/
engine_mvlt_pretrain.py
146 lines (112 loc) · 5.35 KB
/
engine_mvlt_pretrain.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
# Copyright (c) Westone Information Industry Inc.
# --------------------------------------------------------
# References:
# DeiT: https://github.com/facebookresearch/deit
# BEiT: https://github.com/microsoft/unilm/tree/master/beit
# MAE: https://github.com/facebookresearch/mae
# --------------------------------------------------------
import math
import sys
from typing import Iterable
import torch
import util.misc as misc
import util.lr_sched as lr_sched
import util.token_converter as token_converter
import numpy as np
from PIL import Image
def train(model: torch.nn.Module,
data_loader, optimizer: torch.optim.Optimizer,
device: torch.device, num_iter: int, loss_scaler,
log_writer=None,
args=None):
model.train(True)
metric_logger = misc.MetricLogger(delimiter=" ")
metric_logger.add_meter('lr', misc.SmoothedValue(
window_size=1, fmt='{value:.6f}'))
print_freq = 2
converter = token_converter.TokenLabelConverter(args)
accum_iter = args.accum_iter
optimizer.zero_grad()
if log_writer is not None:
print('log_dir: {}'.format(log_writer.log_dir))
for data_iter_step, (images, _words, w_mask) in enumerate(metric_logger.log_every_iter(data_loader, num_iter, args.start_iter, print_freq), args.start_iter):
# we use a per iteration (instead of per epoch) lr scheduler
if data_iter_step % accum_iter == 0:
lr_sched.adjust_learning_rate_iter(optimizer, data_iter_step, args)
images = images.to(device, non_blocking=True)
words = converter.encode(_words)
w_mask = w_mask.to(device, non_blocking=True).flatten(1).to(torch.bool)
with torch.cuda.amp.autocast():
loss_dict, res_dict, mask = model(
images, words, w_mask, mask_ratio=args.mask_ratio)
loss = loss_dict["loss"]
loss_value = loss.item()
if not math.isfinite(loss_value):
print("Loss is {}, stopping training".format(loss_value))
sys.exit(1)
loss /= accum_iter
loss_scaler(loss, optimizer, parameters=model.parameters(),
update_grad=(data_iter_step + 1) % accum_iter == 0)
if (data_iter_step + 1) % accum_iter == 0:
optimizer.zero_grad()
torch.cuda.synchronize()
metric_logger.update(loss=loss_value)
if args.exp_semantic:
exp_v_loss = loss_dict["exp_v_loss"].item()
metric_logger.update(exp_v_loss=exp_v_loss)
exp_t_loss = loss_dict["exp_t_loss"].item()
metric_logger.update(exp_t_loss=exp_t_loss)
if args.imp_semantic:
imp_v_loss = loss_dict["imp_v_loss"].item()
metric_logger.update(imp_v_loss=imp_v_loss)
imp_t_loss = loss_dict["imp_t_loss"].item()
metric_logger.update(imp_t_loss=imp_t_loss)
lr = optimizer.param_groups[0]["lr"]
metric_logger.update(lr=lr)
loss_value_reduce = misc.all_reduce_mean(loss_value)
if log_writer is not None and (data_iter_step + 1) % accum_iter == 0:
log_writer.add_scalar('train_loss', loss_value_reduce, data_iter_step)
log_writer.add_scalar('lr', lr, data_iter_step)
if args.output_dir and ((data_iter_step + 1) % (args.save_model_freq * accum_iter) == 0 or data_iter_step + 1 == args.num_iter):
misc.save_model_iter(
args=args, model=model, model_without_ddp=model.module, optimizer=optimizer,
loss_scaler=loss_scaler, iteration=data_iter_step + 1)
if misc.is_main_process() and data_iter_step % (1000 * accum_iter) == 0:
log_writer_add_images(data_iter_step, model, images, mask, res_dict, log_writer, args)
# gather the stats from all processes
metric_logger.synchronize_between_processes()
print("Averaged stats:", metric_logger)
return {k: meter.global_avg for k, meter in metric_logger.meters.items()}
def log_writer_add_images(data_iter_step, model, images, mask, res_dict, log_writer, args):
im_list = []
images = images.detach().to('cpu')
im = images[0].numpy()
im = im * 0.5+0.5
im_list.append(im)
mask = mask.detach().to('cpu')
mask = mask.unsqueeze(-1).repeat(1, 1,
model.module.patch_embed.patch_size[0]**2 * 3)
mask = model.module.unpatchify(mask)
im_masked = images[0]*(1-mask[0])
im = im_masked.numpy()
im = im * 0.5+0.5
im_list.append(im)
str_list = ["gt","mask"]
if args.exp_semantic:
exp_v = res_dict["exp_v"]
pred_imgs = model.module.unpatchify(exp_v.detach().to('cpu'))
im_paste = images[0] * (1 - mask[0]) + pred_imgs[0] * mask[0]
im = im_paste.numpy()
im = im * 0.5+0.5
im_list.append(im)
str_list.append("exp-predict")
if args.imp_semantic:
imp_v = res_dict["imp_v"]
pred_imgs = model.module.unpatchify(imp_v.detach().to('cpu'))
im_paste = images[0] * (1 - mask[0]) + pred_imgs[0] * mask[0]
im = im_paste.to('cpu').numpy()
im = im * 0.5+0.5
im_list.append(im)
str_list.append("imp-predict")
im_list = np.concatenate(im_list, 1)
log_writer.add_image(" | ".join(str_list), im_list, data_iter_step)