,cv=5,scoring='r2')\n",
214 |     "        holdout_summary.loc[name,:] = [scores.mean()]\n",
215 |     "        print(f\"{name} processed\")\n",
216 |     "    print(holdout_summary)\n"
217 |    ]
218 |   }
219 |  ],
220 |  "metadata": {
221 |   "interpreter": {
222 |    "hash": "fb662769bf5854fc00024041705291ada44874f0671c421db64edbd11228109f"
223 |   },
224 |   "kernelspec": {
225 |    "display_name": "Python 3.8.8 ('dl')",
226 |    "language": "python",
227 |    "name": "python3"
228 |   },
229 |   "language_info": {
230 |    "codemirror_mode": {
231 |     "name": "ipython",
232 |     "version": 3
233 |    },
234 |    "file_extension": ".py",
235 |    "mimetype": "text/x-python",
236 |    "name": "python",
237 |    "nbconvert_exporter": "python",
238 |    "pygments_lexer": "ipython3",
239 |    "version": "3.8.8"
240 |   },
241 |   "orig_nbformat": 4
242 |  },
243 |  "nbformat": 4,
244 |  "nbformat_minor": 2
245 | }
246 | 


--------------------------------------------------------------------------------
/mymontage.xml:
--------------------------------------------------------------------------------
  1 | <?xml version="1.0" encoding="utf-8"?>
  2 | <CHANNELS>
  3 | 	<CHANNEL>
  4 | 		<ID>0</ID>
  5 | 		<NAME>1</NAME>
  6 | 		<LABEL>Fp1</LABEL>
  7 | 		<POSITION>29.000000 -104.536000 33.125500</POSITION>
  8 | 		<TYPE>0</TYPE>
  9 | 		<SAMPLERATE>1000</SAMPLERATE>
 10 | 		<UNIT>V</UNIT>
 11 | 		<HPFILTER>0.000000</HPFILTER>
 12 | 		<RANGE>400.000000</RANGE>
 13 | 	</CHANNEL>
 14 | 	<CHANNEL>
 15 | 		<ID>9</ID>
 16 | 		<NAME>10</NAME>
 17 | 		<LABEL>FT11</LABEL>
 18 | 		<POSITION>80.000000 -18.563800 -39.413000</POSITION>
 19 | 		<TYPE>0</TYPE>
 20 | 		<SAMPLERATE>1000</SAMPLERATE>
 21 | 		<UNIT>V</UNIT>
 22 | 		<HPFILTER>0.000000</HPFILTER>
 23 | 		<RANGE>400.000000</RANGE>
 24 | 	</CHANNEL>
 25 | 	<CHANNEL>
 26 | 		<ID>10</ID>
 27 | 		<NAME>11</NAME>
 28 | 		<LABEL>FC3</LABEL>
 29 | 		<POSITION>62.000000 -47.610000 106.204000</POSITION>
 30 | 		<TYPE>0</TYPE>
 31 | 		<SAMPLERATE>1000</SAMPLERATE>
 32 | 		<UNIT>V</UNIT>
 33 | 		<HPFILTER>0.000000</HPFILTER>
 34 | 		<RANGE>400.000000</RANGE>
 35 | 	</CHANNEL>
 36 | 	<CHANNEL>
 37 | 		<ID>11</ID>
 38 | 		<NAME>12</NAME>
 39 | 		<LABEL>FCz</LABEL>
 40 | 		<POSITION>-3.000000 -57.272800 137.295000</POSITION>
 41 | 		<TYPE>0</TYPE>
 42 | 		<SAMPLERATE>1000</SAMPLERATE>
 43 | 		<UNIT>V</UNIT>
 44 | 		<HPFILTER>0.000000</HPFILTER>
 45 | 		<RANGE>400.000000</RANGE>
 46 | 	</CHANNEL>
 47 | 	<CHANNEL>
 48 | 		<ID>12</ID>
 49 | 		<NAME>13</NAME>
 50 | 		<LABEL>FC4</LABEL>
 51 | 		<POSITION>-63.000000 -48.371900 104.102000</POSITION>
 52 | 		<TYPE>0</TYPE>
 53 | 		<SAMPLERATE>1000</SAMPLERATE>
 54 | 		<UNIT>V</UNIT>
 55 | 		<HPFILTER>0.000000</HPFILTER>
 56 | 		<RANGE>400.000000</RANGE>
 57 | 	</CHANNEL>
 58 | 	<CHANNEL>
 59 | 		<ID>13</ID>
 60 | 		<NAME>14</NAME>
 61 | 		<LABEL>FT12</LABEL>
 62 | 		<POSITION>-79.000000 -16.394000 -32.112900</POSITION>
 63 | 		<TYPE>0</TYPE>
 64 | 		<SAMPLERATE>1000</SAMPLERATE>
 65 | 		<UNIT>V</UNIT>
 66 | 		<HPFILTER>0.000000</HPFILTER>
 67 | 		<RANGE>400.000000</RANGE>
 68 | 	</CHANNEL>
 69 | 	<CHANNEL>
 70 | 		<ID>14</ID>
 71 | 		<NAME>15</NAME>
 72 | 		<LABEL>T7</LABEL>
 73 | 		<POSITION>84.000000 -3.519880 38.853700</POSITION>
 74 | 		<TYPE>0</TYPE>
 75 | 		<SAMPLERATE>1000</SAMPLERATE>
 76 | 		<UNIT>V</UNIT>
 77 | 		<HPFILTER>0.000000</HPFILTER>
 78 | 		<RANGE>400.000000</RANGE>
 79 | 	</CHANNEL>
 80 | 	<CHANNEL>
 81 | 		<ID>15</ID>
 82 | 		<NAME>16</NAME>
 83 | 		<LABEL>C3</LABEL>
 84 | 		<POSITION>69.000000 -14.764200 118.085000</POSITION>
 85 | 		<TYPE>0</TYPE>
 86 | 		<SAMPLERATE>1000</SAMPLERATE>
 87 | 		<UNIT>V</UNIT>
 88 | 		<HPFILTER>0.000000</HPFILTER>
 89 | 		<RANGE>400.000000</RANGE>
 90 | 	</CHANNEL>
 91 | 	<CHANNEL>
 92 | 		<ID>16</ID>
 93 | 		<NAME>17</NAME>
 94 | 		<LABEL>Cz</LABEL>
 95 | 		<POSITION>-3.000000 -22.025800 154.489000</POSITION>
 96 | 		<TYPE>0</TYPE>
 97 | 		<SAMPLERATE>1000</SAMPLERATE>
 98 | 		<UNIT>V</UNIT>
 99 | 		<HPFILTER>0.000000</HPFILTER>
100 | 		<RANGE>400.000000</RANGE>
101 | 	</CHANNEL>
102 | 	<CHANNEL>
103 | 		<ID>17</ID>
104 | 		<NAME>18</NAME>
105 | 		<LABEL>C4</LABEL>
106 | 		<POSITION>-69.000000 -15.757500 117.969000</POSITION>
107 | 		<TYPE>0</TYPE>
108 | 		<SAMPLERATE>1000</SAMPLERATE>
109 | 		<UNIT>V</UNIT>
110 | 		<HPFILTER>0.000000</HPFILTER>
111 | 		<RANGE>400.000000</RANGE>
112 | 	</CHANNEL>
113 | 	<CHANNEL>
114 | 		<ID>18</ID>
115 | 		<NAME>19</NAME>
116 | 		<LABEL>T8</LABEL>
117 | 		<POSITION>-84.000000 -3.982770 42.826800</POSITION>
118 | 		<TYPE>0</TYPE>
119 | 		<SAMPLERATE>1000</SAMPLERATE>
120 | 		<UNIT>V</UNIT>
121 | 		<HPFILTER>0.000000</HPFILTER>
122 | 		<RANGE>400.000000</RANGE>
123 | 	</CHANNEL>
124 | 	<CHANNEL>
125 | 		<ID>1</ID>
126 | 		<NAME>2</NAME>
127 | 		<LABEL>Fp2</LABEL>
128 | 		<POSITION>-30.000000 -104.651000 34.118700</POSITION>
129 | 		<TYPE>0</TYPE>
130 | 		<SAMPLERATE>1000</SAMPLERATE>
131 | 		<UNIT>V</UNIT>
132 | 		<HPFILTER>0.000000</HPFILTER>
133 | 		<RANGE>400.000000</RANGE>
134 | 	</CHANNEL>
135 | 	<CHANNEL>
136 | 		<ID>19</ID>
137 | 		<NAME>20</NAME>
138 | 		<LABEL>CP3</LABEL>
139 | 		<POSITION>67.000000 22.980500 122.482000</POSITION>
140 | 		<TYPE>0</TYPE>
141 | 		<SAMPLERATE>1000</SAMPLERATE>
142 | 		<UNIT>V</UNIT>
143 | 		<HPFILTER>0.000000</HPFILTER>
144 | 		<RANGE>400.000000</RANGE>
145 | 	</CHANNEL>
146 | 	<CHANNEL>
147 | 		<ID>20</ID>
148 | 		<NAME>21</NAME>
149 | 		<LABEL>CPz</LABEL>
150 | 		<POSITION>-3.000000 17.821200 158.125000</POSITION>
151 | 		<TYPE>0</TYPE>
152 | 		<SAMPLERATE>1000</SAMPLERATE>
