-
Notifications
You must be signed in to change notification settings - Fork 0
/
open_close_principle.py
152 lines (130 loc) · 5.18 KB
/
open_close_principle.py
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
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
import os
import pickle
from typing import Tuple
import matplotlib.pyplot as plt
from matplotlib.axes._subplots import Axes
import re
import warnings
warnings.filterwarnings("ignore", category=DeprecationWarning)
def _get_exp_and_histories(exp_keys: list) -> dict:
"""
根據輸入的實驗名稱，透過`get_histories`函數拿取相應的fold.pkl
Args:
exp_keys (list): 輸入的實驗名稱list
Returns:
dict : 輸出的實驗名稱 & fold_1.pkl, fold_2.pkl, fold_3.pkl pair
"""
exp_collection = {}
for k in exp_keys:
exp_collection[k] = _get_histories(k)
return exp_collection
def _get_histories(exp_name: str) -> dict:
"""
根據輸入的實驗名稱拿取對應的history
Args:
exp_name (str): 輸入的實驗名稱
Returns:
dict : 輸出的實驗名稱 & fold_1.pkl, fold_2.pkl, fold_3.pkl pair
"""
exp_histroy_pair = {}
for n in range(1, 4):
fold_path = os.path.join('experiments', exp_name, f'fold_{n}.pkl')
with open(fold_path, 'rb') as save_history:
exp_histroy_pair[f'fold_{n}'] = pickle.load(save_history)
return exp_histroy_pair
def plot_history(history: dict,
lbl: str,
ax1: Axes,
ax2: Axes,
color: str) -> Tuple[Axes, Axes]:
"""
根據給定的history畫圖，左為loss vs epoch，右為accuracy vs epoch
Args:
history (dict): keras中的history dict
lbl (str): 該history對應的legend
ax1 (Axes) 物件，畫線圖 (loss vs epoch)
ax2 (Axes) 物件，畫線圖 (accuracy vs epoch)
color (str) : 該history對應的顏色
Returns:
Tuple[Axes, Axes] : 畫好線的Axes物件
"""
acc_keyword = ['acc', 'accuracy']
val_acc_keyword = ['val_acc', 'val_accuracy']
ax1.plot(history['loss'],
label=f'{lbl}_train', c=color)
ax1.plot(history['val_loss'], label=f'{lbl}_val',
c=color, linestyle='--')
ax1.set_ylabel('Loss')
ax1.set_xlabel('Epoch')
######### [START] for differenct keras version acc keywords #############
try:
# 'acc' for keras 2.1.6
ax2.plot(history[acc_keyword[0]],
label=f'{lbl}_train', c=color)
ax2.plot(history[val_acc_keyword[0]], label=f'{lbl}_val',
c=color, linestyle='--')
except KeyError:
# 'accuracy' for keras > 2.1.6
ax2.plot(history[acc_keyword[1]],
label=f'{lbl}_train', c=color)
ax2.plot(history[val_acc_keyword[1]], label=f'{lbl}_val',
c=color, linestyle='--')
######### [END] for differenct keras version acc keywords #############
ax2.set_ylabel('accuracy')
ax2.set_xlabel('Epoch')
return ax1, ax2
def _extract_label_keyword_func(exp_name: str, extract_pattern: re.Pattern) -> str:
"""
Args:
exp_name (str): 輸入的實驗名稱
extract_pattern(re.Pattern) : 擷取標籤的正則表達式
Returns:
str: 輸出的histroy plot label
"""
return extract_pattern.search(exp_name).group()
def main(pattern_to_extract_label: re.Pattern):
display_exp = sorted(os.listdir('experiments'))
exp_s = "\n".join(display_exp)
print(f'all experiments are : {exp_s}')
exp_keys = []
keep_asking = True
while keep_asking:
exp_name = input("enter experiments names, if end, key [end]")
if exp_name == 'end':
keep_asking = False
else:
exp_keys.append(exp_name)
print(exp_keys)
exp_collections = _get_exp_and_histories(exp_keys=exp_keys)
colors = 'bgrcmykw'[:len(exp_collections.keys())]
with plt.style.context(['science', 'grid', 'no-latex']):
fig, axes = plt.subplots(nrows=3, ncols=2, figsize=(3, 4), dpi=150)
for color_i, (exp_name, all_fold_d) in enumerate(exp_collections.items()):
color = colors[color_i]
for fold_i, history in all_fold_d.items():
if fold_i == 'fold_1':
row = 0
elif fold_i == 'fold_2':
row = 1
elif fold_i == 'fold_3':
row = 2
else:
raise ValueError('should be 3 fold in this experiments')
short_exp_name = _extract_label_keyword_func(exp_name,
pattern_to_extract_label)
axes[row][0], axes[row][1] = plot_history(history=history,
ax1=axes[row][0], ax2=axes[row][1],
lbl=short_exp_name,
color=color)
handles, labels = axes[0][0].get_legend_handles_labels()
fig.legend(handles, labels,
loc='upper center',
ncol=len(exp_collections.keys()),
shadow=True
)
fig.tight_layout(pad=0.4, w_pad=0.5, h_pad=0.5)
fig.subplots_adjust(top=0.85)
plt.show()
if __name__ == "__main__":
pattern_to_extract_label = re.compile(r'dataset_1_(aug_[0-9]+)')
main(pattern_to_extract_label)