-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain.py
193 lines (159 loc) · 7.15 KB
/
main.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
import os
import sys
import time
import torch.nn as nn
from datetime import timedelta
import argparse
import torch
import torch.multiprocessing
import os.path as osp
import tabulate
from termcolor import colored
import copy
sys.path.append('/home/server1-ailab/Desktop/Bach/Text_Image_Matching/')
from configs.default import get_default_config
from models import build_model
#from models.gnn import graph_net, utils, trainer, networks, preprocess
from preprocessing.transforms import *
from utils import Logger
from utils import set_seed
from datasets.build import build_dataloader
from optimizer.build import build_vanilla_optimizer, FreezeBackbone
from tools.evaluate import evaluate
from tools.train import train_epoch, train_epoch_multiview, train_epoch_multiview_gnn, train_epoch_multigrained, train_epoch_multiview_mlm, train_epoch_mlm
import wandb
wandb.login(key="016c2b13617e9b971749b50884ef7bdf9ec68f1a")
best_r1_eval = 0.
best_r1_eval_by_test = 0.
table = []
max_record = ['Max', 'Max', 0, 0, 0, 0, 0]
header = ['Method', 'Dataset', 'Epoch', 'R-1', 'R-5', 'R-10', 'mAP']
table.append(header)
print_interval = 20 # csv
save_interval = 200 # checkpoint
def results_record(name, dataset, epoch, r1, r5, r10, mAP, is_test=False):
# result csv
record = list()
# name = args.name
record.append(name)
record.append(dataset)
record.append(epoch)
record.append(r1)
record.append(r5)
record.append(r10)
record.append(mAP)
print_table = copy.deepcopy(table)
global max_record
if is_test and record[3] > max_record[3]:
max_record = copy.deepcopy(record)
max_record[2] = 'Max_' + str(max_record[2])
print_table.append(max_record)
display = tabulate.tabulate(
print_table,
tablefmt="pipe",
headers='firstrow',
numalign="left",
floatfmt='.3f')
print(f"====> results in csv format: \n" + colored(display, "cyan"))
def prepare_start():
parser = argparse.ArgumentParser(description='Training')
parser.add_argument('--resume', '-r', action='store_true', help='resume from checkpoint')
parser.add_argument('--config', default="configs/baseline.yaml", type=str,
help='config_file')
parser.add_argument('--name', default="baseline", type=str,
help='experiments')
parser.add_argument('--logs-dir', type=str, metavar='PATH', default='logs/')
parser.add_argument('--eval_only', '-eval', action='store_true', help='only eval')
parser.add_argument('--use_wandb', type=bool, help='use weight and biases', default=False)
parser.add_argument('--rerank_top', type=int, help='Rerank', default=20)
parser.add_argument('--use_cuda', type=bool, help='use cuda', default=True)
parser.add_argument(
"opts",
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
args = parser.parse_args()
cfg = get_default_config()
cfg.merge_from_file(args.config)
cfg.merge_from_list(args.opts)
args.cfg = cfg
sys.stdout = Logger(osp.join(args.logs_dir, 'log.txt'))
print(f"====> load config from {args.config}")
return args, cfg
def main():
args, cfg = prepare_start()
set_seed(cfg.TRAIN.SEED, cfg.TRAIN.DETERMINISTIC)
os.makedirs(args.logs_dir, exist_ok=True)
if args.use_wandb:
print("Using Weight and Biases for logging")
wandb.init(project="Text_Image_Matching",name=cfg.MODEL.NAME, config=cfg)
# Cross-Modal Model
trainloader, val_text_loader, val_image_loader, train_data = build_dataloader(cfg)
if cfg.MODEL.NUM_CLASS == 0:
cfg.MODEL.NUM_CLASS = len(train_data)
if args.eval_only:
args.resume = True
model = build_model(cfg, args)
# print(model)
model_freeze = FreezeBackbone(model, freeze_epoch=cfg.TRAIN.FREEZE_EPOCH)
optimizer, scheduler = build_vanilla_optimizer(cfg, model, trainloader)
if args.eval_only:
evaluate(model, val_text_loader, val_image_loader, 0, args, optimizer)
return
# Training
start_time = time.monotonic()
model.train()
global_step = 0
model_freeze.start_freeze_backbone()
selected_idx = None
train_fn = None
train_fn_gnn = None
scaler = torch.cuda.amp.GradScaler()
if cfg.TRAIN.MODE == 'clip_multiview_gnn':
print("USING CLIP_MULTIVIEW_GNN TRAINING")
train_fn_gnn = train_epoch_multiview_gnn(args=args, step=0, cfg=cfg, model=model, optimizer=optimizer, model_freeze=model_freeze, scheduler=scheduler,
trainloader=trainloader, val_image_loader=val_image_loader, val_text_loader=val_text_loader,
global_step=global_step, result_record=results_record, save_interval=save_interval, wandb=wandb if args.use_wandb else None, v=selected_idx)
losses, global_step, metrics = train_fn_gnn.train()
elif cfg.TRAIN.MODE == 'multiview':
print("USING MULTIVIEW TRAINING")
train_fn = train_epoch_multiview
elif cfg.TRAIN.MODE == 'multiview_mlm':
print("USING MULTIVIEW MLM TRAINING")
train_fn = train_epoch_multiview_mlm
elif cfg.TRAIN.MODE == 'clip_mlm':
print("USING CLIP MLM")
train_fn = train_epoch_mlm
elif cfg.TRAIN.MODE == 'multigrained':
print("USING MULTIGRAINED TRAINING")
train_fn = train_epoch_multigrained
else:
print("USING DEFAULT TRAINING")
train_fn = train_epoch
if train_fn is not None:
for epoch in range(cfg.TRAIN.EPOCH):
epo_start_time = time.monotonic()
if cfg.TRAIN.MODE == 'multigrained':
losses, global_step, metrics = train_fn(cfg=cfg, epoch=epoch, model=model, optimizer=optimizer, scheduler=scheduler,
trainloader=trainloader, val_image_loader=val_image_loader, val_text_loader=val_text_loader,
global_step=global_step, result_record=results_record, args=args, wandb=wandb if args.use_wandb else None)
else:
losses, global_step, metrics = train_fn(cfg=cfg, epoch=epoch, model=model, optimizer=optimizer, model_freeze=model_freeze, scheduler=scheduler,
trainloader=trainloader, val_image_loader=val_image_loader, val_text_loader=val_text_loader,
global_step=global_step, result_record=results_record, args=args, wandb=wandb if args.use_wandb else None, scaler=scaler)
if epoch % save_interval == 1:
checkpoint_file = args.logs_dir + "/checkpoint_%d.pth" % epoch
torch.save({"epoch": epoch,"state_dict": model.state_dict(), "optimizer": optimizer.state_dict()}, checkpoint_file)
scheduler.step()
epo_end_time = time.monotonic()
print(f'Epoch {epoch} running time: ', timedelta(seconds=epo_end_time - epo_start_time))
print(f'Logs dir: {args.logs_dir} ')
if args.use_wandb:
wandb.log({"AVG Loss": losses})
del train_data, trainloader
end_time = time.monotonic()
print('Total running time: ', timedelta(seconds=end_time - start_time))
print(f'Logs dir: {args.logs_dir} ')
if __name__ == "__main__":
main()