153 | 		<UNIT>V</UNIT>
154 | 		<HPFILTER>0.000000</HPFILTER>
155 | 		<RANGE>400.000000</RANGE>
156 | 	</CHANNEL>
157 | 	<CHANNEL>
158 | 		<ID>21</ID>
159 | 		<NAME>22</NAME>
160 | 		<LABEL>CP4</LABEL>
161 | 		<POSITION>-67.000000 19.007400 122.019000</POSITION>
162 | 		<TYPE>0</TYPE>
163 | 		<SAMPLERATE>1000</SAMPLERATE>
164 | 		<UNIT>V</UNIT>
165 | 		<HPFILTER>0.000000</HPFILTER>
166 | 		<RANGE>400.000000</RANGE>
167 | 	</CHANNEL>
168 | 	<CHANNEL>
169 | 		<ID>22</ID>
170 | 		<NAME>23</NAME>
171 | 		<LABEL>M1</LABEL>
172 | 		<POSITION>76.000000 41.254900 8.833460</POSITION>
173 | 		<TYPE>0</TYPE>
174 | 		<SAMPLERATE>1000</SAMPLERATE>
175 | 		<UNIT>V</UNIT>
176 | 		<HPFILTER>0.000000</HPFILTER>
177 | 		<RANGE>400.000000</RANGE>
178 | 	</CHANNEL>
179 | 	<CHANNEL>
180 | 		<ID>23</ID>
181 | 		<NAME>24</NAME>
182 | 		<LABEL>M2</LABEL>
183 | 		<POSITION>-77.000000 38.458300 15.555000</POSITION>
184 | 		<TYPE>0</TYPE>
185 | 		<SAMPLERATE>1000</SAMPLERATE>
186 | 		<UNIT>V</UNIT>
187 | 		<HPFILTER>0.000000</HPFILTER>
188 | 		<RANGE>400.000000</RANGE>
189 | 	</CHANNEL>
190 | 	<CHANNEL>
191 | 		<ID>24</ID>
192 | 		<NAME>25</NAME>
193 | 		<LABEL>P7</LABEL>
194 | 		<POSITION>73.000000 54.505100 50.647700</POSITION>
195 | 		<TYPE>0</TYPE>
196 | 		<SAMPLERATE>1000</SAMPLERATE>
197 | 		<UNIT>V</UNIT>
198 | 		<HPFILTER>0.000000</HPFILTER>
199 | 		<RANGE>400.000000</RANGE>
200 | 	</CHANNEL>
201 | 	<CHANNEL>
202 | 		<ID>25</ID>
203 | 		<NAME>26</NAME>
204 | 		<LABEL>P3</LABEL>
205 | 		<POSITION>56.000000 55.160900 114.150000</POSITION>
206 | 		<TYPE>0</TYPE>
207 | 		<SAMPLERATE>1000</SAMPLERATE>
208 | 		<UNIT>V</UNIT>
209 | 		<HPFILTER>0.000000</HPFILTER>
210 | 		<RANGE>400.000000</RANGE>
211 | 	</CHANNEL>
212 | 	<CHANNEL>
213 | 		<ID>26</ID>
214 | 		<NAME>27</NAME>
215 | 		<LABEL>Pz</LABEL>
216 | 		<POSITION>-2.000000 52.913900 142.078000</POSITION>
217 | 		<TYPE>0</TYPE>
218 | 		<SAMPLERATE>1000</SAMPLERATE>
219 | 		<UNIT>V</UNIT>
220 | 		<HPFILTER>0.000000</HPFILTER>
221 | 		<RANGE>400.000000</RANGE>
222 | 	</CHANNEL>
223 | 	<CHANNEL>
224 | 		<ID>27</ID>
225 | 		<NAME>28</NAME>
226 | 		<LABEL>P4</LABEL>
227 | 		<POSITION>-56.000000 53.174300 113.919000</POSITION>
228 | 		<TYPE>0</TYPE>
229 | 		<SAMPLERATE>1000</SAMPLERATE>
230 | 		<UNIT>V</UNIT>
231 | 		<HPFILTER>0.000000</HPFILTER>
232 | 		<RANGE>400.000000</RANGE>
233 | 	</CHANNEL>
234 | 	<CHANNEL>
235 | 		<ID>28</ID>
236 | 		<NAME>29</NAME>
237 | 		<LABEL>P8</LABEL>
238 | 		<POSITION>-72.000000 52.171400 53.396100</POSITION>
239 | 		<TYPE>0</TYPE>
240 | 		<SAMPLERATE>1000</SAMPLERATE>
241 | 		<UNIT>V</UNIT>
242 | 		<HPFILTER>0.000000</HPFILTER>
243 | 		<RANGE>400.000000</RANGE>
244 | 	</CHANNEL>
245 | 	<CHANNEL>
246 | 		<ID>2</ID>
247 | 		<NAME>3</NAME>
248 | 		<LABEL>F11</LABEL>
249 | 		<POSITION>74.000000 -46.771000 -30.618100</POSITION>
250 | 		<TYPE>0</TYPE>
251 | 		<SAMPLERATE>1000</SAMPLERATE>
252 | 		<UNIT>V</UNIT>
253 | 		<HPFILTER>0.000000</HPFILTER>
254 | 		<RANGE>400.000000</RANGE>
255 | 	</CHANNEL>
256 | 	<CHANNEL>
257 | 		<ID>29</ID>
258 | 		<NAME>30</NAME>
259 | 		<LABEL>O1</LABEL>
260 | 		<POSITION>29.000000 91.854400 67.080300</POSITION>
261 | 		<TYPE>0</TYPE>
262 | 		<SAMPLERATE>1000</SAMPLERATE>
263 | 		<UNIT>V</UNIT>
264 | 		<HPFILTER>0.000000</HPFILTER>
265 | 		<RANGE>400.000000</RANGE>
266 | 	</CHANNEL>
267 | 	<CHANNEL>
268 | 		<ID>30</ID>
269 | 		<NAME>31</NAME>
270 | 		<LABEL>Oz</LABEL>
271 | 		<POSITION>-1.000000 94.024200 74.380400</POSITION>
272 | 		<TYPE>0</TYPE>
273 | 		<SAMPLERATE>1000</SAMPLERATE>
274 | 		<UNIT>V</UNIT>
275 | 		<HPFILTER>0.000000</HPFILTER>
276 | 		<RANGE>400.000000</RANGE>
277 | 	</CHANNEL>
278 | 	<CHANNEL>
279 | 		<ID>31</ID>
280 | 		<NAME>32</NAME>
281 | 		<LABEL>O2</LABEL>
282 | 		<POSITION>-30.000000 90.745400 67.957800</POSITION>
283 | 		<TYPE>0</TYPE>
284 | 		<SAMPLERATE>1000</SAMPLERATE>
285 | 		<UNIT>V</UNIT>
286 | 		<HPFILTER>0.000000</HPFILTER>
287 | 		<RANGE>400.000000</RANGE>
288 | 	</CHANNEL>
289 | 	<CHANNEL>
290 | 		<ID>32</ID>
291 | 		<NAME>33</NAME>
292 | 		<LABEL>33</LABEL>
293 | 		<POSITION>-1.000000 -106.763000 78.170300</POSITION>
294 | 		<TYPE>0</TYPE>
295 | 		<SAMPLERATE>0</SAMPLERATE>
296 | 		<UNIT>V</UNIT>
297 | 		<HPFILTER>0.000000</HPFILTER>
298 | 		<RANGE>400.000000</RANGE>
299 | 	</CHANNEL>
300 | 	<CHANNEL>
301 | 		<ID>33</ID>
302 | 		<NAME>34</NAME>
303 | 		<LABEL>34</LABEL>
304 | 		<POSITION>-3.000000 -2.391590 158.790000</POSITION>
305 | 		<TYPE>0</TYPE>
306 | 		<SAMPLERATE>0</SAMPLERATE>
307 | 		<UNIT>V</UNIT>
308 | 		<HPFILTER>0.000000</HPFILTER>
309 | 		<RANGE>400.000000</RANGE>
310 | 	</CHANNEL>
311 | 	<CHANNEL>
312 | 		<ID>34</ID>
313 | 		<NAME>35</NAME>
314 | 		<LABEL>35</LABEL>
315 | 		<TYPE>0</TYPE>
316 | 		<SAMPLERATE>0</SAMPLERATE>
317 | 		<UNIT>V</UNIT>
318 | 		<HPFILTER>0.000000</HPFILTER>
319 | 		<RANGE>400.000000</RANGE>
320 | 	</CHANNEL>
321 | 	<CHANNEL>
322 | 		<ID>35</ID>
323 | 		<NAME>36</NAME>
324 | 		<LABEL>36</LABEL>
325 | 		<TYPE>0</TYPE>
326 | 		<SAMPLERATE>0</SAMPLERATE>
327 | 		<UNIT>V</UNIT>
328 | 		<HPFILTER>0.000000</HPFILTER>
329 | 		<RANGE>400.000000</RANGE>
330 | 	</CHANNEL>
331 | 	<CHANNEL>
332 | 		<ID>36</ID>
333 | 		<NAME>37</NAME>
334 | 		<LABEL>37</LABEL>
335 | 		<TYPE>0</TYPE>
336 | 		<SAMPLERATE>0</SAMPLERATE>
337 | 		<UNIT>V</UNIT>
338 | 		<HPFILTER>0.000000</HPFILTER>
339 | 		<RANGE>400.000000</RANGE>
340 | 	</CHANNEL>
341 | 	<CHANNEL>
342 | 		<ID>37</ID>
343 | 		<NAME>38</NAME>
344 | 		<LABEL>38</LABEL>
345 | 		<TYPE>0</TYPE>
346 | 		<SAMPLERATE>0</SAMPLERATE>
347 | 		<UNIT>V</UNIT>
348 | 		<HPFILTER>0.000000</HPFILTER>
