This repository has been archived by the owner on Jan 3, 2024. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 3
/
engine_finetune.py
165 lines (130 loc) · 6.29 KB
/
engine_finetune.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
# 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.
# --------------------------------------------------------
# References:
# DeiT: https://github.com/facebookresearch/deit
# BEiT: https://github.com/microsoft/unilm/tree/master/beit
# --------------------------------------------------------
import math
import shutil
import sys
from typing import Iterable, Optional
import os
import torch
from timm.data import Mixup
from timm.utils import accuracy
import util.misc as misc
import util.lr_sched as lr_sched
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
def train_one_epoch(model: torch.nn.Module, criterion: torch.nn.Module,
data_loader: Iterable, optimizer: torch.optim.Optimizer,
device: torch.device, epoch: int, loss_scaler, max_norm: float = 0,
mixup_fn: Optional[Mixup] = None, 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}'))
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, targets) 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 = samples.to(device, non_blocking=True)
targets = targets.to(device, non_blocking=True)
if mixup_fn is not None:
samples, targets = mixup_fn(samples, targets)
with torch.cuda.amp.autocast():
outputs = model(samples)
loss = criterion(outputs, targets)
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, clip_grad=max_norm,
parameters=model.parameters(), create_graph=False,
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)
min_lr = 10.
max_lr = 0.
for group in optimizer.param_groups:
min_lr = min(min_lr, group["lr"])
max_lr = max(max_lr, group["lr"])
metric_logger.update(lr=max_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('loss', loss_value_reduce, epoch_1000x)
log_writer.add_scalar('lr', max_lr, 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()}
@torch.no_grad()
def evaluate(data_loader, model, device, class_idx_map=None, visualize_epoch=0, max_num=30):
criterion = torch.nn.CrossEntropyLoss()
metric_logger = misc.MetricLogger(delimiter=" ")
header = 'Test:'
# switch to evaluation mode
model.eval()
if visualize_epoch:
save_dir = f"./results/clef_fungi_results_epoch{visualize_epoch}"
if os.path.exists(save_dir):
shutil.rmtree(save_dir)
os.makedirs(save_dir)
else:
os.makedirs(save_dir)
for i, batch in enumerate(metric_logger.log_every(data_loader, 10, header)):
images = batch[0]
file_names = data_loader.dataset.samples[i:i+images.size(0)]
target = batch[-1]
images = images.to(device, non_blocking=True)
target = target.to(device, non_blocking=True)
# compute output
with torch.cuda.amp.autocast():
output = model(images)
loss = criterion(output, target)
###################################################################
# to save the results
if visualize_epoch:
scores, predictions = output.softmax(dim=1).topk(k=max_num, dim=1)
for batch_i in range(images.size(0)):
for num in range(max_num):
file_name_ = os.path.splitext(os.path.basename(file_names[batch_i][0]))[0]
label = predictions[batch_i, num].cpu().item()
label = [k for k, v in class_idx_map.items() if v == label][0]
score = scores[batch_i, num].cpu().item()
single = str(label) + ";" + str(score) + ";" + str(num+1) + '\n'
with open(f'{save_dir}/{file_name_}.csv', 'a+') as file:
file.writelines(str(single))
# the end
###################################################################
if output.size(1) >= 6:
acc1, acc5 = accuracy(output, target, topk=(1, 5))
else:
acc1, acc5 = accuracy(output, target, topk=(1, 2))
batch_size = images.shape[0]
metric_logger.update(loss=loss.item())
metric_logger.meters['acc1'].update(acc1.item(), n=batch_size)
metric_logger.meters['acc5'].update(acc5.item(), n=batch_size)
# gather the stats from all processes
metric_logger.synchronize_between_processes()
print('* Acc@1 {top1.global_avg:.3f} Acc@5 {top5.global_avg:.3f} loss {losses.global_avg:.3f}'
.format(top1=metric_logger.acc1, top5=metric_logger.acc5, losses=metric_logger.loss))
# with open('results.txt', 'w+') as file:
# file.writelines(str(save_results))
return {k: meter.global_avg for k, meter in metric_logger.meters.items()}