-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtrain.py
163 lines (138 loc) · 8.23 KB
/
train.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
# Copyright (c) 2024 Qihang Zhou
# Licensed under the MIT License (refer to the LICENSE file for details)
# Modifications made by Jitao Ma, 2024
import AnomalyCLIP_lib
import torch
import argparse
import torch.nn.functional as F
from prompt_ensemble import AnomalyCLIP_PromptLearner
from loss import FocalLoss, BinaryDiceLoss
from utils import normalize
from dataset import Dataset
from logger import get_logger
from tqdm import tqdm
import numpy as np
import os
import random
from utils import get_transform
def setup_seed(seed):
torch.manual_seed(seed)
torch.cuda.manual_seed_all(seed)
np.random.seed(seed)
random.seed(seed)
torch.backends.cudnn.deterministic = True
torch.backends.cudnn.benchmark = False
def train(args):
logger = get_logger(args.save_path)
preprocess, target_transform = get_transform(args)
device = "cuda" if torch.cuda.is_available() else "cpu"
AnomalyCLIP_parameters = {"Prompt_length": args.n_ctx, "learnabel_text_embedding_depth": args.depth, "learnabel_text_embedding_length": args.t_n_ctx}
model, _ = AnomalyCLIP_lib.load("ViT-L/14@336px", device=device, design_details = AnomalyCLIP_parameters)
model.eval()
train_data = Dataset(root=args.train_data_path, transform=preprocess, target_transform=target_transform, dataset_name = args.dataset)
train_dataloader = torch.utils.data.DataLoader(train_data, batch_size=args.batch_size, shuffle=True)
##########################################################################################
prompt_learner = AnomalyCLIP_PromptLearner(model.to("cpu"), AnomalyCLIP_parameters)
prompt_learner.to(device)
class_learner = AnomalyCLIP_PromptLearner(model.to("cpu"), AnomalyCLIP_parameters, classnames=['part'], state_list=['the flawless {}', 'the damaged {}'])
class_learner.to(device)
former = AnomalyCLIP_lib.PathWay(model.text_projection, model.visual.proj)
former.to(device)
model.to(device)
model.visual.DAPM_replace(DPAM_layer = None)
##########################################################################################
optimizer = torch.optim.Adam([{'params': prompt_learner.parameters()},
{'params': class_learner.parameters()},
{'params': former.parameters(), 'lr': 1e-4},], args.learning_rate, betas=(0.5, 0.999))
scheduler = torch.optim.lr_scheduler.StepLR(optimizer, 7, 0.1)
# losses
loss_focal = FocalLoss()
loss_dice = BinaryDiceLoss()
model.eval()
prompt_learner.train()
class_learner.train()
former.train()
for epoch in range(args.epoch):
model.eval()
prompt_learner.train()
class_learner.train()
former.train()
loss_list = []
image_loss_list = []
for items in tqdm(train_dataloader):
image = items['img'].to(device)
label = items['anomaly']
gt = items['img_mask'].squeeze().to(device)
gt[gt > 0.5] = 1
gt[gt <= 0.5] = 0
with torch.no_grad():
image_features, patch_features, all_image_features = model.encode_image(image, args.features_list, DPAM_layer = None)
image_features = image_features / image_features.norm(dim=-1, keepdim=True)
###################################
prompts, tokenized_prompts, compound_prompts_text = prompt_learner(cls_id=None)
text_features, text_features_all = model.encode_text_learn(prompts, tokenized_prompts, compound_prompts_text)
text_features = torch.stack(torch.chunk(text_features, dim=0, chunks=2), dim=1)
text_features_all = torch.stack(torch.chunk(text_features_all, dim=0, chunks=2), dim=1)
text_features_norm = text_features / text_features.norm(dim=-1, keepdim=True)
prompts, tokenized_prompts, compound_prompts_text = class_learner(cls_id = None)
patch_classes, patch_classes_all = model.encode_text_learn(prompts, tokenized_prompts, compound_prompts_text)
