-
Notifications
You must be signed in to change notification settings - Fork 0
/
plot.py
267 lines (201 loc) · 9.42 KB
/
plot.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
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
import matplotlib.pyplot as plt
import numpy as np
from PIL import Image, ImageOps
import glob
from os import listdir
from os.path import join, splitext
def plot_PSNRs_SSIMs(data_dirs):
PSNR_epochs = []
SSIMs_epochs = []
PSNRs = []
SSIMs = []
for data_dir in data_dirs:
PSNR_log_name = data_dir + 'PSNR_log.txt'
PSNR_log = np.loadtxt(PSNR_log_name)
PSNR_epochs_list = list(PSNR_log[:, 0])
PSNRs_list = list(PSNR_log[:, 1])
PSNR_epochs += [PSNR_epochs_list]
PSNRs += [PSNRs_list]
SSIM_log_name = data_dir + 'SSIM_log.txt'
SSIM_log = np.loadtxt(SSIM_log_name)
SSIMs_epochs_list = list(SSIM_log[:, 0])
SSIMs_list = list(SSIM_log[:, 1])
SSIMs_epochs += [SSIMs_epochs_list]
SSIMs += [SSIMs_list]
plt.figure()
plt.subplots_adjust(wspace =0.08, hspace =0.5)
plt.subplot(2, 1, 1)
plt.title('PSNR Results')
PSNR_unet = plt.plot(PSNR_epochs[0], PSNRs[0], color='green', label='unet lr=0.0002')
PSNR_unet0005 = plt.plot(PSNR_epochs[2], PSNRs[2], color='red', label='unet lr=0.0005')
PSNR_unet_noDecay = plt.plot(PSNR_epochs[3], PSNRs[3], color='black', label='unet lr=0.0002 no decay')
PSNR_resnet = plt.plot(PSNR_epochs[1], PSNRs[1], color='blue', label='resnet_9blocks lr=0.0002')
# plt.legend()
plt.xlabel('epoch')
plt.ylabel('PSNR')
plt.subplot(2, 1, 2)
plt.title('SSIM Results')
SSIM_unet = plt.plot(SSIMs_epochs[0], SSIMs[0], color='green', label='unet lr=0.0002')
SSIM_unet0005 = plt.plot(SSIMs_epochs[2], SSIMs[2], color='red', label='unet lr=0.0005')
SSIM_unet_noDecay = plt.plot(SSIMs_epochs[3], SSIMs[3], color='black', label='unet lr=0.0002 no decay')
SSIM_resnet = plt.plot(SSIMs_epochs[1], SSIMs[1], color='blue', label='resnet_9blocks lr=0.0002')
plt.legend()
plt.xlabel('epoch')
plt.ylabel('SSIM')
plt.show()
def plot_Loss(data_dirs):
Loss_epochs = []
iterations = []
D_Losses = []
GAN_Losses = []
L1_Losses = []
for data_dir in data_dirs:
Loss_log_name = data_dir + 'Loss_log.txt'
Loss_log = np.loadtxt(Loss_log_name)
Loss_epochs_list = list(Loss_log[:, 0])
Losses_iterations_list = list(Loss_log[:, 1])
Losses_D_list = list(Loss_log[:, 2])
Losses_GAN_list = list(Loss_log[:, 3])
Losses_L1_list = list(Loss_log[:, 4])
total_iterations_array = Loss_log[:, 0:2]
# print(total_iterations_array.shape[0])
# print(Loss_epochs_list)
# calculate iterations
total_train_data = 3300 #22509
Losses_iterations = np.zeros(total_iterations_array.shape[0])
for i in range(total_iterations_array.shape[0]):
Losses_iterations[i] = (total_iterations_array[i, 0] - 1) * total_train_data + total_iterations_array[i, 1]
# print(Losses_iterations)
# print(Losses_iterations.shape)
Losses_iterations_list = list(Losses_iterations)