349 | 		<RANGE>400.000000</RANGE>
350 | 	</CHANNEL>
351 | 	<CHANNEL>
352 | 		<ID>38</ID>
353 | 		<NAME>39</NAME>
354 | 		<LABEL>39</LABEL>
355 | 		<TYPE>0</TYPE>
356 | 		<SAMPLERATE>0</SAMPLERATE>
357 | 		<UNIT>V</UNIT>
358 | 		<HPFILTER>0.000000</HPFILTER>
359 | 		<RANGE>400.000000</RANGE>
360 | 	</CHANNEL>
361 | 	<CHANNEL>
362 | 		<ID>3</ID>
363 | 		<NAME>4</NAME>
364 | 		<LABEL>F7</LABEL>
365 | 		<POSITION>70.000000 -61.892100 30.039500</POSITION>
366 | 		<TYPE>0</TYPE>
367 | 		<SAMPLERATE>1000</SAMPLERATE>
368 | 		<UNIT>V</UNIT>
369 | 		<HPFILTER>0.000000</HPFILTER>
370 | 		<RANGE>400.000000</RANGE>
371 | 	</CHANNEL>
372 | 	<CHANNEL>
373 | 		<ID>39</ID>
374 | 		<NAME>40</NAME>
375 | 		<LABEL>40</LABEL>
376 | 		<TYPE>0</TYPE>
377 | 		<SAMPLERATE>0</SAMPLERATE>
378 | 		<UNIT>V</UNIT>
379 | 		<HPFILTER>0.000000</HPFILTER>
380 | 		<RANGE>400.000000</RANGE>
381 | 	</CHANNEL>
382 | 	<CHANNEL>
383 | 		<ID>40</ID>
384 | 		<NAME>41</NAME>
385 | 		<LABEL>41</LABEL>
386 | 		<TYPE>0</TYPE>
387 | 		<SAMPLERATE>0</SAMPLERATE>
388 | 		<UNIT>V</UNIT>
389 | 		<HPFILTER>0.000000</HPFILTER>
390 | 		<RANGE>400.000000</RANGE>
391 | 	</CHANNEL>
392 | 	<CHANNEL>
393 | 		<ID>41</ID>
394 | 		<NAME>42</NAME>
395 | 		<LABEL>42</LABEL>
396 | 		<TYPE>0</TYPE>
397 | 		<SAMPLERATE>0</SAMPLERATE>
398 | 		<UNIT>V</UNIT>
399 | 		<HPFILTER>0.000000</HPFILTER>
400 | 		<RANGE>400.000000</RANGE>
401 | 	</CHANNEL>
402 | 	<CHANNEL>
403 | 		<ID>42</ID>
404 | 		<NAME>43</NAME>
405 | 		<LABEL>43</LABEL>
406 | 		<TYPE>0</TYPE>
407 | 		<SAMPLERATE>0</SAMPLERATE>
408 | 		<UNIT>V</UNIT>
409 | 		<HPFILTER>0.000000</HPFILTER>
410 | 		<RANGE>400.000000</RANGE>
411 | 	</CHANNEL>
412 | 	<CHANNEL>
413 | 		<ID>43</ID>
414 | 		<NAME>44</NAME>
415 | 		<LABEL>44</LABEL>
416 | 		<TYPE>0</TYPE>
417 | 		<SAMPLERATE>0</SAMPLERATE>
418 | 		<UNIT>V</UNIT>
419 | 		<HPFILTER>0.000000</HPFILTER>
420 | 		<RANGE>400.000000</RANGE>
421 | 	</CHANNEL>
422 | 	<CHANNEL>
423 | 		<ID>44</ID>
424 | 		<NAME>45</NAME>
425 | 		<LABEL>45</LABEL>
426 | 		<TYPE>0</TYPE>
427 | 		<SAMPLERATE>0</SAMPLERATE>
428 | 		<UNIT>V</UNIT>
429 | 		<HPFILTER>0.000000</HPFILTER>
430 | 		<RANGE>400.000000</RANGE>
431 | 	</CHANNEL>
432 | 	<CHANNEL>
433 | 		<ID>45</ID>
434 | 		<NAME>46</NAME>
435 | 		<LABEL>46</LABEL>
436 | 		<TYPE>0</TYPE>
437 | 		<SAMPLERATE>0</SAMPLERATE>
438 | 		<UNIT>V</UNIT>
439 | 		<HPFILTER>0.000000</HPFILTER>
440 | 		<RANGE>400.000000</RANGE>
441 | 	</CHANNEL>
442 | 	<CHANNEL>
443 | 		<ID>46</ID>
444 | 		<NAME>47</NAME>
445 | 		<LABEL>47</LABEL>
446 | 		<TYPE>0</TYPE>
447 | 		<SAMPLERATE>0</SAMPLERATE>
448 | 		<UNIT>V</UNIT>
449 | 		<HPFILTER>0.000000</HPFILTER>
450 | 		<RANGE>400.000000</RANGE>
451 | 	</CHANNEL>
452 | 	<CHANNEL>
453 | 		<ID>47</ID>
454 | 		<NAME>48</NAME>
455 | 		<LABEL>48</LABEL>
456 | 		<TYPE>0</TYPE>
457 | 		<SAMPLERATE>0</SAMPLERATE>
458 | 		<UNIT>V</UNIT>
459 | 		<HPFILTER>0.000000</HPFILTER>
460 | 		<RANGE>400.000000</RANGE>
461 | 	</CHANNEL>
462 | 	<CHANNEL>
463 | 		<ID>48</ID>
464 | 		<NAME>49</NAME>
465 | 		<LABEL>49</LABEL>
466 | 		<TYPE>0</TYPE>
467 | 		<SAMPLERATE>0</SAMPLERATE>
468 | 		<UNIT>V</UNIT>
469 | 		<HPFILTER>0.000000</HPFILTER>
470 | 		<RANGE>400.000000</RANGE>
471 | 	</CHANNEL>
472 | 	<CHANNEL>
473 | 		<ID>4</ID>
474 | 		<NAME>5</NAME>
475 | 		<LABEL>F3</LABEL>
476 | 		<POSITION>52.000000 -76.781600 88.710700</POSITION>
477 | 		<TYPE>0</TYPE>
478 | 		<SAMPLERATE>1000</SAMPLERATE>
479 | 		<UNIT>V</UNIT>
480 | 		<HPFILTER>0.000000</HPFILTER>
481 | 		<RANGE>400.000000</RANGE>
482 | 	</CHANNEL>
483 | 	<CHANNEL>
484 | 		<ID>49</ID>
485 | 		<NAME>50</NAME>
486 | 		<LABEL>50</LABEL>
487 | 		<TYPE>0</TYPE>
488 | 		<SAMPLERATE>0</SAMPLERATE>
489 | 		<UNIT>V</UNIT>
490 | 		<HPFILTER>0.000000</HPFILTER>
491 | 		<RANGE>400.000000</RANGE>
492 | 	</CHANNEL>
493 | 	<CHANNEL>
494 | 		<ID>50</ID>
495 | 		<NAME>51</NAME>
496 | 		<LABEL>51</LABEL>
497 | 		<TYPE>0</TYPE>
498 | 		<SAMPLERATE>0</SAMPLERATE>
499 | 		<UNIT>V</UNIT>
500 | 		<HPFILTER>0.000000</HPFILTER>
501 | 		<RANGE>400.000000</RANGE>
502 | 	</CHANNEL>
503 | 	<CHANNEL>
504 | 		<ID>51</ID>
505 | 		<NAME>52</NAME>
506 | 		<LABEL>52</LABEL>
507 | 		<TYPE>0</TYPE>
508 | 		<SAMPLERATE>0</SAMPLERATE>
509 | 		<UNIT>V</UNIT>
510 | 		<HPFILTER>0.000000</HPFILTER>
511 | 		<RANGE>400.000000</RANGE>
512 | 	</CHANNEL>
513 | 	<CHANNEL>
514 | 		<ID>52</ID>
515 | 		<NAME>53</NAME>
516 | 		<LABEL>53</LABEL>
517 | 		<TYPE>0</TYPE>
518 | 		<SAMPLERATE>0</SAMPLERATE>
519 | 		<UNIT>V</UNIT>
520 | 		<HPFILTER>0.000000</HPFILTER>
521 | 		<RANGE>400.000000</RANGE>
522 | 	</CHANNEL>
523 | 	<CHANNEL>
524 | 		<ID>53</ID>
525 | 		<NAME>54</NAME>
526 | 		<LABEL>54</LABEL>
527 | 		<TYPE>0</TYPE>
528 | 		<SAMPLERATE>0</SAMPLERATE>
529 | 		<UNIT>V</UNIT>
530 | 		<HPFILTER>0.000000</HPFILTER>
531 | 		<RANGE>400.000000</RANGE>
532 | 	</CHANNEL>
533 | 	<CHANNEL>
534 | 		<ID>54</ID>
535 | 		<NAME>55</NAME>
536 | 		<LABEL>55</LABEL>
537 | 		<TYPE>0</TYPE>
538 | 		<SAMPLERATE>0</SAMPLERATE>
539 | 		<UNIT>V</UNIT>
540 | 		<HPFILTER>0.000000</HPFILTER>
541 | 		<RANGE>400.000000</RANGE>
542 | 	</CHANNEL>
543 | 	<CHANNEL>
544 | 		<ID>55</ID>
545 | 		<NAME>56</NAME>
546 | 		<LABEL>56</LABEL>
547 | 		<TYPE>0</TYPE>
548 | 		<SAMPLERATE>0</SAMPLERATE>
549 | 		<UNIT>V</UNIT>
550 | 		<HPFILTER>0.000000</HPFILTER>
551 | 		<RANGE>400.000000</RANGE>
552 | 	</CHANNEL>
553 | 	<CHANNEL>
554 | 		<ID>56</ID>
555 | 		<NAME>57</NAME>
556 | 		<LABEL>57</LABEL>
557 | 		<TYPE>0</TYPE>
558 | 		<SAMPLERATE>0</SAMPLERATE>
559 | 		<UNIT>V</UNIT>
560 | 		<HPFILTER>0.000000</HPFILTER>
