-
Notifications
You must be signed in to change notification settings - Fork 29
/
Copy pathtrain.py
160 lines (137 loc) · 5.46 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
import os
import torch
import datetime
from darknet import Darknet19
from datasets.pascal_voc import VOCDataset
import utils.yolo as yolo_utils
import utils.network as net_utils
from utils.timer import Timer
import cfgs.config as cfg
from random import randint
from patch import *
try:
from tensorboardX import SummaryWriter
except ImportError:
SummaryWriter = None
# data loader
imdb = VOCDataset(cfg.imdb_train, cfg.DATA_DIR, cfg.train_batch_size,
yolo_utils.preprocess_train, processes=2, shuffle=True,
dst_size=cfg.multi_scale_inp_size)
# dst_size=cfg.inp_size)
print('load data succ...')
net = Darknet19()
# net_utils.load_net(cfg.trained_model, net)
#pretrained_model = os.path.join(cfg.train_output_dir,
# 'darknet19_voc07trainval_exp3_79.h5')
pretrained_model = 'yolo-voc.weights.h5'
# training from scratch: with_patch=False
# continue training or testing: with_patch=True
net_utils.load_net(pretrained_model, net, with_patch=False)
#net.load_from_npz(cfg.pretrained_model, num_conv=18)
net.cuda()
net.train()
print('load net succ...')
# optimizer
start_epoch = 0
lr = 1e-2
optimizer = torch.optim.SGD([net.patch], lr=lr, momentum=cfg.momentum,
weight_decay=cfg.weight_decay)
# tensorboad
use_tensorboard = cfg.use_tensorboard and SummaryWriter is not None
# use_tensorboard = False
if use_tensorboard:
summary_writer = SummaryWriter(os.path.join(cfg.TRAIN_DIR, 'runs', cfg.exp_name))
else:
summary_writer = None
batch_per_epoch = imdb.batch_per_epoch
train_loss = 0
bbox_loss, iou_loss, cls_loss = 0., 0., 0.
cnt = 0
t = Timer()
step_cnt = 0
size_index = 0
for step in range(start_epoch * imdb.batch_per_epoch,
cfg.max_epoch * imdb.batch_per_epoch):
t.tic()
if step % imdb.batch_per_epoch == 0:
print('-----------save %d patch to ------------'%step)
save_patch(net.patch, step)
print(net.patch)
print('\n')
# batch
batch = imdb.next_batch(size_index)
im = batch['images']
gt_boxes = batch['gt_boxes']
gt_classes = batch['gt_classes']
dontcare = batch['dontcare']
orgin_im = batch['origin_im']
# reset patch class here
for g, gt_cls in enumerate(gt_classes):
for c,the_cls in enumerate(gt_cls):
gt_classes[g][c] = cfg.target_class
#print('gt_boxes', gt_boxes)
# reset patch bbox here
#for g, gt_box in enumerate(gt_boxes):
# for b, bbox in enumerate(gt_box):
# gt_boxes[g][b] = [cfg.patch_x, cfg.patch_y, cfg.patch_x+cfg.patch_w, cfg.patch_y+cfg.patch_h]
# forward
im_data = net_utils.np_to_variable(im,
is_cuda=True,
volatile=False).permute(0, 3, 1, 2)
bbox_pred, iou_pred, prob_pred = net(im_data, gt_boxes, gt_classes, dontcare, size_index)
# backward
loss = net.loss
bbox_loss += net.bbox_loss.data.cpu()#.numpy()[0]
iou_loss += net.iou_loss.data.cpu()#.numpy()[0]
cls_loss += net.cls_loss.data.cpu()#.numpy()[0]
train_loss += loss.data.cpu()#.numpy()[0]
optimizer.zero_grad()
loss.backward()
optimizer.step()
cnt += 1
step_cnt += 1
duration = t.toc()
if step % cfg.disp_interval == 0:
train_loss /= cnt
bbox_loss /= cnt
iou_loss /= cnt
cls_loss /= cnt
print(('epoch %d[%d/%d], loss: %.3f, bbox_loss: %.3f, iou_loss: %.3f, '
'cls_loss: %.3f (%.2f s/batch, rest:%s)' %
(imdb.epoch, step_cnt, batch_per_epoch, train_loss, bbox_loss,
iou_loss, cls_loss, duration,
str(datetime.timedelta(seconds=int((batch_per_epoch - step_cnt) * duration)))))) # noqa
if summary_writer and step % cfg.log_interval == 0:
summary_writer.add_scalar('loss_train', train_loss, step)
summary_writer.add_scalar('loss_bbox', bbox_loss, step)
summary_writer.add_scalar('loss_iou', iou_loss, step)
summary_writer.add_scalar('loss_cls', cls_loss, step)
summary_writer.add_scalar('learning_rate', lr, step)
# plot results
bbox_pred = bbox_pred.data[0:1].cpu().numpy()
iou_pred = iou_pred.data[0:1].cpu().numpy()
prob_pred = prob_pred.data[0:1].cpu().numpy()
image = im[0]
bboxes, scores, cls_inds = yolo_utils.postprocess(
bbox_pred, iou_pred, prob_pred, image.shape, cfg, thresh=0.3, size_index=size_index)
im2show = yolo_utils.draw_detection(image, bboxes, scores, cls_inds, cfg)
#summary_writer.add_image('predict', im2show, step)
train_loss = 0
bbox_loss, iou_loss, cls_loss = 0., 0., 0.
cnt = 0
t.clear()
size_index = randint(0, len(cfg.multi_scale_inp_size) - 1)
print("image_size {}".format(cfg.multi_scale_inp_size[size_index]))
if step > 0 and (step % imdb.batch_per_epoch == 0):
if imdb.epoch in cfg.lr_decay_epochs:
lr *= cfg.lr_decay
print("learning rate", lr)
optimizer = torch.optim.SGD([net.patch], lr=lr,
momentum=cfg.momentum,
weight_decay=cfg.weight_decay)
save_name = os.path.join(cfg.train_output_dir,
'{}_{}.h5'.format(cfg.exp_name, imdb.epoch))
net_utils.save_net(save_name, net)
print(('save model: {}'.format(save_name)))
step_cnt = 0
imdb.close()