# print(len(Losses_iterations_list))
# a
# PSNR_epochs += [PSNR_epochs_list]
# PSNRs += [PSNRs_list]
iterations += [Losses_iterations_list]
D_Losses += [Losses_D_list]
GAN_Losses += [Losses_GAN_list]
L1_Losses += [Losses_L1_list]
# print(iterations[0])
# print(D_Losses[0])
plt.figure()
plt.subplots_adjust(wspace =0.08, hspace =0.5)
plt.subplot(2, 1, 1)
plt.title('Loss Results')
plt.plot(iterations[0], L1_Losses[0], color='red', label='L1_Loss')
plt.plot(iterations[0], GAN_Losses[0], color='blue', label='GAN_Loss')
plt.plot(iterations[0], D_Losses[0], color='green', label='D_Loss')
plt.xlim(0, 100000)
plt.ylim(0, 40)
plt.legend()
# plt.xlabel('iteration')
plt.ylabel('Loss')
plt.subplot(2, 1, 2)
# plt.title('Loss Results')
# plt.plot(iterations[1], D_Losses[1], color='green', label='D_Losses')
# plt.plot(iterations[1], GAN_Losses[1], color='blue', label='GAN_Losses')
# plt.plot(iterations[1], L1_Losses[1], color='red', label='L1_Losses')
plt.plot(iterations[0], D_Losses[0], color='green', label='D_Loss')
plt.plot(iterations[0], GAN_Losses[0], color='blue', label='GAN_Loss')
# plt.plot(iterations[0], L1_Losses[0], color='red', label='L1_Losses')
plt.xlim(0, 100000)
plt.ylim(0, 1.75)
# plt.legend()
plt.xlabel('iteration')
plt.ylabel('Loss')
plt.show()
def splice_epoch_images(image_dir):
# plt.figure(figsize=(16,8))
# plt.subplots_adjust(wspace =0.08, hspace =0.002)
# plt.axis('off')
# for num in range(1,10):
# filenum = num + 18
# img_name = "epoch_{}_predict_{}_A.png".format(2, filenum)
# # print(img_name)
# img = Image.open(image_dir+img_name)
# plt.subplot(1,9,num)
# plt.axis('off')
# # plt.title('img')
# plt.rcParams['savefig.dpi'] = (256, 256)
# plt.imshow(img)
# plt.show()
UNIT_SIZE = 369
TARGET_WIDTH = 369*9 # 拼接完后的横向长度
target = Image.new('RGB', (TARGET_WIDTH, UNIT_SIZE))
left = 0
right = UNIT_SIZE
epoch = 0
for num in range(1,10):
filenum = num + 18
# img_name = "epoch_{}_predict_{}_A.png".format(epoch, filenum)
img_name = "{}_A.png".format(filenum)
# print(img_name)
img = Image.open(image_dir+img_name)
img = img.resize((UNIT_SIZE, UNIT_SIZE),Image.ANTIALIAS)
print(img.size)
target.paste(img, (left, 0, right, UNIT_SIZE))# 将image复制到target的指定位置中
left += UNIT_SIZE # left是左上角的横坐标,依次递增
right += UNIT_SIZE # right是右下的横坐标,依次递增
quality_value = 100 # quality来指定生成图片的质量,范围是0~100
target.save(image_dir+"epoch_{}.png".format(epoch), quality = quality_value)
target.show()
# for n in range(1,11):
# epoch = n * 2
# for num in range(1,10):
# filenum = num + 18
# img_name = "epoch_{}_predict_{}_A.png".format(epoch, filenum)
# print(img_name)
# apple = Image.open(image_dir+img_name)
# plt.subplot(10,9,num)
# plt.axis('off')
# # plt.title('apple')
# plt.imshow(apple)
# plt.show()
def splice_test_images(image_dir, number_of_images, imageNameSave):
