-
Notifications
You must be signed in to change notification settings - Fork 5
/
eval_cd.py
308 lines (274 loc) · 12.8 KB
/
eval_cd.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
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
import math
import time
import numpy as np
import torch
import argparse
from pathlib import Path
from utils.logger import *
from utils.config import *
from utils.misc import *
import os
from tools import builder
from torch.utils.data import Dataset
import json
from pytorch3d.ops import sample_farthest_points, knn_points
class PairDataset_final(Dataset):
def __init__(self, args, eval_task, corrupt_level):
self.data_root = args.data_path
self.corrupt_level = corrupt_level - 1
self.data_list_file = os.path.join(self.data_root, 'test_list.json')
self.train_list_file = os.path.join(self.data_root, 'train_list.json')
print_log(f'[DATASET] Open file {self.data_list_file}', logger = 'Point_Painter_Dataset')
data_list = json.load(open(self.data_list_file))
train_list = json.load(open(self.train_list_file))
self.file_list = []
self.train_list = []
for line in data_list:
task = line.split('/')[1]
level = line.split('/')[3]
if task == eval_task and int(level[5]) == corrupt_level:
self.file_list.append({'task': task, 'file_path': line})
for line in train_list:
task = line.split('-', 1)[0]
if task == eval_task:
self.train_list.append({'task': task, 'file_path': line.split('-', 1)[1]})
self.task_dict = {}
for idx, file in enumerate(self.train_list):
task = file['task']
if task not in self.task_dict:
self.task_dict[task] = [idx]
else:
self.task_dict[task].append(idx)
print_log(f'[DATASET] {len(self.file_list)} instances were loaded', logger = 'Point_Painter_Dataset')
# print(len(self.file_list))
def random_rotate_together(self, pointcloud1, pointcloud2, level=0):
"""
Randomly rotate the point cloud
:param pointcloud: input point cloud
:param level: severity level
:return: corrupted point cloud
"""
angle_clip = math.pi / 3
angle_clip = angle_clip / 3 * (level + 1)
angles = np.random.uniform(-angle_clip, angle_clip, size=(3))
Rx = np.array([[1, 0, 0],
[0, np.cos(angles[0]), -np.sin(angles[0])],
[0, np.sin(angles[0]), np.cos(angles[0])]])
Ry = np.array([[np.cos(angles[1]), 0, np.sin(angles[1])],
[0, 1, 0],
[-np.sin(angles[1]), 0, np.cos(angles[1])]])
Rz = np.array([[np.cos(angles[2]), -np.sin(angles[2]), 0],
[np.sin(angles[2]), np.cos(angles[2]), 0],
[0, 0, 1]])
R = np.dot(Rz, np.dot(Ry, Rx))
pointcloud1 = np.dot(pointcloud1, R)
pointcloud2 = np.dot(pointcloud2, R)
return pointcloud1, pointcloud2
def random_rotate(self, pointcloud, level=0):
"""
Randomly rotate the point cloud
:param pointcloud: input point cloud
:param level: severity level
:return: corrupted point cloud
"""
angle_clip = math.pi / 3
angle_clip = angle_clip / 3 * (level + 1)
angles = np.random.uniform(-angle_clip, angle_clip, size=(3))
Rx = np.array([[1, 0, 0],
[0, np.cos(angles[0]), -np.sin(angles[0])],
[0, np.sin(angles[0]), np.cos(angles[0])]])
Ry = np.array([[np.cos(angles[1]), 0, np.sin(angles[1])],
[0, 1, 0],
[-np.sin(angles[1]), 0, np.cos(angles[1])]])
Rz = np.array([[np.cos(angles[2]), -np.sin(angles[2]), 0],
[np.sin(angles[2]), np.cos(angles[2]), 0],
[0, 0, 1]])
R = np.dot(Rz, np.dot(Ry, Rx))
pointcloud = np.dot(pointcloud, R)
return pointcloud
def y_flip(self, pointcloud1, pointcloud2):
angles = [0, 0, math.pi]
Rx = np.array([[1, 0, 0],
[0, np.cos(angles[0]), -np.sin(angles[0])],
[0, np.sin(angles[0]), np.cos(angles[0])]])
Ry = np.array([[np.cos(angles[1]), 0, np.sin(angles[1])],
[0, 1, 0],
[-np.sin(angles[1]), 0, np.cos(angles[1])]])
Rz = np.array([[np.cos(angles[2]), -np.sin(angles[2]), 0],
[np.sin(angles[2]), np.cos(angles[2]), 0],
[0, 0, 1]])
R = np.dot(Rz, np.dot(Ry, Rx))
