-
Notifications
You must be signed in to change notification settings - Fork 9
/
train_AACVPMVSNet.py
222 lines (187 loc) · 7.51 KB
/
train_AACVPMVSNet.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
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2020/8/3 0016 11:52
# @Author : Anzhu Yu
# @Site :
# @File : train_AACVPMVSNet.py
# @Software: PyCharm
# some packages used in this project
from argsParser import getArgsParser, checkArgs
import os
import logging
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.backends.cudnn as cudnn
import torch.optim as optim
from datasets import dtu_loader
from torch.utils.data import DataLoader
from torch.autograd import Variable
import torch.nn.functional as F
import numpy as np
import time
from tensorboardX import SummaryWriter
from datasets import find_dataset_def
from models import *
from utils import *
import gc
import sys
import datetime
import torch.utils
import torch.utils.checkpoint
from torchscan import summary
# CUDA_LAUNCH_BLOCKING=1
parser = getArgsParser()
args = parser.parse_args()
assert args.mode == "train", 'HERE IS THE TRAINING MODE!'
checkArgs(args)
torch.manual_seed(args.seed)
torch.cuda.manual_seed(args.seed)
cudnn.benchmark = True
# logger
logger = logging.getLogger()
logger.setLevel(logging.INFO)
curTime = time.strftime('%Y%m%d-%H%M', time.localtime(time.time()))
log_path = args.loggingdir + args.info.replace(" ", "_") + "/"
if not os.path.isdir(args.loggingdir):
os.mkdir(args.loggingdir)
if not os.path.isdir(log_path):
os.mkdir(log_path)
log_name = log_path + curTime + '.log'
logfile = log_name
formatter = logging.Formatter("%(asctime)s - %(filename)s[line:%(lineno)d] - %(levelname)s: %(message)s")
fileHandler = logging.FileHandler(logfile, mode='a')
fileHandler.setFormatter(formatter)
logger.addHandler(fileHandler)
consoleHandler = logging.StreamHandler(sys.stdout)
consoleHandler.setFormatter(formatter)
logger.addHandler(consoleHandler)
logger.info("Logger initialized.")
logger.info("Writing logs to file:" + logfile)
settings_str = "All settings:\n"
line_width = 20
for k, v in vars(args).items():
settings_str += '{0}: {1}\n'.format(k, v)
logger.info(settings_str)
# Read the Data,
train_dataset = dtu_loader.MVSDataset(args, logger)
train_loader = DataLoader(train_dataset, args.batch_size, shuffle=True, num_workers=16, drop_last=True)
# Build the model
model = AACVPMVSNet(args, group=args.groups, num_heads=args.num_heads)
# Use the cuda_ids to determine the GPUs used for experiments
device_id_list = [int(idd) for idd in args.cuda_ids.split(',')]
if len(device_id_list) == 1 and (device_id_list[0] != 666):
print("Now multi-GPUs mode activated!")
device_ids = [int(args.cuda_ids)]
elif (device_id_list[0] == 666):
model = model.cpu()
else:
device_ids = device_id_list
del device_id_list
# GPUs
if args.mode == "train" and torch.cuda.is_available():
model = nn.DataParallel(model, device_ids=device_ids, output_device=device_ids[0])
if torch.cuda.is_available():
model.cuda()
model.train()
if args.loss_function == "sl1":
logger.info("Using smoothed L1 loss")
model_loss = sL1_loss
else: # MSE
logger.info("Using MSE loss")
model_loss = MSE_loss
logger.info(">>> total params: {:.2f}M".format(sum(p.numel() for p in model.parameters()) / 1000000.0))
# model_loss = mvsnet_loss
optimizer = optim.Adam(model.parameters(), lr=args.lr, betas=(0.9, 0.999), weight_decay=args.wd)
# Start at a given checkpoint
sw_path = args.logckptdir + args.info + "/"
start_epoch = 0
if (args.mode == "train" and args.resume) or (args.mode == "test" and not args.loadckpt):
logger.info("Resuming or testing...")
