-
Notifications
You must be signed in to change notification settings - Fork 0
/
rock_train_c3_rgbd1.py
153 lines (133 loc) · 6.93 KB
/
rock_train_c3_rgbd1.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
"""
eureka_train.py
Zhiang Chen, Dec 25 2019
train mask rcnn with eureka data
"""
import transforms as T
from engine import train_one_epoch, evaluate
import utils
import torch
from c3 import Dataset
from model import get_rock_model_instance_segmentation
import numpy as np
torch.manual_seed(0)
class ToTensor(object):
def __call__(self, image, target):
# image = F.to_tensor(image).float()
image = torch.from_numpy(image / 255.0).float()
image = image.permute((2, 0, 1))
return image, target
def get_transform(train):
transforms = []
transforms.append(ToTensor()) # torchvision.transforms.functional is a garbage, sorry guys
return T.Compose(transforms)
'''
def get_mean_std(input_channel, image_mean, image_std):
if input_channel == 8:
return image_mean, image_std
elif input_channel == 3:
return image_mean[:3], image_std[:3]
elif input_channel == 5:
return image_mean[:5], image_std[:5]
elif input_channel == 6:
return image_mean[:3] + image_mean[-3:], image_std[:3] + image_mean[-3:]
elif input_channel == 4:
return image_mean[:3] + [np.mean(image_mean[-3:]).tolist()], image_std[:3] + [np.mean(image_std[-3:]).tolist()]
elif input_channel == 'dem':
return image_mean[-3:], image_std[-3:]
'''
def get_mean_std(input_channel, image_mean, image_std):
if input_channel == 6:
return image_mean, image_std
elif input_channel == 3:
return image_mean[:3], image_std[:3]
elif input_channel == 4:
return image_mean[:3] + [np.mean(image_mean[-3:]).tolist()], image_std[:3] + [np.mean(image_std[-3:]).tolist()]
elif input_channel == -3:
return image_mean[-3:], image_std[-3:]
elif input_channel == 1:
return [np.mean(image_mean[-3:]).tolist()], [np.mean(image_std[-3:]).tolist()]
if __name__ == '__main__':
# train on the GPU or on the CPU, if a GPU is not available
device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
device = torch.device('cuda:0')
# our dataset has three classes only - background, non-damaged, and damaged
num_classes = 2
input_c = 4
# use our dataset and defined transformations
dataset = Dataset("./datasets/iros/c3/aug/", transforms=get_transform(train=True), include_name=False, input_channel=input_c)
##dataset_test = Dataset("./datasets/Rock/data_test/", transforms=get_transform(train=False), include_name=False, input_channel=input_c)
#dataset_test = Dataset("./datasets/Rock_test/mult/", transforms=get_transform(train=False), include_name=False, input_channel=input_c)
dataset_test = Dataset("./datasets/iros/c3_test/rgbd_masks/", transforms=get_transform(train=False), include_name=False, input_channel=input_c)
# image_mean, image_std, _, _ = dataset.imageStat()
#image_mean = [0.2635908247051704, 0.2565450032962188, 0.24311759802366928, 0.3007502338171828, 0.35639093071269307, 0.24508291731782375]
#image_std = [0.23736204788409055, 0.23722317885795837, 0.22990199087409235, 0.25134661805921518, 0.26149370275679834, 0.26121346688112296]
#image_mean = [0.5561083275961, 0.4762376069094619, 0.4467060158591905, 0.10208016634218137, 0.5028495619538484, 0.5806285215840447]
#image_std = [0.23056259201528329, 0.21163784115481502, 0.19388422413947035, 0.0311074642376007, 0.18291436237313496, 0.20893878043377523]
image_mean = [0.5620081496860253, 0.48239326529405263, 0.452555695791186, 0.10258982219688167, 0.5069069552967871, 0.5892437452304018]
image_std = [0.22786734424312816, 0.21037678690930353, 0.19248594128629354, 0.030205125607848885, 0.1780133951799179, 0.20065141490899788]
image_mean, image_std = get_mean_std(input_c, image_mean, image_std)
# split the dataset in train and test set
#indices = torch.randperm(len(dataset)).tolist()
#dataset = torch.utils.data.Subset(dataset, indices)
#indices_test = torch.randperm(len(dataset_test)).tolist()
#dataset_test = torch.utils.data.Subset(dataset_test, indices_test)
# define training and validation data loaders
data_loader = torch.utils.data.DataLoader(
dataset, batch_size=4, shuffle=False, num_workers=8,
collate_fn=utils.collate_fn)
data_loader_test = torch.utils.data.DataLoader(
dataset_test, batch_size=1, shuffle=False, num_workers=4,
collate_fn=utils.collate_fn)
# get the model using our helper function
mask_rcnn = get_rock_model_instance_segmentation(num_classes, input_channel=input_c, image_mean=image_mean, image_std=image_std, pretrained=True)
read_param = False
if read_param:
mask_rcnn.load_state_dict(torch.load("trained_param_6/epoch_0050.param"))
print("Loaded weights")
# move model to the right device
mask_rcnn.to(device)
# construct an optimizer
params = [p for p in mask_rcnn.parameters() if p.requires_grad]
optimizer = torch.optim.SGD(params, lr=0.01, momentum=0.9, weight_decay=0.0005)
# and a learning rate scheduler
lr_scheduler = torch.optim.lr_scheduler.StepLR(optimizer,
step_size=3,
gamma=0.5)
init_epoch = 0
num_epochs = 12
save_param = "trained_param_c3_tl_rgbd1/epoch_{:04d}.param".format(init_epoch)
torch.save(mask_rcnn.state_dict(), save_param)
#'''
for epoch in range(init_epoch, init_epoch + num_epochs):
save_param = "trained_param_c3_tl_rgbd1/epoch_{:04d}.param".format(epoch)
#torch.save(mask_rcnn.state_dict(), save_param)
# train for one epoch, printing every 10 iterations
print(save_param)
train_one_epoch(mask_rcnn, optimizer, data_loader, device, epoch, print_freq=100)
# update the learning rate
lr_scheduler.step()
# evaluate on the test dataset
#print('\n')
#print("trained_param_4/epoch_00%02d.param" % epoch)
#mask_rcnn.load_state_dict(torch.load("trained_param_4/epoch_00%02d.param" % epoch))
evaluate(mask_rcnn, data_loader_test, device=device)
#save_param = "trained_param_8_fresh/epoch_{:04d}.param".format(epoch)
torch.save(mask_rcnn.state_dict(), save_param)
'''
for epoch in range(init_epoch, init_epoch + num_epochs):
#save_param = "trained_param_3_fresh/epoch_{:04d}.param".format(epoch)
#torch.save(mask_rcnn.state_dict(), save_param)
# train for one epoch, printing every 10 iterations
#train_one_epoch(mask_rcnn, optimizer, data_loader, device, epoch, print_freq=100)
# update the learning rate
#lr_scheduler.step()
# evaluate on the test dataset
print('\n')
name = "trained_param_8/epoch_00%02d.param" % epoch
print(name)
mask_rcnn.load_state_dict(torch.load(name))
evaluate(mask_rcnn, data_loader_test, device=device)
#save_param = "trained_param_8_fresh/epoch_{:04d}.param".format(epoch)
#torch.save(mask_rcnn.state_dict(), save_param)
'''