forked from dazinovic/neural-rgbd-surface-reconstruction
-
Notifications
You must be signed in to change notification settings - Fork 0
/
load_network_model.py
34 lines (24 loc) · 1.11 KB
/
load_network_model.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
import os
from load_scannet import get_num_training_frames
import optimize
def load_network_model_from_disk(expname, iter, basedir='./logs'):
config = os.path.join(basedir, expname, 'config.txt')
print('Args:')
print(open(config, 'r').read())
parser = optimize.config_parser()
ft_str = ''
if iter is not None:
ft_str = '--ft_path {}'.format(os.path.join(basedir, expname, f'model_{iter:06}.npy'))
args = parser.parse_args('--config {} '.format(config) + ft_str)
args.num_training_frames = get_num_training_frames(args.datadir, trainskip=args.trainskip)
print(args.num_training_frames)
# Create nerf model
_, render_kwargs_test, _, _, models = optimize.create_nerf(args)
query_fn = render_kwargs_test['network_query_fn']
network_fn = render_kwargs_test['network_fn']
if args.N_importance > 0 and not args.share_coarse_fine:
network_fn = render_kwargs_test['network_fine']
feature_array = None
if 'feature_array' in models:
feature_array = models['feature_array']
return args, render_kwargs_test, query_fn, feature_array, network_fn