gan-exp/myutils.py at master · whyguu/gan-exp · 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
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
import shutil
import numpy as np
import os
import multiprocessing
from tqdm import tqdm
import tensorflow as tf
import glob
from skimage import io
import warnings


def fake_result(samples):
    pid = os.getpid()
    print("pid :", pid)
    path = '../nas/SRAD2018/SRAD2018_Test_1/'
    for sample in tqdm(samples):
        # print(sample)
        os.mkdir('../nas/SRAD2018Model/result/fake_rlt/'+sample)

        for i in range(1, 7):
            suffix = '_f00{}.png'.format(i)
            shutil.copy(path+sample+'/'+sample+'_030.png',
                        '../nas/SRAD2018Model/result/fake_rlt/'+sample+'/'+sample+suffix)


def fake_result1(samples):
    mean_idx = 2
    rlt_dir = '../nas/SRAD2018Model/result/fake_rlt_mean3_at_{}/'.format(mean_idx)
    try:
        os.mkdir(rlt_dir)
    except:
        pass
    pid = os.getpid()
    print("pid :", pid)
    data_path = '../nas/SRAD2018/SRAD2018_Test_1/'
    imgs = np.zeros(shape=(501, 501, 3), dtype=np.uint8)
    dx = [1, 2, 3, 4, 5, 6]
    del dx[mean_idx - 1]
    print(dx)
    for sample in tqdm(samples):
        # print(sample)
        os.mkdir(rlt_dir+sample)

        # mean data
        for i in range(30, 27, -1):
            suffix = '_{:03d}.png'.format(i)
            imgs[:, :, i-30] = io.imread(data_path+sample+'/'+sample+suffix)
        mean_img = np.mean(imgs, axis=2)
        mean_img[mean_img > 80] = 255
        with warnings.catch_warnings():
            warnings.simplefilter("ignore")
            io.imsave(rlt_dir+sample+'/'+sample+'_f00{}.png'.format(mean_idx), mean_img.astype(np.uint8))

        for i in dx:
            shutil.copy(data_path+sample+'/'+sample+'_030.png',
                        rlt_dir+sample+'/'+sample+'_f00{}.png'.format(i))


def parallel_():
    parallel_num = 16
    pools = multiprocessing.Pool(parallel_num)

    path = '../nas/SRAD2018/SRAD2018_Test_1/'
    samples = os.listdir(path)
    samples = [sample for sample in samples if os.path.isdir(path + sample)]

    ls = len(samples) // parallel_num
    print(ls)
    for i in range(parallel_num):
        if i == parallel_num-1:
            sp = samples[i * ls:]
            pools.apply_async(fake_result1, (sp,))

        else:
            sp = samples[i*ls:(i+1)*ls]
            pools.apply_async(fake_result1, (sp, ))
    pools.close()
    pools.join()


def zip_file():
    compacted_zip_name = 'haha_grad'
    dir_to_be_compact = 'unet_grad'
    shutil.make_archive(base_name='../nas/SRAD2018Model/result/'+compacted_zip_name, format='zip',
                        root_dir='../nas/SRAD2018Model/result/', base_dir=dir_to_be_compact)
    print('make archive done!')


def data_loader_test():
    train_list = glob.glob('../nas/SRAD2018Model/train_val_files/train*')
    train_list = sorted(train_list, key=lambda x: int(x[-7:-4]))
    gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.9, allow_growth=True)
    os.environ['CUDA_VISIBLE_DEVICES'] = '2'  # 使用 GPU id
    config = tf.ConfigProto(log_device_placement=False,
                            allow_soft_placement=True,
                            gpu_options=gpu_options)
    train_set = tf.data.TextLineDataset(train_list[3:], buffer_size=200, ).make_one_shot_iterator()
    txt = train_set.get_next()
    p = tf.placeholder(tf.string, )
    img_content = tf.read_file(p)
    img = tf.image.decode_image(img_content, channels=0)
    # f = open('loader_test.txt')
    with tf.Session(config=config) as sess:
        count = 0
        while True:
            path = sess.run(txt)
            path = np.array(path).astype(np.str)
            path = '../nas/SRAD2018/' + str(path)
            img_names = os.listdir(path)
            img_names = [name for name in img_names if '.png' in name]
            img_names = sorted(img_names, key=lambda x: int(x[-7:-4]))

            for idx, name in enumerate(img_names):
                try:
                    sess.run(img, feed_dict={p: path+'/'+name})
                except:
                    print(count, path+'/'+name)
            count += 1

            if count % 1000 == 0:
                print('decode image num: ', count)


if __name__ == '__main__':
    # parallel_()
    zip_file()
    # data_loader_test()
    pass