561 | 		<RANGE>400.000000</RANGE>
562 | 	</CHANNEL>
563 | 	<CHANNEL>
564 | 		<ID>57</ID>
565 | 		<NAME>58</NAME>
566 | 		<LABEL>58</LABEL>
567 | 		<TYPE>0</TYPE>
568 | 		<SAMPLERATE>0</SAMPLERATE>
569 | 		<UNIT>V</UNIT>
570 | 		<HPFILTER>0.000000</HPFILTER>
571 | 		<RANGE>400.000000</RANGE>
572 | 	</CHANNEL>
573 | 	<CHANNEL>
574 | 		<ID>58</ID>
575 | 		<NAME>59</NAME>
576 | 		<LABEL>59</LABEL>
577 | 		<TYPE>0</TYPE>
578 | 		<SAMPLERATE>0</SAMPLERATE>
579 | 		<UNIT>V</UNIT>
580 | 		<HPFILTER>0.000000</HPFILTER>
581 | 		<RANGE>400.000000</RANGE>
582 | 	</CHANNEL>
583 | 	<CHANNEL>
584 | 		<ID>5</ID>
585 | 		<NAME>6</NAME>
586 | 		<LABEL>Fz</LABEL>
587 | 		<POSITION>-2.000000 -88.729900 113.494000</POSITION>
588 | 		<TYPE>0</TYPE>
589 | 		<SAMPLERATE>1000</SAMPLERATE>
590 | 		<UNIT>V</UNIT>
591 | 		<HPFILTER>0.000000</HPFILTER>
592 | 		<RANGE>400.000000</RANGE>
593 | 	</CHANNEL>
594 | 	<CHANNEL>
595 | 		<ID>59</ID>
596 | 		<NAME>60</NAME>
597 | 		<LABEL>60</LABEL>
598 | 		<TYPE>0</TYPE>
599 | 		<SAMPLERATE>0</SAMPLERATE>
600 | 		<UNIT>V</UNIT>
601 | 		<HPFILTER>0.000000</HPFILTER>
602 | 		<RANGE>400.000000</RANGE>
603 | 	</CHANNEL>
604 | 	<CHANNEL>
605 | 		<ID>60</ID>
606 | 		<NAME>61</NAME>
607 | 		<LABEL>61</LABEL>
608 | 		<TYPE>0</TYPE>
609 | 		<SAMPLERATE>0</SAMPLERATE>
610 | 		<UNIT>V</UNIT>
611 | 		<HPFILTER>0.000000</HPFILTER>
612 | 		<RANGE>400.000000</RANGE>
613 | 	</CHANNEL>
614 | 	<CHANNEL>
615 | 		<ID>61</ID>
616 | 		<NAME>62</NAME>
617 | 		<LABEL>62</LABEL>
618 | 		<TYPE>0</TYPE>
619 | 		<SAMPLERATE>0</SAMPLERATE>
620 | 		<UNIT>V</UNIT>
621 | 		<HPFILTER>0.000000</HPFILTER>
622 | 		<RANGE>400.000000</RANGE>
623 | 	</CHANNEL>
624 | 	<CHANNEL>
625 | 		<ID>62</ID>
626 | 		<NAME>63</NAME>
627 | 		<LABEL>63</LABEL>
628 | 		<TYPE>0</TYPE>
629 | 		<SAMPLERATE>0</SAMPLERATE>
630 | 		<UNIT>V</UNIT>
631 | 		<HPFILTER>0.000000</HPFILTER>
632 | 		<RANGE>400.000000</RANGE>
633 | 	</CHANNEL>
634 | 	<CHANNEL>
635 | 		<ID>63</ID>
636 | 		<NAME>64</NAME>
637 | 		<LABEL>64</LABEL>
638 | 		<TYPE>0</TYPE>
639 | 		<SAMPLERATE>0</SAMPLERATE>
640 | 		<UNIT>V</UNIT>
641 | 		<HPFILTER>0.000000</HPFILTER>
642 | 		<RANGE>400.000000</RANGE>
643 | 	</CHANNEL>
644 | 	<CHANNEL>
645 | 		<ID>6</ID>
646 | 		<NAME>7</NAME>
647 | 		<LABEL>F4</LABEL>
648 | 		<POSITION>-53.000000 -78.652500 87.485900</POSITION>
649 | 		<TYPE>0</TYPE>
650 | 		<SAMPLERATE>1000</SAMPLERATE>
651 | 		<UNIT>V</UNIT>
652 | 		<HPFILTER>0.000000</HPFILTER>
653 | 		<RANGE>400.000000</RANGE>
654 | 	</CHANNEL>
655 | 	<CHANNEL>
656 | 		<ID>7</ID>
657 | 		<NAME>8</NAME>
658 | 		<LABEL>F8</LABEL>
659 | 		<POSITION>-70.000000 -61.245900 33.135100</POSITION>
660 | 		<TYPE>0</TYPE>
661 | 		<SAMPLERATE>1000</SAMPLERATE>
662 | 		<UNIT>V</UNIT>
663 | 		<HPFILTER>0.000000</HPFILTER>
664 | 		<RANGE>400.000000</RANGE>
665 | 	</CHANNEL>
666 | 	<CHANNEL>
667 | 		<ID>8</ID>
668 | 		<NAME>9</NAME>
669 | 		<LABEL>F12</LABEL>
670 | 		<POSITION>-74.000000 -44.485500 -24.311300</POSITION>
671 | 		<TYPE>0</TYPE>
672 | 		<SAMPLERATE>1000</SAMPLERATE>
673 | 		<UNIT>V</UNIT>
674 | 		<HPFILTER>0.000000</HPFILTER>
675 | 		<RANGE>400.000000</RANGE>
676 | 	</CHANNEL>
677 | 	<CHANNEL>
678 | 		<ID>64</ID>
679 | 		<NAME>BP 1</NAME>
680 | 		<LABEL>VEOG</LABEL>
681 | 		<TYPE>1</TYPE>
682 | 		<SAMPLERATE>1000</SAMPLERATE>
683 | 		<UNIT>V</UNIT>
684 | 		<HPFILTER>0.000000</HPFILTER>
685 | 		<RANGE>400.000000</RANGE>
686 | 		<UPPER>0.000000</UPPER>
687 | 		<LOWER>0.000000</LOWER>
688 | 	</CHANNEL>
689 | 	<CHANNEL>
690 | 		<ID>65</ID>
691 | 		<NAME>BP 2</NAME>
692 | 		<LABEL>HEOG</LABEL>
693 | 		<TYPE>1</TYPE>
694 | 		<SAMPLERATE>1000</SAMPLERATE>
695 | 		<UNIT>V</UNIT>
696 | 		<HPFILTER>0.000000</HPFILTER>
697 | 		<RANGE>400.000000</RANGE>
698 | 		<UPPER>0.000000</UPPER>
699 | 		<LOWER>0.000000</LOWER>
700 | 	</CHANNEL>
701 | 	<CHANNEL>
702 | 		<ID>66</ID>
703 | 		<NAME>BP 3</NAME>
704 | 		<LABEL>EKG</LABEL>
705 | 		<TYPE>1</TYPE>
706 | 		<SAMPLERATE>1000</SAMPLERATE>
707 | 		<UNIT>V</UNIT>
708 | 		<HPFILTER>0.000000</HPFILTER>
709 | 		<RANGE>400.000000</RANGE>
710 | 		<UPPER>0.000000</UPPER>
711 | 		<LOWER>0.000000</LOWER>
712 | 	</CHANNEL>
713 | 	<CHANNEL>
714 | 		<ID>67</ID>
715 | 		<NAME>BP 4</NAME>
716 | 		<LABEL>EMG</LABEL>
717 | 		<TYPE>1</TYPE>
718 | 		<SAMPLERATE>1000</SAMPLERATE>
719 | 		<UNIT>V</UNIT>
720 | 		<HPFILTER>0.000000</HPFILTER>
721 | 		<RANGE>400.000000</RANGE>
722 | 		<UPPER>0.000000</UPPER>
723 | 		<LOWER>0.000000</LOWER>
724 | 	</CHANNEL>
725 | 	<CHANNEL>
726 | 		<ID>280</ID>
727 | 		<NAME>Trigger</NAME>
728 | 		<LABEL/>
729 | 		<LABEL_ALT>Trigger</LABEL_ALT>
730 | 		<TYPE>3</TYPE>
731 | 		<SAMPLERATE>1000</SAMPLERATE>
732 | 		<UNIT/>
733 | 		<HPFILTER>0.000000</HPFILTER>
734 | 		<RANGE>400.000000</RANGE>
735 | 	</CHANNEL>
736 | </CHANNELS>
737 | 


--------------------------------------------------------------------------------
/permutation_importance.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 3,
  6 |    "metadata": {},
  7 |    "outputs": [
  8 |     {
  9 |      "data": {
 10 |       "text/plain": [
 11 |        "<WinDLL 'G:\\CUDA\\cudnn\\cuda\\bin\\cudnn_adv_train64_8.dll', handle 7ff81cca0000 at 0x18b94f8ae50>"
 12 |       ]
 13 |      },
 14 |      "execution_count": 3,
 15 |      "metadata": {},
 16 |      "output_type": "execute_result"
 17 |     }
 18 |    ],
 19 |    "source": [
 20 |     "import numpy as np\n",
 21 |     "import pandas as pd\n",
 22 |     "from sklearn.model_selection import train_test_split\n",
 23 |     "from sklearn.preprocessing import StandardScaler\n",
 24 |     "import keras\n",
 25 |     "import sklearn\n",
