/
show_results.py
46 lines (34 loc) · 1.71 KB
/
show_results.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
from utils.lib import record_JSON
import pdb
def write_to(filename, lines):
with open(filename, 'w') as file:
file.write(lines)
accuracy_record = record_JSON()
accuracy_record.load()
dataset_name = 'CIFAR10'
model_name = 'efficientnet'
combined_class=True
batch_sizes = ['single', 'batch_8', 'batch_16', 'batch', 'batch_64', 'batch_128', 'batch_256']
model_train_modes = [0,2,3,5,10]
filename = 'accuracy_record_'+dataset_name+'.csv'
accuracy_record.print_all()
lines = 'training_mode, single, batch_8, batch_16, batch_32, batch_64, batch_128, batch_256'
for model_train_mode in model_train_modes:
lines += f'\n{model_train_mode}, '
for batch_size in batch_sizes:
accuracy,idx = accuracy_record.lookup( combined_class=combined_class,
model_name = model_name ,
model_train_mode = model_train_mode,
batch_size = batch_size,
dataset_name = dataset_name)
# accuracy = accuracy_record.lookup( combined_class=True, model_name = 'efficientnet', model_train_mode = 2, batch_size = 'single', dataset_name = dataset_name)
lines += f'{accuracy}, '
# if model_train_mode == 5:
# match = accuracy_record.match (model_name, combined_class, model_train_mode, 'batch', dataset_name)
# df = accuracy_record.get_df()
# accuracy = df[match]['accuracy'].values[-1]
# print(f'model_name = {model_name}, combined_class={combined_class}, model_train_mode= {model_train_mode}, batch_size={batch_size},dataset_name= {dataset_name},match = {match}')
# print(accuracy)
# pdb.set_trace()
print (lines)
write_to(filename, lines)