saved_models = [fn for fn in os.listdir(sw_path) if fn.endswith(".ckpt")]
saved_models = sorted(saved_models, key=lambda x: int(x.split('_')[-1].split('.')[0]))
# use the latest checkpoint file
loadckpt = os.path.join(sw_path, saved_models[-1])
logger.info("Resuming " + loadckpt)
state_dict = torch.load(loadckpt)
model.load_state_dict(state_dict['model'])
optimizer.load_state_dict(state_dict['optimizer'])
start_epoch = state_dict['epoch'] + 1
elif args.loadckpt:
# load checkpoint file specified by args.loadckpt
logger.info("loading model {}".format(args.loadckpt))
state_dict = torch.load(args.loadckpt)
model.load_state_dict(state_dict['model'])
print("start at epoch {}".format(start_epoch))
print('Number of model parameters: {}'.format(sum([p.data.nelement() for p in model.parameters()])))
# Training at each epoch
def train():
milestones = [int(epoch_idx) for epoch_idx in args.lrepochs.split(':')[0].split(',')]
lr_gamma = 1 / float(args.lrepochs.split(':')[1])
lr_scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones, gamma=lr_gamma,
last_epoch=start_epoch - 1)
# epoch stat
last_loss = None
this_loss = None
for epoch_idx in range(start_epoch, args.epochs):
logger.info('Epoch {}:'.format(epoch_idx))
global_step = len(train_loader) * epoch_idx
if last_loss is None:
last_loss = 999999
else:
last_loss = this_loss
this_loss = []
for batch_idx, sample in enumerate(train_loader):
start_time = time.time()
global_step = len(train_loader) * epoch_idx + batch_idx
do_summary = global_step % args.summary_freq == 0
loss = train_sample(sample, detailed_summary=do_summary)
this_loss.append(loss)
logger.info(
'Epoch {}/{}, Iter {}/{}, train loss = {:.3f}, time = {:.3f}'.format(epoch_idx, args.epochs, batch_idx,
len(train_loader), loss,
time.time() - start_time))
# checkpoint
if (epoch_idx + 1) % args.save_freq == 0:
torch.save({
'epoch': epoch_idx,
'model': model.state_dict(),
'optimizer': optimizer.state_dict()},
"{}/model_{:0>6}.ckpt".format(args.logckptdir + args.info.replace(" ", "_"), epoch_idx))
logger.info("model_{:0>6}.ckpt saved".format(epoch_idx))
this_loss = np.mean(this_loss)
logger.info("Epoch loss: {:.5f} --> {:.5f}".format(last_loss, this_loss))
lr_scheduler.step()
# Training for each batch
def train_sample(sample, detailed_summary=False):
"""
:param sample: each batch
:param detailed_summary: whether the detailed logs are needed.
:return: the loss
"""
# model.train() is not needed here, however it is often used to state this script is not for evaluation.
model.train()
optimizer.zero_grad()
sample_cuda = tocuda(sample)
ref_depths = sample_cuda["ref_depths"]
# forward
outputs = model(sample_cuda["ref_img"].float(), sample_cuda["src_imgs"].float(), sample_cuda["ref_intrinsics"],
sample_cuda["src_intrinsics"], sample_cuda["ref_extrinsics"], sample_cuda["src_extrinsics"],
sample_cuda["depth_min"], sample_cuda["depth_max"])
depth_est_list = outputs["depth_est_list"]
dHeight = ref_depths.shape[2]
dWidth = ref_depths.shape[3]
loss = []
for i in range(0, args.nscale):
# generate the masks.
depth_gt = ref_depths[:, i, :int(dHeight / 2 ** i), :int(dWidth / 2 ** i)]
mask = depth_gt > 425
loss.append(model_loss(depth_est_list[i], depth_gt.float(), mask))
loss = sum(loss)
loss.backward()
optimizer.step()
return loss.data.cpu().item()
# main function, the start of this program
if __name__ == '__main__':
if args.mode == "train":
train()