 26 |     "from ctypes import *\n",
 27 |     "WinDLL('G:\\\\CUDA\\\\cudnn\\\\cuda\\\\bin\\\\cudnn_ops_infer64_8.dll')\n",
 28 |     "WinDLL('G:\\\\CUDA\\\\cudnn\\\\cuda\\\\bin\\\\cudnn_adv_infer64_8.dll')\n",
 29 |     "WinDLL('G:\\\\CUDA\\\\cudnn\\\\cuda\\\\bin\\\\cudnn_adv_train64_8.dll')"
 30 |    ]
 31 |   },
 32 |   {
 33 |    "cell_type": "code",
 34 |    "execution_count": 4,
 35 |    "metadata": {},
 36 |    "outputs": [],
 37 |    "source": [
 38 |     "eeg_df = pd.read_excel('eegdata.xlsx')\n",
 39 |     "ques_df = pd.read_excel('questionare.xlsx')\n",
 40 |     "ques_df = ques_df.loc[:,['sub','group','gender','age','外向性','宜人性','尽责性','神经质','开放性']]"
 41 |    ]
 42 |   },
 43 |   {
 44 |    "cell_type": "code",
 45 |    "execution_count": 5,
 46 |    "metadata": {},
 47 |    "outputs": [],
 48 |    "source": [
 49 |     "df_initial = pd.merge(eeg_df,ques_df,on=['sub'])\n",
 50 |     "#将分类变量转为0与1\n",
 51 |     "df_initial.replace('male',1,inplace=True)\n",
 52 |     "df_initial.replace('female',0,inplace=True)\n",
 53 |     "df_initial['group'].replace(1,0,inplace=True)  #group为0的是被社会拒斥的        ##！！！！group实际上不应该出现在我的实验中，考虑删掉这个变量\n",
 54 |     "df_initial['group'].replace(2,1,inplace=True)  #group为1的是未被社会拒斥的\n",
 55 |     "df_initial['con'].replace('neg',0,inplace=True)\n",
 56 |     "df_initial['con'].replace('pos',1,inplace=True)\n",
 57 |     "\n",
 58 |     "\n",
 59 |     "df_initial.drop(columns=['sub','group'],inplace=True)\n",
 60 |     "x = df_initial.drop(columns=['外向性','宜人性','尽责性','神经质','开放性'])\n",
 61 |     "x = pd.DataFrame(StandardScaler().fit_transform(x),columns=x.columns)\n",
 62 |     "\n",
 63 |     "y = df_initial.loc[:,['外向性','宜人性','尽责性','神经质','开放性']]"
 64 |    ]
 65 |   },
 66 |   {
 67 |    "cell_type": "code",
 68 |    "execution_count": 6,
 69 |    "metadata": {},
 70 |    "outputs": [
 71 |     {
 72 |      "ename": "KeyboardInterrupt",
 73 |      "evalue": "",
 74 |      "output_type": "error",
 75 |      "traceback": [
 76 |       "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
 77 |       "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
 78 |       "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_4936/137572014.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[0mx_train\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mx_test\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0my_train\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0my_test\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtrain_test_split\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0my_temp\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mtest_size\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m0.15\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mrandom_state\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      6\u001b[0m \u001b[0mmodel\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mkeras\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmodels\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mload_model\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'models\\\\'\u001b[0m\u001b[1;33m+\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m+\u001b[0m\u001b[1;34m'.h5'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 7\u001b[1;33m \u001b[0mperm\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mPermutationImportance\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscoring\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m'neg_mean_squared_error'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mrandom_state\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_test\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my_test\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      8\u001b[0m \u001b[0meli5\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshow_weights\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mperm\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfeature_names\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mx_test\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mtolist\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 79 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\eli5\\sklearn\\permutation_importance.py\u001b[0m in \u001b[0;36mfit\u001b[1;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[0;32m    202\u001b[0m             \u001b[0msi\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_cv_scores_importances\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mgroups\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mgroups\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mfit_params\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    203\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 204\u001b[1;33m             \u001b[0msi\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_non_cv_scores_importances\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    205\u001b[0m         \u001b[0mscores\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mresults\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0msi\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    206\u001b[0m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mscores_\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0marray\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mscores\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 80 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\eli5\\sklearn\\permutation_importance.py\u001b[0m in \u001b[0;36m_non_cv_scores_importances\u001b[1;34m(self, X, y)\u001b[0m\n\u001b[0;32m    