-
Notifications
You must be signed in to change notification settings - Fork 6
/
extract_features.py
210 lines (169 loc) · 8.75 KB
/
extract_features.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
"""
Script for extracting features from the 3D patches, assuming that preprocess/create_patches_3D.py has already been run
For fast-processing version, refer to extract_patches_fp.py
"""
import argparse
import os
from tqdm import tqdm
import h5py
import numpy as np
import pandas as pd
import torch
import torch.nn as nn
from models.feature_extractor import get_extractor_model
from utils.exp_utils import update_config
from utils.feature_utils import extract_patch_features
from data.ThreeDimDataset import ImgBag
from data.transforms import get_basic_data_transforms
device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
def extract_features(conf):
"""
Extract features from the patches
Args:
- conf (dict)
Returns:
- None
"""
print("\n================")
print('Loading model...')
if conf['target_patch_size_z'] is None:
patch_size = (conf['target_patch_size'], ) * 3
else:
patch_size = (conf['target_patch_size_z'], conf['target_patch_size'], conf['target_patch_size'])
print("Patch size ", patch_size)
model = get_extractor_model(encoder=conf['encoder'],
mode=conf['patch_mode'],
input=patch_size)
model.load_weights(**conf['pretrained'])
model.eval()
model = model.to(device)
print(model)
channel = model.get_channel_dim()
if torch.cuda.device_count() > 1:
model = nn.DataParallel(model)
aug_suffix = '_aug' if conf['augment_fold'] > 0 else ''
conf['extracted_dir'] = os.path.join(conf['dataroot'], conf['extracted_dir']) if (conf['extracted_dir'][0] != '/') else conf['extracted_dir']
patches_subdir = os.path.join(conf['extracted_dir'], 'patches')
## Define subfolder name
if conf['pretrained']['load_weights']:
subfolder = conf['pretrained']['pretrained_name'] + aug_suffix
else:
subfolder = 'random' + aug_suffix
feats_h5_subdir = os.path.join(conf['extracted_dir'],
'{}_h5_patch_features'.format(conf['encoder']),
subfolder)
os.makedirs(feats_h5_subdir, exist_ok=True)
### Setting Up For-Loop for feature extraction
df = pd.read_csv(os.path.join(conf['extracted_dir'], conf['process_list']))
total = len(df)
pbar_stack = tqdm(range(total))
if 'process_features' not in df.columns:
df['process_features'] = np.ones(total, dtype=np.int32)
print(df)
print("=============================")
print("\nBeginning {} patch extraction with {} settings".format(conf['patch_mode'],
conf['data_mode']))
######################
# Feature extraction #
######################
for i in pbar_stack:
idx = df.index[i]
slide_id = df.loc[idx, 'slide_id']
patches_h5_path = os.path.join(patches_subdir, slide_id + '_patches.h5')
clip_min = df.loc[idx, 'clip_min'] if 'clip_min' not in conf else conf['clip_min']
clip_max = df.loc[idx, 'clip_max'] if 'clip_max' not in conf else conf['clip_max']
# Error-Handling in disrupted scripts
IS_PROCESSED = df.loc[idx, 'process_features'] == 0
TO_SKIP = df.loc[idx, 'process_features'] == -1
IS_FAILURE = df.loc[idx, 'process_features'] == -2
if TO_SKIP or IS_FAILURE or IS_PROCESSED:
if TO_SKIP:
df.loc[idx, 'status_features'] = 'skip'
elif IS_FAILURE:
df.loc[idx, 'status_features'] = 'failure'
else:
df.loc[idx, 'status_features'] = 'proccessed'
df.loc[idx, 'bag_size'] = 0
continue
if not os.path.isfile(patches_h5_path):
df.loc[idx, 'status_features'] = 'skip'
df.loc[idx, 'bag_size'] = 0
print('Could not find patch file for: %s' % patches_h5_path)
continue
# If no issue, proceed with patch loading
img_dataset = ImgBag(file_path=patches_h5_path,
patch_mode=conf['patch_mode'],
clip_min=clip_min,
clip_max=clip_max)
# Augmentation loop
for aug_idx in range(conf['augment_fold'] + 1):
if aug_idx == 0:
aug_suffix = ''
else:
aug_suffix = '_aug{}'.format(aug_idx)
feats_h5_path = os.path.join(feats_h5_subdir, slide_id + aug_suffix + '.h5')
if os.path.isfile(feats_h5_path):
print(feats_h5_path + " exists!")
try:
_ = h5py.File(feats_h5_path, "r")
df.loc[idx, 'status_features'] = 'processed'
continue
except OSError:
print('Error Opening %s' % feats_h5_path)
os.system('rm %s' % feats_h5_path)
df.loc[idx, 'process_features'] = -2
df.loc[idx, 'status_features'] = 'failure'
# Feature extraction
print("Extracting features from ", slide_id, " for aug ", aug_idx, " with ", clip_min, clip_max)
data_transforms = get_basic_data_transforms(augment=False if aug_idx == 0 else True,
patch_mode=conf['patch_mode'],
data_mode=conf['data_mode'],
invert=conf['invert'])
img_dataset.set_transform(data_transforms)
extract_patch_features(dataset=img_dataset,
output_path=feats_h5_path,
model=model,
model_name=conf['encoder'],
batch_size=conf['batch_size'],
leave=bool(idx == len(pbar_stack) - 1),
channel=channel,
device=device)
if aug_idx == conf['augment_fold']:
df.loc[idx, 'process_features'] = 0
df.loc[idx, 'status_features'] = 'processed'
df.to_csv(os.path.join(conf['extracted_dir'], conf['process_list']), index=False)
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Feature Extraction')
parser.add_argument('--dataroot', type=str, help='The root project folder directory. \
We assume that for most projects, you would want your extracted features to live in the same directory as your WSIs.')
parser.add_argument('--extracted_dir', type=str, default=None, help='Folder to save extracted results for patching, tissue segmentation, and stitching. \
By default, we assume args.extracted_dir is a directory within args.datroot. However, passing an absolute path into args.extracted_dir (by checking \
if "/" is in args.extracted_dir) will override using args.dataroot as a root path.')
parser.add_argument('--config', type=str, default='.',
help='Config files that contain default parameters')
parser.add_argument('--batch_size', type=int, default=50)
parser.add_argument('--patch_mode', default='2D', choices=['2D', '3D'], type=str,
help='2D patching or 3D patching')
parser.add_argument('--clip_min', type=int)
parser.add_argument('--clip_max', type=int)
parser.add_argument('--process_list', type=str, default='process_list_extract.csv',
help='name of list of images to process with parameters (.csv)')
parser.add_argument('--encoder', type=str,
help='cnn feature extractor to use')
parser.add_argument('--target_patch_size_z', type=int, default=96,
help='the desired size of patches for optional scaling before feature embedding')
parser.add_argument('--target_patch_size', type=int, default=96,
help='the desired size of patches for optional scaling before feature embedding')
parser.add_argument('--augment_fold', default=5, type=int,
help='Number of augmentations to perform')
parser.add_argument('--data_mode', type=str,
help='The input device mode, e.g., CT, OTLS')
parser.add_argument('--invert', action='store_true', default=False,
help='Whether to invert intesinty or not')
args = parser.parse_args()
conf = update_config(args) # Update args namespace with parameters in config file
print("\nPARAMETERS: ", conf)
if conf['patch_mode'] == '3D' and conf['batch_size'] >= 100:
print("*************************************")
print("WARNING: Make sure you are using pytorch 2.0 for large batch size of 3D inputs!")
extract_features(conf)