BNeuralT/evaluate_MLP_class_reg_TF.py at master · vojha-code/BNeuralT · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
# -*- coding: utf-8 -*-
"""
Created on Sun Mar 22 14:15:06 2020

@author: Anonymus
"""

import os
import pandas as pd
import numpy as np
from random import randint

# save np.load
np_load_old = np.load

# modify the default parameters of np.load
np.load = lambda *a,**k: np_load_old(*a, allow_pickle=True, **k)

#%%
sfolder_Check = os.path.join(r'C:\Users\yl918888\Desktop\BNeuralT\trained_models\traind_results_keras_TF_sklern\MLP_experiments')
resfolder_Check = os.listdir(sfolder_Check)
plotsPath = os.path.join(r"C:\Users\yl918888\Desktop\BNeuralT\trained_models")

#exp	data	algo	error_trn	error_tst	corr_trn	corr_tst	r2_trn	r2_tst	tree_size	func_node	leaf_node	train_param	compute_time	problem

#%%
listExp = []
listData = []
listAlgo = []
listErrorTrn = []
listErrorTst = []

listR2Trn = []
listR2Tst = []

listPram = []
listTime = []
listProblem = []

listNull = []


count = 0
for fileName in  resfolder_Check:
    lststr = fileName.replace('.', '_').split('B_1_')

    if (
        lststr[1] == 'sigmoid_ES_No_Reg_No_npy' or
        lststr[1] == 'sigmoid_ES_No_Reg_No_defopt_npy' or
        lststr[1] == 'relu_ES_No_Reg_No_npy' or
        lststr[1] == 'relu_ES_No_Reg_No_defopt_npy' or
        lststr[1] == 'sigmoid_ES_No_Reg_l1_l2_npy' or
        lststr[1] == 'sigmoid_ES_No_Reg_l1_l2_defopt_npy' or
        lststr[1] == 'sigmoid_ES_50_Reg_No__npy' or
        lststr[1] == 'sigmoid_ES_50_Reg_No_defopt__npy'
        ):
        print(fileName)

        count +=1


        path_performance = os.path.join(sfolder_Check,fileName)
        read_np = np.load(path_performance).item()
        for key in read_np:
            if(not (key == 'model_config')):
                keyStrList = key.split('_')
                dataV = keyStrList[0]
                algoV = keyStrList[2]
                keyVal = read_np[key]

                listExp.append('MLP_'+lststr[1])
                listData.append(dataV)
                listAlgo.append(algoV)

                if len(keyVal) == 7:
                    listProblem.append("reg")

                    listErrorTrn.append(keyVal[0]) # keyVal[0] has mse trn
                    listErrorTst.append(keyVal[1]) # keyVal[1] has mse tst

                    listR2Trn.append(keyVal[2]) #  keyVal[2] has accuracy R2 trn
                    listR2Tst.append(keyVal[3]) #  keyVal[3] has accuracy R2 tst

                    listPram.append(keyVal[4])  # trainable paramters
                    listTime.append(keyVal[5])  # compute time
                else:
                    listProblem.append("class")

                    listErrorTrn.append(1.0 - keyVal[0]) # keyVal[0] has accuracy trn
                    listErrorTst.append(1.0 - keyVal[1]) # keyVal[1] has accuracy tst

                    listR2Trn.append(keyVal[0]) #  keyVal[0] has accuracy trn
                    listR2Tst.append(keyVal[1]) #  keyVal[1] has accuracy trn

                    listPram.append(keyVal[2]) # trainable paramters
                    listTime.append(keyVal[3]) # compute time
                #common null
                listNull.append('')

print('total expirments',(8*14),'=',count)
#%%
# exp	data	algo
# error_trn	error_tst	corr_trn	corr_tst	r2_trn	r2_tst
# tree_size	func_node	leaf_node	train_param	compute_time
# problem


df_performance = pd.DataFrame({'A' : []})

df_performance['exp'] = listExp
df_performance['data'] = listData
df_performance['algo'] = listAlgo
# performance
df_performance['error_trn'] = listErrorTrn
df_performance['error_tst'] = listErrorTst
df_performance['corr_trn'] = listNull # null
df_performance['corr_tst'] = listNull # null
df_performance['r2_trn'] = listR2Trn
df_performance['r2_tst'] = listR2Tst
# other params
df_performance['tree_size'] = listNull #null
df_performance['func_node'] = listNull # null
df_performance['leaf_node'] = listNull # null
df_performance['train_param'] = listPram
df_performance['compute_time'] = listTime
# problem
df_performance['problem'] = listProblem

del df_performance['A']
#%%◘
df_performance.to_csv(os.path.join(plotsPath,'MLP_All_results.csv'),index=False)
print('data saved')