230\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m_non_cv_scores_importances\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    231\u001b[0m         \u001b[0mscore_func\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mpartial\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mscorer_\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwrapped_estimator_\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 232\u001b[1;33m         \u001b[0mbase_score\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mimportances\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_score_importances\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mscore_func\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    233\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mbase_score\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m*\u001b[0m \u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mimportances\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mimportances\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    234\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
 81 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\eli5\\sklearn\\permutation_importance.py\u001b[0m in \u001b[0;36m_get_score_importances\u001b[1;34m(self, score_func, X, y)\u001b[0m\n\u001b[0;32m    234\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    235\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0m_get_score_importances\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mscore_func\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 236\u001b[1;33m         return get_score_importances(score_func, X, y, n_iter=self.n_iter,\n\u001b[0m\u001b[0;32m    237\u001b[0m                                      random_state=self.rng_)\n\u001b[0;32m    238\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
 82 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\eli5\\permutation_importance.py\u001b[0m in \u001b[0;36mget_score_importances\u001b[1;34m(score_func, X, y, n_iter, columns_to_shuffle, random_state)\u001b[0m\n\u001b[0;32m     87\u001b[0m     \u001b[0mscores_decreases\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     88\u001b[0m     \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mn_iter\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 89\u001b[1;33m         scores_shuffled = _get_scores_shufled(\n\u001b[0m\u001b[0;32m     90\u001b[0m             \u001b[0mscore_func\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcolumns_to_shuffle\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mcolumns_to_shuffle\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     91\u001b[0m             \u001b[0mrandom_state\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mrng\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 83 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\eli5\\permutation_importance.py\u001b[0m in \u001b[0;36m_get_scores_shufled\u001b[1;34m(score_func, X, y, columns_to_shuffle, random_state)\u001b[0m\n\u001b[0;32m     98\u001b[0m                         random_state=None):\n\u001b[0;32m     99\u001b[0m     \u001b[0mXs\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0miter_shuffled\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcolumns_to_shuffle\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mrandom_state\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mrandom_state\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 100\u001b[1;33m     \u001b[1;32mreturn\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0marray\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mscore_func\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX_shuffled\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mX_shuffled\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mXs\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
 84 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\eli5\\permutation_importance.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m     98\u001b[0m                         random_state=None):\n\u001b[0;32m     99\u001b[0m     \u001b[0mXs\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0miter_shuffled\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcolumns_to_shuffle\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mrandom_state\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mrandom_state\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 100\u001b[1;33m     \u001b[1;32mreturn\u001b[0m \u001b[0mnp\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0marray\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mscore_func\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX_shuffled\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mX_shuffled\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mXs\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
 85 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\eli5\\sklearn\\permutation_importance.py\u001b[0m in \u001b[0;36mpd_scorer\u001b[1;34m(model, X, y)\u001b[0m\n\u001b[0;32m    160\u001b[0m         \u001b[1;32mdef\u001b[0m \u001b[0mpd_scorer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    161\u001b[0m             \u001b[0mX\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDataFrame\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcolumns\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mpd_columns\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 162\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mbase_scorer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    163\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mpd_scorer\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    164\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
