-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathmain_scan.py
337 lines (305 loc) · 15.3 KB
/
main_scan.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
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
@Author: Yue Wang
@Contact: [email protected]
@File: main_cls.py
@Time: 2018/10/13 10:39 PM
Modified by
@Author: An Tao
@Contact: [email protected]
@Time: 2019/12/30 9:32 PM
"""
from __future__ import print_function
import os
import argparse
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.optim.lr_scheduler import CosineAnnealingLR, StepLR
from data import ScanObject
from model import PointNet_scan, DGCNN_scan
import numpy as np
from torch.utils.data import DataLoader
from util import cal_loss, compute_chamfer_distance, IOStream
import sklearn.metrics as metrics
class CrossEntropyLossSeg(nn.Module):
def __init__(self, weight=None, size_average=True):
super(CrossEntropyLossSeg, self).__init__()
self.nll_loss = nn.NLLLoss(weight, size_average)
def forward(self, inputs, targets):
'''
:param inputs: BxclassxN
:param targets: BxN
:return:
'''
inputs = inputs.unsqueeze(3)
targets = targets.unsqueeze(2)
return self.nll_loss(F.log_softmax(inputs, dim=1), targets)
def _init_():
ckpt_dir = BASE_DIR + '/ckpt/scan'
if not os.path.exists(ckpt_dir):
os.makedirs(ckpt_dir)
if not os.path.exists(ckpt_dir + '/' + args.exp_name):
os.makedirs(ckpt_dir + '/' + args.exp_name)
if not os.path.exists(ckpt_dir + '/' + args.exp_name + '/' + 'models'):
os.makedirs(ckpt_dir + '/' + args.exp_name + '/' + 'models')
if not os.path.exists(ckpt_dir + '/' + args.exp_name + '/' + 'visu'):
os.makedirs(ckpt_dir + '/' + args.exp_name + '/' + 'visu')
os.system('cp main_scan.py ' + ckpt_dir + '/' + args.exp_name + '/' + 'main_scan.py.backup')
os.system('cp model.py ' + ckpt_dir + '/' + args.exp_name + '/' + 'model.py.backup')
os.system('cp util.py ' + ckpt_dir + '/' + args.exp_name + '/' + 'util.py.backup')
os.system('cp data.py ' + ckpt_dir + '/' + args.exp_name + '/' + 'data.py.backup')
def train(args, io):
train_loader = DataLoader(
ScanObject(h5_filename=BASE_DIR + '/data/scanobjectnn/{}'.format(args.file_name), num_points=args.num_points),
num_workers=8,
batch_size=args.batch_size, shuffle=True, drop_last=True)
test_loader = DataLoader(
ScanObject(h5_filename=BASE_DIR + '/data/scanobjectnn/{}'.format(args.file_name.replace('training', 'test')),
num_points=args.num_points), num_workers=8, batch_size=args.test_batch_size, shuffle=True,
drop_last=False)
device = torch.device("cuda" if args.cuda else "cpu")
#Try to load models
if args.model == 'pointnet':
model = PointNet_scan(args).to(device)
elif args.model == 'dgcnn':
model = DGCNN_scan(args).to(device)
else:
raise Exception("Not implemented")
print(str(model))
model = nn.DataParallel(model)
softmax_segmenter = CrossEntropyLossSeg()
softmax_segmenter.to(device)
seg_num_all = 2
print("Let's use", torch.cuda.device_count(), "GPUs!")
if args.use_sgd:
print("Use SGD")
opt = optim.SGD(model.parameters(), lr=args.lr*100, momentum=args.momentum, weight_decay=1e-4)
else:
print("Use Adam")
opt = optim.Adam(model.parameters(), lr=args.lr, weight_decay=1e-4)
if args.scheduler == 'cos':
scheduler = CosineAnnealingLR(opt, args.epochs, eta_min=1e-3)
elif args.scheduler == 'step':
scheduler = StepLR(opt, step_size=20, gamma=0.7)
criterion = cal_loss
best_test_acc = 0
for epoch in range(args.epochs):
####################
# Train
####################
train_loss = 0.0
train_cd_loss = 0.0
train_cls_loss = 0.0
train_seg_loss = 0.0
count = 0.0
model.train()
train_pred = []
train_true = []
for data, label, seg in train_loader:
label = label.type(torch.LongTensor)
seg = seg.type(torch.LongTensor)
data, label, seg = data.to(device), label.to(device), seg.to(device)
# data = data.permute(0, 2, 1)
batch_size = data.size()[0]
opt.zero_grad()
logits_cls, logits_seg, node1, node1_static = model(data)
loss_cls = criterion(logits_cls, label)
logits_seg = logits_seg.permute(0, 2, 1).contiguous()
loss_seg = criterion(logits_seg.view(-1, seg_num_all), seg.view(-1, 1).squeeze())
# loss_seg = softmax_segmenter(logits_seg, seg)
loss_cd = compute_chamfer_distance(node1, data)
loss = loss_cls + loss_seg + args.cd_weights * loss_cd
loss.backward()
opt.step()
preds = logits_cls.max(dim=1)[1]
count += batch_size
train_loss += loss.item() * batch_size
train_cls_loss += loss_cls.item() * batch_size
train_seg_loss += loss_seg.item() * batch_size
train_cd_loss += loss_cd.item() * batch_size
train_true.append(label.cpu().numpy())
train_pred.append(preds.detach().cpu().numpy())
if args.scheduler == 'cos':
