-
Notifications
You must be signed in to change notification settings - Fork 0
/
predict_gen.py
188 lines (154 loc) · 6.54 KB
/
predict_gen.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
'''
# Reconstruct image and evalute the performance by Generalization
# Author: Yuki Saeki
# Reference: "https://github.com/Silver-L/beta-VAE"
'''
import tensorflow as tf
import os
import argparse
import numpy as np
import SimpleITK as sitk
from tqdm import tqdm
import csv
import dataIO as io
from network import *
from model import Variational_Autoencoder
import matplotlib.pyplot as plt
import matplotlib.cm as cm
import utils
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' #for windows
def main():
# tf flag
flags = tf.flags
flags.DEFINE_string("test_data_txt", "E:/git/beta-VAE/input/CT/shift/test.txt", "i1")
flags.DEFINE_string("model1", 'D:/vae_result/n1+n2/all/fine/beta_1/model/model_{}'.format(198500), "i2")
flags.DEFINE_string("model2", 'D:/vae_result/n1+n2/all/fine/beta_1/model2/model_{}'.format(198500), "i3")
flags.DEFINE_string("outdir", "D:/vae_result/n1+n2/all/fine/beta_1/gen/", "i4")
flags.DEFINE_float("beta", 1, "hyperparameter beta")
flags.DEFINE_integer("num_of_test", 600, "number of test data")
flags.DEFINE_integer("batch_size", 1, "batch size")
flags.DEFINE_integer("latent_dim", 6, "latent dim")
flags.DEFINE_list("image_size", [9*9*9], "image size")
FLAGS = flags.FLAGS
# check folder
if not (os.path.exists(FLAGS.outdir)):
os.makedirs(FLAGS.outdir + 'ori/')
os.makedirs(FLAGS.outdir + 'preds/')
os.makedirs(FLAGS.outdir + 'rec/')
# read list
test_data_list = io.load_list(FLAGS.test_data_txt)
# test step
test_step = FLAGS.num_of_test // FLAGS.batch_size
if FLAGS.num_of_test % FLAGS.batch_size != 0:
test_step += 1
# load test data
test_set = tf.data.TFRecordDataset(test_data_list)
test_set = test_set.map(lambda x: _parse_function(x, image_size=FLAGS.image_size),
num_parallel_calls=os.cpu_count())
test_set = test_set.batch(FLAGS.batch_size)
test_iter = test_set.make_one_shot_iterator()
test_data = test_iter.get_next()
# initializer
init_op = tf.group(tf.initializers.global_variables(),
tf.initializers.local_variables())
with tf.Session(config = utils.config) as sess:
sess.run(init_op)
# set network
kwargs = {
'sess': sess,
'outdir': FLAGS.outdir,
'beta': FLAGS.beta,
'latent_dim': FLAGS.latent_dim,
'batch_size': FLAGS.batch_size,
'image_size': FLAGS.image_size,
'encoder': encoder_mlp,
'decoder': decoder_mlp,
'is_res': False
}
VAE = Variational_Autoencoder(**kwargs)
kwargs_2 = {
'sess': sess,
'outdir': FLAGS.outdir,
'beta': FLAGS.beta,
'latent_dim': 8,
'batch_size': FLAGS.batch_size,
'image_size': FLAGS.image_size,
'encoder': encoder_mlp2,
'decoder': decoder_mlp_tanh,
'is_res': True,
'is_constraints': False
}
VAE_2 = Variational_Autoencoder(**kwargs_2)
# testing
VAE.restore_model(FLAGS.model1)
VAE_2.restore_model(FLAGS.model2)
tbar = tqdm(range(test_step), ascii=True)
preds = []
ori = []
rec = []
for k in tbar:
test_data_batch = sess.run(test_data)
ori_single = test_data_batch
preds_single = VAE.reconstruction_image(ori_single)
rec_single = VAE_2.reconstruction_image2(ori_single, preds_single)
preds_single = preds_single[0, :]
ori_single = ori_single[0, :]
rec_single = rec_single[0, :]
preds.append(preds_single)
ori.append(ori_single)
rec.append(rec_single)
patch_side = 9
preds = np.reshape(preds, [FLAGS.num_of_test, patch_side, patch_side, patch_side])
ori = np.reshape(ori, [FLAGS.num_of_test, patch_side, patch_side, patch_side])
rec = np.reshape(rec, [FLAGS.num_of_test, patch_side, patch_side, patch_side])
# label
generalization_single = []
file_ori = open(FLAGS.outdir + 'ori/list.txt', 'w')
file_preds = open(FLAGS.outdir + 'preds/list.txt', 'w')
file_rec = open(FLAGS.outdir + 'rec/list.txt', 'w')
for j in range(len(preds)):
# EUDT
ori_image = sitk.GetImageFromArray(ori[j])
ori_image.SetOrigin([0, 0, 0])
ori_image.SetSpacing([0.885,0.885,1])
preds_image = sitk.GetImageFromArray(preds[j])
preds_image.SetOrigin([0, 0, 0])
preds_image.SetSpacing([0.885,0.885,1])
rec_image = sitk.GetImageFromArray(rec[j])
rec_image.SetOrigin([0, 0, 0])
rec_image.SetSpacing([0.885,0.885,1])
# output image
io.write_mhd_and_raw(ori_image, '{}.mhd'.format(os.path.join(FLAGS.outdir, 'ori','ori_{}'.format(j + 1))))
io.write_mhd_and_raw(preds_image, '{}.mhd'.format(os.path.join(FLAGS.outdir, 'preds','preds_{}'.format(j + 1))))
io.write_mhd_and_raw(rec_image, '{}.mhd'.format(os.path.join(FLAGS.outdir, 'rec', 'rec_{}'.format(j + 1))))
file_ori.write('{}.mhd'.format(os.path.join(FLAGS.outdir, 'ori', 'ori_{}'.format(j + 1))) + "/n")
file_rec.write('{}.mhd'.format(os.path.join(FLAGS.outdir, 'preds', 'preds_{}'.format(j + 1))) + "/n")
file_rec.write('{}.mhd'.format(os.path.join(FLAGS.outdir, 'rec', 'rec_{}'.format(j + 1))) + "/n")
generalization_single.append(utils.L1norm(ori[j], rec[j]))
file_ori.close()
file_preds.close()
file_rec.close()
generalization = np.average(generalization_single)
print('generalization = %f' % generalization)
np.savetxt(os.path.join(FLAGS.outdir, 'generalization.csv'), generalization_single, delimiter=",")
# plot reconstruction
a_X = ori[:, 4, :]
a_Xe = rec[:, 4, :]
c_X = ori[:, :, 4, :]
c_Xe = rec[:, :, 4, :]
s_X = ori[:, :, :, 4]
s_Xe = rec[:, :, :, 4]
utils.visualize_slices(a_X, a_Xe, FLAGS.outdir + "axial_")
utils.visualize_slices(c_X, c_Xe, FLAGS.outdir + "coronal_")
utils.visualize_slices(s_X, s_Xe, FLAGS.outdir + "sagital_")
# # load tfrecord function
def _parse_function(record, image_size=[9 * 9 * 9]):
keys_to_features = {
'img_raw': tf.FixedLenFeature(np.prod(image_size), tf.float32),
}
parsed_features = tf.parse_single_example(record, keys_to_features)
image = parsed_features['img_raw']
image = tf.reshape(image, image_size)
return image
if __name__ == '__main__':
main()