-
Notifications
You must be signed in to change notification settings - Fork 31
/
config.py
110 lines (95 loc) · 3.39 KB
/
config.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
import os
import os.path as osp
import sys
import time
import numpy as np
from easydict import EasyDict as edict
import argparse
C = edict()
config = C
cfg = C
C.seed = 12345
remoteip = os.popen('pwd').read()
C.root_dir = os.path.abspath(os.path.join(os.getcwd(), './'))
C.abs_dir = osp.realpath(".")
# Dataset config
"""Dataset Path"""
C.dataset_name = 'NYUDepthv2'
C.dataset_path = osp.join(C.root_dir, 'datasets', 'NYUDepthv2')
C.rgb_root_folder = osp.join(C.dataset_path, 'RGB')
C.rgb_format = '.jpg'
C.gt_root_folder = osp.join(C.dataset_path, 'Label')
C.gt_format = '.png'
C.gt_transform = True
# True when label 0 is invalid, you can also modify the function _transform_gt in dataloader.RGBXDataset
# True for most dataset valid, Faslse for MFNet(?)
C.x_root_folder = osp.join(C.dataset_path, 'HHA')
C.x_format = '.jpg'
C.x_is_single_channel = False # True for raw depth, thermal and aolp/dolp(not aolp/dolp tri) input
C.train_source = osp.join(C.dataset_path, "train.txt")
C.eval_source = osp.join(C.dataset_path, "test.txt")
C.is_test = False
C.num_train_imgs = 795
C.num_eval_imgs = 654
C.num_classes = 40
C.class_names = ['wall','floor','cabinet','bed','chair','sofa','table','door','window','bookshelf','picture','counter','blinds',
'desk','shelves','curtain','dresser','pillow','mirror','floor mat','clothes','ceiling','books','refridgerator',
'television','paper','towel','shower curtain','box','whiteboard','person','night stand','toilet',
'sink','lamp','bathtub','bag','otherstructure','otherfurniture','otherprop']
"""Image Config"""
C.background = 255
C.image_height = 480
C.image_width = 640
C.norm_mean = np.array([0.485, 0.456, 0.406])
C.norm_std = np.array([0.229, 0.224, 0.225])
""" Settings for network, this would be different for each kind of model"""
C.backbone = 'mit_b2' # Remember change the path below.
C.pretrained_model = C.root_dir + '/pretrained/segformer/mit_b2.pth'
C.decoder = 'MLPDecoder'
C.decoder_embed_dim = 512
C.optimizer = 'AdamW'
"""Train Config"""
C.lr = 6e-5
C.lr_power = 0.9
C.momentum = 0.9
C.weight_decay = 0.01
C.batch_size = 8
C.nepochs = 500
C.niters_per_epoch = C.num_train_imgs // C.batch_size + 1
C.num_workers = 16
C.train_scale_array = [0.5, 0.75, 1, 1.25, 1.5, 1.75]
C.warm_up_epoch = 10
C.fix_bias = True
C.bn_eps = 1e-3
C.bn_momentum = 0.1
"""Eval Config"""
C.eval_iter = 25
C.eval_stride_rate = 2 / 3
C.eval_scale_array = [1] # [0.75, 1, 1.25] #
C.eval_flip = False # True #
C.eval_crop_size = [480, 640] # [height weight]
"""Store Config"""
C.checkpoint_start_epoch = 250
C.checkpoint_step = 25
"""Path Config"""
def add_path(path):
if path not in sys.path:
sys.path.insert(0, path)
add_path(osp.join(C.root_dir))
C.log_dir = osp.abspath('log_' + C.dataset_name + '_' + C.backbone)
C.tb_dir = osp.abspath(osp.join(C.log_dir, "tb"))
C.log_dir_link = C.log_dir
C.checkpoint_dir = osp.abspath(osp.join(C.log_dir, "checkpoint"))
exp_time = time.strftime('%Y_%m_%d_%H_%M_%S', time.localtime())
C.log_file = C.log_dir + '/log_' + exp_time + '.log'
C.link_log_file = C.log_file + '/log_last.log'
C.val_log_file = C.log_dir + '/val_' + exp_time + '.log'
C.link_val_log_file = C.log_dir + '/val_last.log'
if __name__ == '__main__':
print(config.nepochs)
parser = argparse.ArgumentParser()
parser.add_argument(
'-tb', '--tensorboard', default=False, action='store_true')
args = parser.parse_args()
if args.tensorboard:
open_tensorboard()