patch_classes = torch.stack(torch.chunk(patch_classes, dim=0, chunks=2), dim=1)
patch_classes_all = torch.stack(torch.chunk(patch_classes_all, dim=0, chunks=2), dim=1)
# Apply DPAM surgery
text_probs = image_features.unsqueeze(1) @ text_features_norm.permute(0, 2, 1)
text_probs = text_probs[:, 0, ...]/0.07
image_loss = F.cross_entropy(text_probs, label.long().cuda())
image_loss_list.append(image_loss.item())
#########################################################################
similarity_map_list = []
for idx, patch_feature in enumerate(patch_features):
if idx >= args.feature_map_layer[0]:
class_feature, _ = former(text_features.repeat(patch_feature.shape[0], 1, 1), patch_classes.repeat(patch_feature.shape[0], 1, 1), patch_feature[:, 1:, :])
class_feature = class_feature / class_feature.norm(dim = -1, keepdim = True)
patch_feature = patch_feature / patch_feature.norm(dim = -1, keepdim = True)
similarity = AnomalyCLIP_lib.compute_similarity(patch_feature, class_feature)
similarity_map = AnomalyCLIP_lib.get_similarity_map(similarity[:, 1:, :], args.image_size).permute(0, 3, 1, 2)
similarity_map_list.append(similarity_map)
loss = 0
for i in range(len(similarity_map_list)):
loss += loss_focal(similarity_map_list[i], gt)
loss += loss_dice(similarity_map_list[i][:, 1, :, :], gt)
loss += loss_dice(similarity_map_list[i][:, 0, :, :], 1-gt)
optimizer.zero_grad()
# optimizer_c.zero_grad()
(loss + image_loss).backward()
optimizer.step()
# optimizer_c.step()
loss_list.append(loss.item())
# logs
if (epoch + 1) % args.print_freq == 0:
logger.info('epoch [{}/{}], loss:{:.4f}, image_loss:{:.4f}'.format(epoch + 1, args.epoch, np.mean(loss_list), np.mean(image_loss_list)))
# save model
if (epoch + 1) % args.save_freq == 0:
ckp_path = os.path.join(args.save_path, 'epoch_' + str(epoch + 1) + '.pth')
torch.save({"prompt_learner": prompt_learner.state_dict(),
"class_learner": class_learner.state_dict(),
"former": former.state_dict()}, ckp_path)
if __name__ == '__main__':
parser = argparse.ArgumentParser("AnomalyCLIP", add_help=True)
# parser.add_argument("--train_data_path", type=str, default="/home/worker1/AD-datasets/mvtec-dataset", help="train dataset path")
# parser.add_argument("--train_data_path", type=str, default="/home/worker1/AD-datasets/VisA_20220922-dataset", help="train dataset path")
# parser.add_argument("--train_data_path", type=str, default="/home/worker1/AD-datasets/HeadCT-dataset")
parser.add_argument("--train_data_path", type=str, default="/home/worker1/AD-datasets/ISBI2016")
parser.add_argument("--save_path", type=str, default='./checkpoint', help='path to save results')
parser.add_argument("--dataset", type=str, default='ISBI', help="train dataset name")
parser.add_argument("--depth", type=int, default=9, help="image size")
parser.add_argument("--n_ctx", type=int, default=12, help="zero shot")
parser.add_argument("--t_n_ctx", type=int, default=4, help="zero shot")
parser.add_argument("--feature_map_layer", type=int, nargs="+", default=[0, 1, 2, 3], help="zero shot")
parser.add_argument("--features_list", type=int, nargs="+", default=[6, 12, 18, 24], help="features used")
parser.add_argument("--epoch", type=int, default=15, help="epochs")
parser.add_argument("--learning_rate", type=float, default=0.001, help="learning rate")
parser.add_argument("--batch_size", type=int, default=8, help="batch size")
parser.add_argument("--image_size", type=int, default=518, help="image size")
parser.add_argument("--print_freq", type=int, default=1, help="print frequency")
parser.add_argument("--save_freq", type=int, default=1, help="save frequency")
parser.add_argument("--seed", type=int, default=111, help="random seed")
args = parser.parse_args()
setup_seed(args.seed)
train(args)