-
Notifications
You must be signed in to change notification settings - Fork 21
/
test_Camelyon16.py
211 lines (154 loc) · 6.63 KB
/
test_Camelyon16.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
"""
Test script for Camelyon16 Probability map generation
"""
import argparse
import os
import time
import random
import numpy as np
from PIL import Image
import cv2
import glob
from tqdm import tqdm
import torch.backends.cudnn as cudnn
import torch
from torch.utils.data import Dataset
import torch.optim as optim
import torch.nn as nn
from util import AverageMeter
from collections import OrderedDict
from torchvision import transforms, datasets
import torch.nn.functional as F
from dataset import DatasetCamelyon16_test
import models.net as net
import matplotlib.pyplot as plt
import matplotlib.cm as cm
#####
def test(args, model, classifier, test_loader):
# switch to evaluate mode
model.eval()
classifier.eval()
batch_time = AverageMeter()
with torch.no_grad():
end = time.time()
probs_map = np.zeros(test_loader.dataset.mask.shape)
for batch_idx, (input, x_mask, y_mask) in enumerate(tqdm(test_loader, disable=False)):
# Get inputs and target
input = input.cuda()
x_mask = x_mask.data.numpy()
y_mask = y_mask.data.numpy()
# compute output ############
feats = model(input)
output = classifier(feats)
#######
probs = torch.softmax(output, dim=1).cpu()
probs = probs[:, -1] # second column 'tumor'
probs = probs.data.numpy()
# measure elapsed time
batch_time.update(time.time() - end)
end = time.time()
# print statistics and write summary every N batch
if (batch_idx + 1) % 10 == 0:
print('Test: [{0}/{1}]\t'
'BT {batch_time.val:.3f} ({batch_time.avg:.3f})'.format(batch_idx, len(test_loader), batch_time=batch_time))
probs_map[x_mask, y_mask] = probs
return probs_map
def parse_args():
parser = argparse.ArgumentParser('Argument for Camelyon16 test predictions')
parser.add_argument('--gpu', default='0', help='GPU id to use.')
parser.add_argument('--num_workers', type=int, default=8, help='num of workers to use.')
parser.add_argument('--seed', type=int, default=42, help='seed for initializing training.')
# model definition
parser.add_argument('--model', type=str, default='resnet18', help='choice of network architecture.')
parser.add_argument('--num_classes', type=int, default=2, help='# of classes.')
parser.add_argument('--batch_size', type=int, default=32, help='batch_size.')
parser.add_argument('--finetune_model_path', type=str, default='/home/srinidhi/projects/Camelyon16/',
help='path to load fine-tuned model for evaluation/test')
# Data paths
parser.add_argument('--test_image_pth', default='/home/srinidhi/projects/Camelyon16/testing/Images/')
parser.add_argument('--test_mask_pth', default='/home/srinidhi/projects/Camelyon16/test_mask/')
parser.add_argument('--probs_map_path', default='/home/srinidhi/projects/Camelyon16/Results/SSL/')
# Tiling parameters
parser.add_argument('--image_size', default=256, type=int, help='patch size width 256')
args = parser.parse_args()
return args
########
def main():
# parse the args
args = parse_args()
# set the model
if args.model == 'resnet18':
model = net.TripletNet_Finetune(args.model)
# original model saved file with DataParallel (Multi-GPU)
state_dict = torch.load(args.finetune_model_path)
# create new OrderedDict that does not contain `module.`
new_state_dict = OrderedDict()
for k, v in state_dict['model'].items():
name = k[7:] # remove `module.`
new_state_dict[name] = v
# Load pre-trained model
print('==> loading pre-trained model')
model.load_state_dict(new_state_dict)
print('==> classification')
classifier = net.FinetuneResNet(args.num_classes)
else:
raise NotImplementedError('model not supported {}'.format(args.model))
# Load model to CUDA
if torch.cuda.is_available():
model = model.cuda()
classifier = classifier.cuda()
cudnn.benchmark = True
####### Camelyon16 Evaluation Script ########################
wsipaths = []
maskpaths = []
for file_ext in ['tif', 'svs', 'npy']:
wsipaths = wsipaths + glob.glob('{}/*.{}'.format(args.test_image_pth, file_ext))
maskpaths = maskpaths + glob.glob('{}/*.{}'.format(args.test_mask_pth, file_ext))
wsipaths, maskpaths = sorted(wsipaths), sorted(maskpaths)
for file_ID in range(len(wsipaths)):
wsi_pth = wsipaths[file_ID]
mask_pth = maskpaths[file_ID]
wsi_id = str(os.path.split(wsi_pth)[-1])
wsi_id = os.path.splitext(wsi_id)[0]
# Test set
test_dataset = DatasetCamelyon16_test(wsi_pth, mask_pth, args.image_size)
test_loader = torch.utils.data.DataLoader(test_dataset, batch_size=args.batch_size, shuffle=False, num_workers=args.num_workers, pin_memory=True)
#####
n_data = len(test_dataset)
print('number of testing samples: {}'.format(n_data))
#################
# Testing Model
print("==> testing final test data...")
probs_map = test(args, model, classifier, test_loader)
# Save predictions
np.save(os.path.join(args.probs_map_path, wsi_id), probs_map)
probs_map = np.transpose(probs_map)
predicted_img = Image.fromarray(np.uint8(probs_map * 255))
predicted_img.save(os.path.join(args.probs_map_path, wsi_id + "." + 'png'), "PNG")
predicted_img.close()
# Save Heat-map
cmapper = cm.get_cmap('jet')
probs_heatmap = Image.fromarray(np.uint8(cmapper(np.clip(probs_map, 0, 1)) * 255))
probs_heatmap.save(os.path.join(args.probs_map_path, wsi_id + "_" + 'heatmap' + "." + 'png'), "PNG")
probs_heatmap.close()
# Plot heatmap-colorbar
plt.imshow(probs_map, cmap='jet', interpolation='nearest')
plt.colorbar()
plt.clim(0.00, 1.00)
plt.axis('off')
plt.savefig(os.path.join(args.probs_map_path, wsi_id + "_" + 'heatmap_bar' + "." + 'png'), bbox_inches='tight', dpi=300)
plt.clf()
del probs_map, cmapper, predicted_img, probs_heatmap
if __name__ == "__main__":
args = parse_args()
print(vars(args))
# Force the pytorch to create context on the specific device
os.environ["CUDA_VISIBLE_DEVICES"] = str(args.gpu)
if args.seed:
random.seed(args.seed)
np.random.seed(args.seed)
torch.manual_seed(args.seed)
if args.gpu:
torch.cuda.manual_seed_all(args.seed)
# Main function
main()