pointcloud1 = np.dot(pointcloud1, R)
pointcloud2 = np.dot(pointcloud2, R)
return pointcloud1, pointcloud2
def y_flip_single(self, pointcloud1):
angles = [0, 0, math.pi]
Rx = np.array([[1, 0, 0],
[0, np.cos(angles[0]), -np.sin(angles[0])],
[0, np.sin(angles[0]), np.cos(angles[0])]])
Ry = np.array([[np.cos(angles[1]), 0, np.sin(angles[1])],
[0, 1, 0],
[-np.sin(angles[1]), 0, np.cos(angles[1])]])
Rz = np.array([[np.cos(angles[2]), -np.sin(angles[2]), 0],
[np.sin(angles[2]), np.cos(angles[2]), 0],
[0, 0, 1]])
R = np.dot(Rz, np.dot(Ry, Rx))
pointcloud1 = np.dot(pointcloud1, R)
return pointcloud1
def random_dropout_global(self, pointcloud, level=0):
"""
Drop random points globally
:param pointcloud: input point cloud
:param level: severity level
:return: corrupted point cloud
"""
# drop_rate = [0.25, 0.375, 0.5, 0.625, 0.75][level]
drop_rate = [0.5, 0.75, 0.875, 0.9375, 0.96875][level]
num_points = pointcloud.shape[0]
# choice = random.sample(range(0, num_points), int(drop_rate * num_points))
pointcloud[(1 - int(drop_rate * num_points)):, :] = 0
return pointcloud
def random_add_noise(self, pointcloud, level=0, sigma=0.2):
"""
Randomly add noise data to point cloud
:param pointcloud: input point cloud
:param num_noise: number of noise points
:return: corrupted point cloud
"""
N, _ = pointcloud.shape
num_noise = 100 * (level + 1)
noise = np.clip(sigma * np.random.randn(num_noise, 3), -1, 1)
idx = np.random.randint(0, N, num_noise)
pointcloud[idx, :3] = pointcloud[idx, :3] + noise
return pointcloud
def __getitem__(self, idx):
pointset1 = self.file_list[idx]
task = pointset1['task']
if task == "registration":
pointset1_pc = np.load(os.path.join(self.data_root, pointset1['file_path']))
else:
pointset1_pc = np.load(os.path.join(self.data_root, pointset1['file_path'])).astype(np.float32)
pointset2_index = random.choice(self.task_dict[task])
pointset2 = self.train_list[pointset2_index]
pointset2_pc = np.load(os.path.join(self.data_root, pointset2['file_path'])).astype(np.float32)
corrupt_level = self.corrupt_level
# print(corrupt_level)
# print(task)
if task == "reconstruction":
target1 = np.load(os.path.join(self.data_root, pointset1['file_path'].replace("sources", "targets"))).astype(np.float32)
target2 = pointset2_pc.copy()
pointset2_pc = self.random_dropout_global(pointset2_pc, corrupt_level)
elif task == "registration":
rotation_matrix = pointset1_pc["rotation_matrix"]
pointset1_pc = pointset1_pc["pointcloud"].astype(np.float32)
target1 = np.load(os.path.join(self.data_root, pointset1['file_path'].replace("sources", "targets")[:-1] + 'y')).astype(np.float32)
target2 = pointset2_pc.copy()
pointset2_pc = np.dot(pointset2_pc, rotation_matrix)
target2 = self.y_flip_single(target2)
elif task == "denoising":
target1 = np.load(os.path.join(self.data_root, pointset1['file_path'].replace("sources", "targets"))).astype(np.float32)
target2 = pointset2_pc.copy()
pointset2_pc = self.random_add_noise(pointset2_pc, corrupt_level)
else:
raise NotImplementedError()
pointset1_pc = torch.from_numpy(pointset1_pc).float()
pointset2_pc = torch.from_numpy(pointset2_pc).float()
target1 = torch.from_numpy(target1).float()
target2 = torch.from_numpy(target2).float()
return pointset2_pc, pointset1_pc, target2, target1
def __len__(self):
return len(self.file_list)
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument('--config', type=str, help='yaml config file')
parser.add_argument('--local_rank', type=int, default=0)
parser.add_argument('--num_workers', type=int, default=8)
# seed
parser.add_argument('--seed', type=int, default=0, help='random seed')
parser.add_argument('--deterministic', action='store_true', help='whether to set deterministic options for CUDNN backend.')
