Transliteration-using-RNNs/plot_loss.py at master · SahanaRamnath/Transliteration-using-RNNs · 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
'''
------------------------------------
PLOT LOSS GRAPH
------------------------------------
'''

import pickle
import matplotlib.pyplot as plt
import argparse
import os

parser=argparse.ArgumentParser()
parser.add_argument("--save_dir",help="directory to load data from",required=True)
args=parser.parse_args()

dir_name=args.save_dir

###################################################################################
# loss and accuracy
with open(os.path.join(dir_name,'train_loss_list.pkl')) as f :
	[epoch_list_train,train_loss_list]=pickle.load(f)

with open(os.path.join(dir_name,'val_loss_list.pkl')) as f :
	[epoch_list_val,val_loss_list]=pickle.load(f)

fig1=plt.figure().add_subplot(111)
fig1.plot(epoch_list_train,train_loss_list)
fig1.plot(epoch_list_val,val_loss_list)
plt.legend(('train loss','val loss'),loc='best')
plt.title('Loss versus Epoch')
plt.xlabel('Epoch')
plt.ylabel('Loss')
plt.savefig('plots/loss_plot_'+dir_name+'.png')

with open(os.path.join(dir_name,'val_acc_list.pkl')) as f :
	[epoch_list_val,val_acc_list]=pickle.load(f)

# fig2=plt.figure().add_subplot(111)
# fig2.plot(epoch_list_train,val_acc_list)
# fig2.plot(epoch_list_val,val_loss_list)
# plt.legend(('val accuracy','val loss'),loc='best')
# plt.title('Validation loss and accuracy versus Epoch')
# plt.xlabel('Epoch')
# plt.ylabel('Loss/Accuracy')
# plt.savefig('plots/val_plot_'+dir_name+'.png')

fig2=plt.figure().add_subplot(111)
fig2.plot(epoch_list_train,val_acc_list)
plt.title('Validation Accuracy versus Epoch')
plt.xlabel('Epoch')
plt.ylabel('Val Accuracy')
plt.savefig('plots/val_plot_'+dir_name+'.png')
#####################################################################################
# hparams
epoch_list=[]
with open(os.path.join(dir_name,'val_acc_list1.pkl')) as f :
	[epoch_list1,val_acc_list1]=pickle.load(f)
if len(epoch_list1)>len(epoch_list) :
	epoch_list=epoch_list1

with open(os.path.join(dir_name,'val_acc_list2.pkl')) as f :
	[epoch_list2,val_acc_list2]=pickle.load(f)
if len(epoch_list2)>len(epoch_list) :
	epoch_list=epoch_list2

with open(os.path.join(dir_name,'val_acc_list3.pkl')) as f :
	[epoch_list3,val_acc_list3]=pickle.load(f)
if len(epoch_list3)>len(epoch_list) :
	epoch_list=epoch_list3

with open(os.path.join(dir_name,'val_acc_list4.pkl')) as f :
	[epoch_list4,val_acc_list4]=pickle.load(f)
if len(epoch_list4)>len(epoch_list) :
	epoch_list=epoch_list4

fig3=plt.figure().add_subplot(111)
fig3.plot(epoch_list[:len(val_acc_list1)],val_acc_list1)
fig3.plot(epoch_list[:len(val_acc_list2)],val_acc_list2)
# fig3.plot(epoch_list[:len(val_acc_list3)],val_acc_list3)
# fig3.plot(epoch_list[:len(val_acc_list4)],val_acc_list4)
plt.legend(('xavier','random uniform'),loc='best')
plt.title('Validation Accuracy versus Epoch')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.savefig('plots/hparams_plot_init.png')