forked from wenbowen123/iros20-6d-pose-tracking
-
Notifications
You must be signed in to change notification settings - Fork 0
/
problems.py
156 lines (133 loc) · 5.46 KB
/
problems.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
#
# Authors: Bowen Wen
# Contact: [email protected]
# 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 os,sys,gc
import glob
dir_path = os.path.dirname(os.path.realpath(__file__))
sys.path.append(dir_path)
from data_augmentation import *
from network_modules import *
from Utils import *
from tensorboardX import SummaryWriter
import time
import torch
torch.multiprocessing.set_sharing_strategy('file_system')
import torch.nn.functional as F
import torch.nn as nn
from torch.autograd import Variable
import numpy as np
import cv2
from PIL import Image
torch.backends.cudnn.benchmark = False
torch.backends.cudnn.deterministic = True
class Problem:
def __init__(self, model, train_data_loader, valid_data_loader, config=None, optimizer=None, scheduler=None):
self.train_data = train_data_loader
self.valid_data = valid_data_loader
self.optimizer = optimizer
self.model = model
self.scheduler = scheduler
self.global_step = 0
self.model = self.model.cuda()
self.config = config
self.loss_weights = self.config['loss_weights']
self.best_eval = np.inf
self.best_val = np.inf
self.best_train = np.inf
self.dataset_info = train_data_loader.dataset.dataset_info
self.K = self.train_data.dataset.cam_K
def train(self, epoch):
self.model.train()
for data, target, A_in_cams, B_in_cams, rgbA, rgbB, maskA, maskB in self.train_data:
dataA = data[0]
dataB = data[1]
dataA = dataA.cuda()
dataB = dataB.cuda()
for i in range(len(target)):
target[i] = target[i].cuda()
pred = self.model(dataA,dataB)
output = self.model.loss((pred['trans'],pred['rot']), target)
loss = output['trans']*self.loss_weights['trans']+output['rot']*self.loss_weights['rot']
self.optimizer.zero_grad()
loss.backward()
self.optimizer.step()
if self.global_step%100==0:
print("epoch={}, iter={}, loss={}".format(epoch, self.global_step, loss.data))
print('trans_label[0] =',target[0][0].data.cpu().numpy().reshape(-1))
print('rot_label[0] =',target[1][0].data.cpu().numpy().reshape(-1))
self.global_step += 1
return loss.data
def validate(self, epoch):
self.model.eval()
trans_losses = []
rot_losses = []
trans_uncertain_losses = []
rot_uncertain_losses = []
seg_losses = []
adi_losses = []
with torch.no_grad():
for data, target, A_in_cams, B_in_cams, rgbA, rgbB, maskA, maskB in self.valid_data:
dataA = data[0]
dataB = data[1]
dataA = dataA.cuda()
dataB = dataB.cuda()
for i in range(len(target)):
target[i] = target[i].cuda()
pred = self.model(dataA,dataB)
output = self.model.loss((pred['trans'],pred['rot']), target)
trans_losses.append(output['trans'].cpu().item())
rot_losses.append(output['rot'].cpu().item())
trans_loss = np.array(trans_losses).mean()
rot_loss = np.array(rot_losses).mean()
total_loss = trans_loss*self.loss_weights['trans'] + rot_loss*self.loss_weights['rot']
return total_loss
def loop(self, total_epochs, output_path, save_all_checkpoints=False):
for epoch in range(0, total_epochs):
print(">>>>>>>>>>>>>> epoch {}".format(epoch))
train_loss = self.train(epoch)
validation_loss_average = self.validate(epoch)
if train_loss<self.best_train:
self.best_train = train_loss
checkpoint_data = {'state_dict': self.model.state_dict()}
dir = "{}/model_best_train.pth.tar".format(output_path)
torch.save(checkpoint_data, dir)
is_val_best = validation_loss_average < self.best_val
self.best_val = min(validation_loss_average, self.best_val)
if is_val_best:
checkpoint_data = {'state_dict': self.model.state_dict()}
dir = "{}/model_best_val.pth.tar".format(output_path)
torch.save(checkpoint_data, dir)
if self.scheduler:
self.scheduler.step()