# some args
parser.add_argument('--exp_name', type=str, default='default', help='experiment name')
parser.add_argument('--loss', type=str, default='cd2', help='loss name')
parser.add_argument('--ckpts', type=str, default=None, help='test used ckpt path')
parser.add_argument('--val_freq', type=int, default=1, help='test freq')
# dataset
parser.add_argument('--data_path', type=str, default='data', help='')
# comment
parser.add_argument('--comment', type=str, default='default', help='')
parser.add_argument(
'--resume',
action='store_true',
default=False,
help = 'autoresume training (interrupted by accident)')
args = parser.parse_args()
args.experiment_path = args.exp_name
args.log_name = Path(args.config).stem
if not os.path.exists(args.experiment_path):
os.makedirs(args.experiment_path)
print('Create experiment path successfully at %s' % args.experiment_path)
return args
def eval_cd(args, config, base_model, eval_task, corrupt_level):
test_dataset = PairDataset_final(args, eval_task, corrupt_level)
test_dataloader = torch.utils.data.DataLoader(test_dataset, batch_size=config.total_bs * 2,
shuffle=False,
drop_last=False,
num_workers=int(args.num_workers),
worker_init_fn=worker_init_fn)
loss = eval(base_model, test_dataloader)
return loss
def eval(base_model, test_dataloader):
base_model.eval() # set model to eval mode
mean_loss = 0
i = 0
with torch.no_grad():
for idx, (pointset1_pc, pointset2_pc, target1, target2) in enumerate(test_dataloader):
pointset1_pc = pointset1_pc.cuda()
pointset2_pc = pointset2_pc.cuda()
target1 = target1.cuda()
target2 = target2.cuda()
_, rebuild_points, loss = base_model(pointset1_pc, pointset2_pc, target1, target2)
rebuild_points, _ = sample_farthest_points(rebuild_points, K=target2.shape[1])
loss = base_model.loss_func(rebuild_points, target2)
loss = loss.mean()
mean_loss += loss.item()*1000
i += 1
mean_loss /= i
return mean_loss
def main():
# args
args = get_args()
# CUDA
args.use_gpu = torch.cuda.is_available()
# logger
timestamp = time.strftime('%Y%m%d_%H%M%S', time.localtime())
log_file = os.path.join(args.experiment_path, f'{timestamp}-{args.seed}.log')
logger = get_root_logger(log_file=log_file, name=args.log_name)
# config
config = get_config(args, logger = logger)
# log
log_args_to_file(args, 'args', logger=logger)
log_config_to_file(config, 'config', logger=logger)
print_log(args.comment)
# set random seeds
if args.seed is not None:
logger.info(f'Set random seed to {args.seed}, '
f'deterministic: {args.deterministic}')
set_random_seed(args.seed + args.local_rank, deterministic=args.deterministic) # seed + rank, for augmentation
base_model = builder.model_builder(config.model)
# load checkpoints
builder.load_model(base_model, args.ckpts, logger=logger)
if args.use_gpu:
base_model.to(args.local_rank)
tasks = ["reconstruction", "denoising", "registration"]
for eval_task in tasks:
task_loss = 0.
print_log("------------------[{}] Tester start !!!!!!!!----------------- ".format(eval_task.upper()), logger=logger)
for corrupt_level in range(5):
loss = eval_cd(args, config, base_model, eval_task, corrupt_level+1)
print_log('[TEST] corrupt_level: {} loss = {:.4f}'.format(corrupt_level+1, loss), logger=logger)
task_loss += loss
task_loss /= 5
print_log('[TEST {}] loss = {:.4f}'.format(eval_task.upper(), task_loss), logger=logger)
print_log("Done!!!!!!!!!!!!!!!!!")
if __name__ == "__main__":
main()