-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathevaluate_seg.py
161 lines (130 loc) · 6.25 KB
/
evaluate_seg.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
from datetime import datetime
import os
import re
import time
import glob
import logging
import ttach as tta
from tqdm import tqdm
import argparse
import torch
from PIL import Image
from skimage import morphology
import numpy as np
import cv2 as cv
import albumentations as A
from albumentations.pytorch import ToTensorV2
import torch.nn.functional as F
from seg_network.model import SegNetwork
from utils.pyutils import str2bool, set_seed, setup_logger
from utils import evaluate
from utils.tta_wrapper import SegmentationTTAWrapper
start_time = datetime.now()
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument("--gpu", type=int, default=0)
parser.add_argument("--dataset", type=str, default="luad")
parser.add_argument("--cls_num_classes", type=int, default=4)
parser.add_argument("--model_path", type=str, default=None)
parser.add_argument("--palette_path", type=str, default="./datasets/luad_palette.npy")
parser.add_argument("--seed", type=int, default=0)
parser.add_argument("--backbone", type=str, default="mit_b3")
parser.add_argument("--split", type=str, default="test")
args = parser.parse_args()
return args
def time_synchronized():
torch.cuda.synchronize() if torch.cuda.is_available() else None
return time.time()
def main():
args = get_args()
PALETTE = list(np.load(args.palette_path))
set_seed(args.seed)
args.dataset = args.dataset.lower()
if args.dataset == "luad":
args.seg_num_classes = 5
args.cls_num_classes = 4
args.img_root = "./data/LUAD-HistoSeg/{}/img/*.png".format(args.split)
args.mask_root = "./data/LUAD-HistoSeg/{}/mask/*.png".format(args.split)
CLASSES = ["TE", "NEC", "LYM", "TAS", "BACK"]
elif args.dataset == "bcss":
args.seg_num_classes = 5
args.cls_num_classes = 4
args.img_root = "./data/BCSS-WSSS/{}/img/*.png".format(args.split)
args.mask_root = "./data/BCSS-WSSS/{}/mask/*.png".format(args.split)
CLASSES = ["TUM", "STR", "LYM", "NEC", "BACK"]
img_paths, mask_paths = glob.glob(args.img_root), glob.glob(args.mask_root)
if len(mask_paths) == 0:
mask_paths = [None] * len(img_paths)
model_dir = re.split("/checkpoints/", args.model_path)[0]
os.makedirs(os.path.join(model_dir, "test_log"), exist_ok=True)
setup_logger(os.path.join(model_dir, "test_log", "eval_{}".format(args.split) + ".log"))
logging.info(args)
# get devices
device = torch.device("cuda:{}".format(args.gpu) if torch.cuda.is_available() else "cpu")
torch.cuda.set_device(args.gpu)
logging.info("using {} device.".format(device))
# create model
model = SegNetwork(backbone=args.backbone,
stride=[4, 2, 2, 1],
seg_num_classes=args.seg_num_classes,
embedding_dim=256,
pretrained=True)
logging.info("=================================================================================")
weights_dict = torch.load(args.model_path, map_location='cpu')
model_iter = weights_dict["iter"]
weights_dict = weights_dict["model"]
weights_dict = {k: v for k, v in weights_dict.items() if k in model.state_dict().keys()}
logging.info('loading from checkpoint: {}'.format(os.path.basename(args.model_path)))
# load weights
model.load_state_dict(weights_dict, strict=True)
model.to(device)
model.eval()
if args.dataset == "luad":
MEAN, STD = [0.69164956, 0.50502165, 0.73429191], [0.15521939, 0.22774934, 0.18846453]
elif args.dataset == "bcss":
MEAN, STD = [0.66791496, 0.47791372, 0.70623304], [0.1736589, 0.22564577, 0.19820057]
transform = A.Compose([
A.Normalize(MEAN, STD),
ToTensorV2(transpose_mask=True)
])
tta_model = SegmentationTTAWrapper(model, tta.aliases.d4_transform(), merge_mode='mean')
seg_matrix = evaluate.ConfusionMatrixAllClass(num_classes=args.seg_num_classes)
seg_matrix.reset()
with torch.no_grad():
for index, img_path in enumerate(tqdm(img_paths, ncols=100, ascii=' >=')):
mask_path = mask_paths[index]
img_name = os.path.basename(img_path)[:-4]
img = cv.imread(img_path, cv.IMREAD_UNCHANGED)
mask = np.array(Image.open(mask_path))[:, :, None].astype(np.uint8) if mask_path is not None else np.zeros((img.shape[0], img.shape[1], 1))
transdormed = transform(image=img, mask=mask)
inputs = transdormed["image"].float().unsqueeze(dim=0).cuda()
gt_mask = transdormed["mask"].float().unsqueeze(dim=0)
if args.split == "train":
if args.dataset == "luad":
term_split = re.split("-\[|\].", img_path)
cls_label = np.array(list(map(int, term_split[1].split(" ")))).reshape((1, -1))
else:
term_split = re.split("\[|\]", img_path)
cls_label = np.array([int(x) for x in term_split[1]]).reshape((1, -1))
else:
cls_label = np.zeros((1, args.cls_num_classes))
x = np.unique(mask) if np.unique(mask)[-1] != 4 else np.unique(mask)[:-1]
cls_label[:, x] = 1
segs = tta_model(inputs)
# priori mask, but no priori_mask is better
if args.dataset == "luad":
gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
ret, binary = cv.threshold(gray, 200, 255, cv.THRESH_BINARY)
binary = np.uint8(binary)
dst = morphology.remove_small_objects(binary == 255, min_size=80, connectivity=1).astype(np.uint8)
priori_bg_mask = dst.reshape(1, img.shape[0], img.shape[1])
priori_bg_mask = torch.from_numpy(priori_bg_mask).cuda()
segs[:, -1, :, :] *= priori_bg_mask
output = torch.argmax(segs, dim=1, keepdim=True).long().cpu()
seg_matrix.update(gt_mask.clone(), output.clone())
logging.info("=============================================================================================================")
logging.info(str(seg_matrix))
end_time = datetime.now()
logging.info("infference finished, cost time: {}".format((end_time - start_time).seconds // 60))
if __name__ == '__main__':
main()