 86 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\sklearn\\metrics\\_scorer.py\u001b[0m in \u001b[0;36m__call__\u001b[1;34m(self, estimator, X, y_true, sample_weight)\u001b[0m\n\u001b[0;32m    214\u001b[0m             \u001b[0mScore\u001b[0m \u001b[0mfunction\u001b[0m \u001b[0mapplied\u001b[0m \u001b[0mto\u001b[0m \u001b[0mprediction\u001b[0m \u001b[0mof\u001b[0m \u001b[0mestimator\u001b[0m \u001b[0mon\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    215\u001b[0m         \"\"\"\n\u001b[1;32m--> 216\u001b[1;33m         return self._score(\n\u001b[0m\u001b[0;32m    217\u001b[0m             \u001b[0mpartial\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0m_cached_call\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    218\u001b[0m             \u001b[0mestimator\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 87 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\sklearn\\metrics\\_scorer.py\u001b[0m in \u001b[0;36m_score\u001b[1;34m(self, method_caller, estimator, X, y_true, sample_weight)\u001b[0m\n\u001b[0;32m    256\u001b[0m         \"\"\"\n\u001b[0;32m    257\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 258\u001b[1;33m         \u001b[0my_pred\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmethod_caller\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mestimator\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"predict\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    259\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0msample_weight\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    260\u001b[0m             return self._sign * self._score_func(\n",
 88 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\sklearn\\metrics\\_scorer.py\u001b[0m in \u001b[0;36m_cached_call\u001b[1;34m(cache, estimator, method, *args, **kwargs)\u001b[0m\n\u001b[0;32m     66\u001b[0m     \u001b[1;34m\"\"\"Call estimator with method and args and kwargs.\"\"\"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     67\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mcache\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 68\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mestimator\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmethod\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     69\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     70\u001b[0m     \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 89 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\keras\\utils\\traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[1;34m(*args, **kwargs)\u001b[0m\n\u001b[0;32m     62\u001b[0m     \u001b[0mfiltered_tb\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     63\u001b[0m     \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 64\u001b[1;33m       \u001b[1;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     65\u001b[0m     \u001b[1;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# pylint: disable=broad-except\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     66\u001b[0m       \u001b[0mfiltered_tb\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0m_process_traceback_frames\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0me\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m__traceback__\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 90 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\keras\\engine\\training.py\u001b[0m in \u001b[0;36mpredict\u001b[1;34m(self, x, batch_size, verbose, steps, callbacks, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[0;32m   1783\u001b[0m       \u001b[0mcallbacks\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mon_predict_begin\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1784\u001b[0m       \u001b[0mbatch_outputs\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1785\u001b[1;33m       \u001b[1;32mfor\u001b[0m \u001b[0m_\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0miterator\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0menumerate_epochs\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# Single epoch.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1786\u001b[0m         \u001b[1;32mwith\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcatch_stop_iteration\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1787\u001b[0m           \u001b[1;32mfor\u001b[0m \u001b[0mstep\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msteps\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 91 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\keras\\engine\\data_adapter.py\u001b[0m in \u001b[0;36menumerate_epochs\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m   1193\u001b[0m     \u001b[1;34m\"\"\"Yields `(epoch, tf.data.Iterator)`.\"\"\"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1194\u001b[0m     \u001b[1;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_truncate_execution_to_epoch\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1195\u001b[1;33m       \u001b[0mdata_iterator\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0miter\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_dataset\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1196\u001b[0m       \u001b[1;32mfor\u001b[0m \u001b[0mepoch\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_initial_epoch\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_epochs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1197\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_insufficient_data\u001b[0m\u001b[1;33m:\u001b[0m  \u001b[1;31m# Set by `catch_stop_iteration`.