input_fullnames = listdir(image_dir)
# self.input_fullnames.sort(key = lambda fullname: int(splitext(fullname)[0]))
# input_fullnames.sort(key = lambda fullname: int(splitext(fullname)[0].split('_')[0]))
# print(input_fullnames)
UNIT_SIZE = 369
TARGET_WIDTH = 369*number_of_images # 拼接完后的横向长度
target = Image.new('RGB', (TARGET_WIDTH, UNIT_SIZE))
left = 0
right = UNIT_SIZE
epoch = 0
for image_name in input_fullnames:
print(image_name)
img = Image.open(image_dir+image_name)
img = img.resize((UNIT_SIZE, UNIT_SIZE),Image.ANTIALIAS)
# print(img.size)
target.paste(img, (left, 0, right, UNIT_SIZE))# 将image复制到target的指定位置中
left += UNIT_SIZE # left是左上角的横坐标,依次递增
right += UNIT_SIZE # right是右下的横坐标,依次递增
quality_value = 100 # quality来指定生成图片的质量,范围是0~100
target.save(imageNameSave)
target.show()
if __name__ == '__main__':
data_dirs = []
data_dirs += ['../pix2pixFiles/GPU_result/cGAN_deblur_simplified_unet/checkpoint/']
data_dirs += ['../pix2pixFiles/GPU_result/cGAN_deblur_simplified_resnet/checkpoint/']
data_dirs += ['../pix2pixFiles/GPU_result/cGAN_deblur_simplified_0005/checkpoint/']
data_dirs += ['../pix2pixFiles/GPU_result/cGAN_deblur_simplified_0002NoDecay/checkpoint/']
# data_dirs = []
# data_dirs += ['../pix2pixFiles/GPU_result/cGAN_depth_simplified_unet/checkpoint/']
# data_dirs += ['../pix2pixFiles/GPU_result/cGAN_depth_simplified_resnet/checkpoint/']
# plot_PSNRs_SSIMs(data_dirs)
data_dirs = ['../pix2pixFiles/GPU_result/cGAN_depth_simplified_unet/checkpoint/']
# plot_Loss(data_dirs)
splice_epoch_images_dir = './GPU_result/cGAN_deblur_simplified_unet/checkpoint/image_for_comparison/'
# splice_epoch_images(splice_epoch_images_dir)
splice_test_images_dir = './GPU_result/cGAN_depth_simplified_resnet/checkpoint/samples_predict_images/YOLO/'
# splice_test_images_dir = './test/input/YOLO/'
splice_test_images_dir = './depth_New_test_sample/netG_unet_256_epoch_80_80_lr_0002/YOLO/'
splice_test_images_dir = './GPU_result/cGAN_deblur_simplified_0005/checkpoint/visualSetSamples/YOLO/'
splice_test_images_dir = './GPU_result/cGAN_deblur_simplified_unet/checkpoint/image_for_comparison/originBlur/'
splice_test_images_dir = './GPU_result/cGAN_deblur_simplified_0005/checkpoint/visualSetSamples/YOLO/'
splice_test_images_dir = './test/netG_unet_256_epoch_10_10_lr_0005/YOLO/'
splice_test_images_dir = './depth_New_test_sample/netG_resnet_9blocks_epoch_80_80_lr_0002/YOLO/'
splice_test_images_dir = '../pix2pixFiles/GPU_result/cGAN_depth_simplified_unet/checkpoint/epoch_predict_images/epoch160/YOLO/'
splice_test_images_dir = '../pix2pixFiles/GPU_result/cGAN_deblur_simplified_0002NoDecay/checkpoint/visualSetSamples/YOLO/'
splice_test_images_dir = '../pix2pixFiles/test_NewBlur/netG_unet_256_epoch_20_0_lr_0002/YOLO/'
splice_test_images(splice_test_images_dir, 9, 'netG_unet_256_epoch_20_0_lr_0002.png')