-
Notifications
You must be signed in to change notification settings - Fork 68
/
train.py
executable file
·165 lines (141 loc) · 6.86 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
161
162
163
164
165
#
# Authors: Bowen Wen
# Contact: wenbowenxjtu@gmail.com
# Created in 2020
#
# Copyright (c) Rutgers University, 2020 All rights reserved.
#
# Wen, B., C. Mitash, B. Ren, and K. E. Bekris. "se (3)-TrackNet:
# Data-driven 6D Pose Tracking by Calibrating Image Residuals in
# Synthetic Domains." In IEEE/RSJ International Conference on Intelligent
# Robots and Systems (IROS). 2020.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
# * Redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution.
# * Neither the name of the PRACSYS, Bowen Wen, Rutgers University,
# nor the names of its contributors may be used to
# endorse or promote products derived from this software without
# specific prior written permission.
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 'AS IS' AND
# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE FOR ANY
# DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
# (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
# LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
# ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
# SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
import open3d as o3d
import sys,shutil
import os
dir_path = os.path.dirname(os.path.realpath(__file__))
sys.path.append(dir_path)
from datasets import *
from multiprocessing import cpu_count
import argparse
from problems import *
from network_modules import *
from se3_tracknet import *
import torch
from torch import optim
import torch.utils.data
import numpy as np
import yaml
import glob
import random
from data_augmentation import *
with open(dir_path+'/config.yml', 'r') as ff:
config = yaml.safe_load(ff)
random.seed(0)
np.random.seed(0)
torch.manual_seed(0)
torch.backends.cudnn.benchmark = True
if __name__=="__main__":
data_path = config['data_path']
validation_path = config['validation_path']
dir_path = os.path.dirname(os.path.realpath(__file__))
output_path = f"{dir_path}/train_output/"
print('output_path',output_path)
if not os.path.exists(output_path):
os.makedirs(output_path)
with open(data_path+'/../dataset_info.yml', 'r') as ff:
dataset_info = yaml.safe_load(ff)
print('loaded dataset info from:',data_path+'/../dataset_info.yml')
shutil.copy(data_path+'/../dataset_info.yml',f'{output_path}/dataset_info.yml')
hsv_noise = config['data_augmentation']['hsv_noise']
batch_size = config['batch_size']
n_workers = config['n_workers']
augmentations = Compose([
HSVJitter(hsv_noise[0],hsv_noise[1],hsv_noise[2]),
ChangeBright(prob=0.5,mag=[config['data_augmentation']['bright_mag'][0], config['data_augmentation']['bright_mag'][1]]),
GaussianNoise(config['data_augmentation']['gaussian_noise']['rgb'], config['data_augmentation']['gaussian_noise']['depth']),
GaussianBlur(config['data_augmentation']['gaussian_blur_kernel']),
BlackCover(prob=0.2),
# DepthMissing(prob=0.5,missing_percent=config['data_augmentation']['depth_missing_percent']),
])
############ Compute mean std
posttransforms = Compose([OffsetDepth(),Transpose(),ToTensor()]) #NOTE Transpose() make image CxHxW
train_dataset = TrackDataset(data_path,mode='train',images_mean=None,images_std=None,pretransforms=None, augmentations=augmentations, posttransforms=posttransforms, dataset_info=dataset_info, trans_normalizer=dataset_info['max_translation'], rot_normalizer=dataset_info['max_rotation']*np.pi/180)
print('len(train_dataset)=',len(train_dataset))
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=batch_size,
shuffle=False,
num_workers=n_workers,
pin_memory=False,
drop_last=True,
)
n = 10000
print('Computing mean std for n={}'.format(n))
total = 0
image_means_array = []
for i, (data, target, A_in_cams, B_in_cams, rgbA, rgbB, maskA, maskB) in enumerate(train_loader):
bufferA, bufferB = data
bufferA_numpy = bufferA.cpu().numpy()[:,:4]
bufferB_numpy = bufferB.cpu().numpy()[:,:4]
buffer_numpy = np.concatenate((bufferA_numpy, bufferB_numpy), axis=1) #NxCxHxW
image_means = np.mean(buffer_numpy, axis=(0, 2, 3))
image_means_array.append(image_means)
total += 1
if i * batch_size >= n:
break
images_mean = np.mean(image_means_array, axis=0)
images_std = np.std(image_means_array, axis=0)
np.save("{}/mean.npy".format(output_path), images_mean)
np.save("{}/std.npy".format(output_path), images_std)
print('images_mean\n',images_mean)
print('images_std\n',images_std)
posttransforms = Compose([OffsetDepth(),NormalizeChannels(images_mean,images_std),ToTensor()])
train_dataset = TrackDataset(data_path,'train',images_mean,images_std,None,augmentations,posttransforms,dataset_info=dataset_info, trans_normalizer=dataset_info['max_translation'], rot_normalizer=dataset_info['max_rotation']*np.pi/180)
valid_dataset = TrackDataset(validation_path,'val',images_mean,images_std,None,augmentations,posttransforms,dataset_info=dataset_info, trans_normalizer=dataset_info['max_translation'], rot_normalizer=dataset_info['max_rotation']*np.pi/180)
with open(os.path.join(output_path, "config_backup.yml"), 'w') as ff:
yaml.dump(config, ff)
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=batch_size,
shuffle=True,
num_workers=n_workers,
pin_memory=False,
drop_last=True,
)
assert len(train_loader)>0
val_loader = torch.utils.data.DataLoader(valid_dataset,
batch_size=200,
shuffle=False,
num_workers=n_workers,
pin_memory=False,
drop_last=False,
)
model = Se3TrackNet(image_size=int(dataset_info["resolution"]))
model = model.cuda()
optimizer = optim.Adam(filter(lambda p: p.requires_grad, model.parameters()), lr=config['learning_rate'], weight_decay=config['weight_decay'], betas=(0.9, 0.99), amsgrad=False)
scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones=[100,200,300], gamma=0.1)
problem = Problem(model, train_loader, val_loader,config=config,optimizer=optimizer, scheduler=scheduler)
print("Training Begins:")
problem.loop(config['epochs'], output_path, save_all_checkpoints=True)
print("Training Complete")