\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 92 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\tensorflow\\python\\data\\ops\\dataset_ops.py\u001b[0m in \u001b[0;36m__iter__\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    488\u001b[0m     \u001b[1;32mif\u001b[0m \u001b[0mcontext\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mexecuting_eagerly\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mor\u001b[0m \u001b[0mops\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0minside_function\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    489\u001b[0m       \u001b[1;32mwith\u001b[0m \u001b[0mops\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolocate_with\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_variant_tensor\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 490\u001b[1;33m         \u001b[1;32mreturn\u001b[0m \u001b[0miterator_ops\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mOwnedIterator\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    491\u001b[0m     \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    492\u001b[0m       raise RuntimeError(\"`tf.data.Dataset` only supports Python-style \"\n",
 93 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\tensorflow\\python\\data\\ops\\iterator_ops.py\u001b[0m in \u001b[0;36m__init__\u001b[1;34m(self, dataset, components, element_spec)\u001b[0m\n\u001b[0;32m    724\u001b[0m             \u001b[1;34m\"When `dataset` is provided, `element_spec` and `components` must \"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    725\u001b[0m             \"not be specified.\")\n\u001b[1;32m--> 726\u001b[1;33m       \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_create_iterator\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdataset\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    727\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    728\u001b[0m     \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_next_call_count\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m0\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 94 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\tensorflow\\python\\data\\ops\\iterator_ops.py\u001b[0m in \u001b[0;36m_create_iterator\u001b[1;34m(self, dataset)\u001b[0m\n\u001b[0;32m    749\u001b[0m               \u001b[0moutput_types\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_flat_output_types\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    750\u001b[0m               output_shapes=self._flat_output_shapes))\n\u001b[1;32m--> 751\u001b[1;33m       \u001b[0mgen_dataset_ops\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmake_iterator\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mds_variant\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_iterator_resource\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    752\u001b[0m       \u001b[1;31m# Delete the resource when this object is deleted\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    753\u001b[0m       self._resource_deleter = IteratorResourceDeleter(\n",
 95 |       "\u001b[1;32mF:\\pythonenv\\dl\\lib\\site-packages\\tensorflow\\python\\ops\\gen_dataset_ops.py\u001b[0m in \u001b[0;36mmake_iterator\u001b[1;34m(dataset, iterator, name)\u001b[0m\n\u001b[0;32m   3234\u001b[0m   \u001b[1;32mif\u001b[0m \u001b[0mtld\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mis_eager\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3235\u001b[0m     \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3236\u001b[1;33m       _result = pywrap_tfe.TFE_Py_FastPathExecute(\n\u001b[0m\u001b[0;32m   3237\u001b[0m         _ctx, \"MakeIterator\", name, dataset, iterator)\n\u001b[0;32m   3238\u001b[0m       \u001b[1;32mreturn\u001b[0m \u001b[0m_result\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
 96 |       "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
 97 |      ]
 98 |     }
 99 |    ],
100 |    "source": [
101 |     "import eli5\n",
102 |     "from eli5.sklearn import PermutationImportance\n",
103 |     "i='外向性'\n",
104 |     "y_temp = y.loc[:,[i]]\n",
105 |     "x_train,x_test,y_train,y_test = train_test_split(x,y_temp,test_size=0.15,random_state=1)\n",
106 |     "model = keras.models.load_model('models\\\\'+i+'.h5')\n",
107 |     "perm = PermutationImportance(model, scoring='neg_mean_squared_error',random_state=1).fit(x_test, y_test)\n",
108 |     "eli5.show_weights(perm, feature_names = x_test.columns.tolist())\n"
109 |    ]
110 |   }
111 |  ],
112 |  "metadata": {
113 |   "interpreter": {
114 |    "hash": "fb662769bf5854fc00024041705291ada44874f0671c421db64edbd11228109f"
115 |   },
116 |   "kernelspec": {
117 |    "display_name": "Python 3.8.8 ('dl')",
118 |    "language": "python",
119 |    "name": "python3"
120 |   },
121 |   "language_info": {
122 |    "codemirror_mode": {
123 |     "name": "ipython",
124 |     "version": 3
125 |    },
126 |    "file_extension": ".py",
127 |    "mimetype": "text/x-python",
128 |    "name": "python",
129 |    "nbconvert_exporter": "python",
130 |    "pygments_lexer": "ipython3",
131 |    "version": "3.8.8"
132 |   },
133 |   "orig_nbformat": 4
134 |  },
135 |  "nbformat": 4,
136 |  "nbformat_minor": 2
137 | }
138 | 


--------------------------------------------------------------------------------