-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathengine_pretrain.py
179 lines (154 loc) · 6.43 KB
/
engine_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
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
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
# --------------------------------------------------------
#
# Original Work:
# Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles
# Chaitanya Ryali, Yuan-Ting Hu, Daniel Bolya et al.
# https://arxiv.org/abs/2306.00989/
#
# Enhanced and modified by Stoffl et al.
#
# For more details on our work, please refer to:
# Elucidating the Hierarchical Nature of Behavior with Masked Autoencoders
# Lucas Stoffl, Andy Bonnetto, Stéphane d'Ascoli, Alexander Mathis
# https://www.biorxiv.org/content/10.1101/2024.08.06.606796v1
# --------------------------------------------------------
import math
from typing import Iterable
import torch
from iopath.common.file_io import g_pathmgr as pathmgr
import util.lr_sched as lr_sched
import util.misc as misc
def train_one_epoch(
model: torch.nn.Module,
data_loader: Iterable,
data_loader_val: Iterable,
optimizer: torch.optim.Optimizer,
device: torch.device,
epoch: int,
loss_scaler,
log_writer=None,
args=None,
fp32=False,
):
model.train(True)
metric_logger = misc.MetricLogger(delimiter=" ")
metric_logger.add_meter("lr", misc.SmoothedValue(window_size=1, fmt="{value:.6f}"))
metric_logger.add_meter(
"cpu_mem", misc.SmoothedValue(window_size=1, fmt="{value:.6f}")
)
metric_logger.add_meter(
"cpu_mem_all", misc.SmoothedValue(window_size=1, fmt="{value:.6f}")
)
metric_logger.add_meter(
"gpu_mem", misc.SmoothedValue(window_size=1, fmt="{value:.6f}")
)
metric_logger.add_meter(
"mask_ratio", misc.SmoothedValue(window_size=1, fmt="{value:.6f}")
)
header = "Epoch: [{}]".format(epoch)
print_freq = 20
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, (samples, _) in enumerate(
for data_iter_step, samples in enumerate(
metric_logger.log_every(data_loader, print_freq, header)
):
# we use a per iteration (instead of per epoch) lr scheduler
if data_iter_step % accum_iter == 0:
lr_sched.adjust_learning_rate(
optimizer, data_iter_step / len(data_loader) + epoch, args
)
samples, targets = samples
samples = samples.to(device, non_blocking=True)
if targets:
targets = [tgt.to(device, non_blocking=True) for tgt in targets]
if len(samples.shape) == 6:
b, r, c, t, h, w = samples.shape
samples = samples.reshape(b * r, c, t, h, w)
mask_ratio = args.mask_ratio
with torch.cuda.amp.autocast(enabled=not fp32):
loss, _, _, _, _ = model(
samples,
targets,
mask_ratio=mask_ratio,
mask_strategy=args.masking_strategy,
)
loss_value = loss.item()
if not math.isfinite(loss_value):
for _ in range(args.num_checkpoint_del):
try:
path = misc.get_last_checkpoint(args)
pathmgr.rm(path)
print(f"remove checkpoint {path}")
except Exception as _:
pass
raise Exception("Loss is {}, stopping training".format(loss_value))
loss /= accum_iter
loss_scaler(
loss,
optimizer,
parameters=model.parameters(),
update_grad=(data_iter_step + 1) % accum_iter == 0,
clip_grad=args.clip_grad,
)
if (data_iter_step + 1) % accum_iter == 0:
optimizer.zero_grad()
torch.cuda.synchronize()
metric_logger.update(loss=loss_value)
metric_logger.update(cpu_mem=misc.cpu_mem_usage()[0])
metric_logger.update(cpu_mem_all=misc.cpu_mem_usage()[1])
metric_logger.update(gpu_mem=misc.gpu_mem_usage())
metric_logger.update(mask_ratio=args.mask_ratio)
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:
"""We use epoch_1000x as the x-axis in tensorboard.
This calibrates different curves when batch size changes.
"""
epoch_1000x = int((data_iter_step / len(data_loader) + epoch) * 1000)
log_writer.add_scalar("train_loss", loss_value_reduce, epoch_1000x)
log_writer.add_scalar("lr", lr, epoch_1000x)
if data_loader_val:
# compute loss on test data
if epoch % 5 == 0:
for data_iter_step, samples in enumerate(
metric_logger.log_every(data_loader_val, print_freq, header)
):
samples, targets = samples
samples = samples.to(device, non_blocking=True)
if targets:
targets = [tgt.to(device, non_blocking=True) for tgt in targets]
if len(samples.shape) == 6:
b, r, c, t, h, w = samples.shape
samples = samples.reshape(b * r, c, t, h, w)
mask_ratio = args.mask_ratio
with torch.no_grad():
loss, _, _, _, _ = model(
samples,
targets,
mask_ratio=mask_ratio,
mask_strategy=args.masking_strategy,
)
loss_value = loss.item()
loss /= accum_iter
torch.cuda.synchronize()
loss_value_reduce = misc.all_reduce_mean(loss_value)
if log_writer is not None and (data_iter_step + 1) % accum_iter == 0:
"""We use epoch_1000x as the x-axis in tensorboard.
This calibrates different curves when batch size changes.
"""
epoch_1000x = int(
(data_iter_step / len(data_loader) + epoch) * 1000
)
log_writer.add_scalar("test_loss", loss_value_reduce, epoch_1000x)
# 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()}