scheduler.step()
elif args.scheduler == 'step':
if opt.param_groups[0]['lr'] > 1e-5:
scheduler.step()
if opt.param_groups[0]['lr'] < 1e-5:
for param_group in opt.param_groups:
param_group['lr'] = 1e-5
train_true = np.concatenate(train_true)
train_pred = np.concatenate(train_pred)
outstr = 'Train %d, loss: %.6f, loss_cls: %.6f, loss_seg: %.6f, loss_cd: %.6f, train acc: %.6f, train avg acc: %.6f' \
% (epoch,
train_loss * 1.0 / count,
train_cls_loss * 1.0 / count,
train_seg_loss * 1.0 / count,
train_cd_loss * 1.0 / count,
metrics.accuracy_score(
train_true, train_pred),
metrics.balanced_accuracy_score(
train_true, train_pred))
io.cprint(outstr)
####################
# Test
####################
test_loss = 0.0
test_cd_loss = 0.0
test_cls_loss = 0.0
test_seg_loss = 0.0
count = 0.0
model.eval()
test_pred = []
test_true = []
with torch.no_grad():
for data, label, seg in test_loader:
label = label.type(torch.LongTensor)
seg = seg.type(torch.LongTensor)
data, label, seg = data.to(device), label.to(device), seg.to(device)
# data = data.permute(0, 2, 1)
batch_size = data.size()[0]
logits_cls, logits_seg, node1, node1_static = model(data)
loss_cls = criterion(logits_cls, label)
logits_seg = logits_seg.permute(0, 2, 1).contiguous()
loss_seg = criterion(logits_seg.view(-1, seg_num_all), seg.view(-1, 1).squeeze())
# loss_seg = softmax_segmenter(logits_seg, seg)
loss_cd = compute_chamfer_distance(node1, data)
loss = loss_cls + loss_seg + args.cd_weights * loss_cd
preds = logits_cls.max(dim=1)[1]
count += batch_size
test_loss += loss.item() * batch_size
test_cls_loss += loss_cls.item() * batch_size
test_seg_loss += loss_seg.item() * batch_size
test_cd_loss += loss_cd.item() * batch_size
test_true.append(label.cpu().numpy())
test_pred.append(preds.detach().cpu().numpy())
test_true = np.concatenate(test_true)
test_pred = np.concatenate(test_pred)
test_acc = metrics.accuracy_score(test_true, test_pred)
avg_per_class_acc = metrics.balanced_accuracy_score(test_true, test_pred)
outstr = 'Test %d, loss: %.6f, loss_cls: %.6f, loss_seg: %.6f, loss_cd: %.6f, test acc: %.6f, test avg acc: %.6f' \
% (epoch,
test_loss * 1.0 / count,
test_cls_loss * 1.0 / count,
test_seg_loss * 1.0 / count,
test_cd_loss * 1.0 / count,
test_acc,
avg_per_class_acc)
io.cprint(outstr)
if test_acc >= best_test_acc:
best_test_acc = test_acc
torch.save(model.state_dict(), BASE_DIR + '/ckpt/scan/%s/models/model.t7' % args.exp_name)
def test(args, io):
test_loader = DataLoader(
ScanObject(h5_filename=BASE_DIR + '/data/scanobjectnn/{}'.format(args.file_name.replace('training', 'test')),
num_points=args.num_points), batch_size=args.test_batch_size, shuffle=False, drop_last=False)
device = torch.device("cuda" if args.cuda else "cpu")
#Try to load models
if args.model == 'pointnet':
model = PointNet_scan(args).to(device)
elif args.model == 'dgcnn':
model = DGCNN_scan(args).to(device)
else:
raise Exception("Not implemented")
model = nn.DataParallel(model)
model.load_state_dict(torch.load(args.model_path))
model = model.eval()
test_true = []
test_pred = []
count = 0
for data, label, seg in test_loader:
count += 1
data, label, seg = data.to(device), label.to(device).squeeze(), seg.to(device)
# data = data.permute(0, 2, 1)
logits_cls, logits_seg, node1, node1_static = model(data)
preds = logits_cls.max(dim=1)[1]
test_true.append(label.cpu().numpy())
test_pred.append(preds.detach().cpu().numpy())
if args.visu:
for i in range(data.shape[0]):
np.save(
BASE_DIR + '/ckpt/scan/%s/visu/node0_%02d_%04d.npy' % (args.exp_name, label[i], count * args.test_batch_size + i),
data[i, :, :].detach().cpu().numpy())
np.save(
BASE_DIR + '/ckpt/scan/%s/visu/node1_%02d_%04d.npy' % (args.exp_name, label[i], count * args.test_batch_size + i),
node1[i, :, :].detach().cpu().numpy())
test_true = np.concatenate(test_true)
test_pred = np.concatenate(test_pred)
test_acc = metrics.accuracy_score(test_true, test_pred)
avg_per_class_acc = metrics.balanced_accuracy_score(test_true, test_pred)
outstr = 'Test :: test acc: %.6f, test avg acc: %.6f'%(test_acc, avg_per_class_acc)
io.cprint(outstr)
if __name__ == "__main__":
# Training settings
parser = argparse.ArgumentParser(description='Point Cloud Recognition')
parser.add_argument('--base_dir', type=str, default='/opt/data/private', metavar='N',
help='Path to data and ckpt')
parser.add_argument('--exp_name', type=str, default='exp', metavar='N',
help='Name of the experiment')
parser.add_argument('--model', type=str, default='dgcnn', metavar='N',
choices=['pointnet', 'dgcnn'],
help='Model to use, [pointnet, dgcnn]')
parser.add_argument('--dataset', type=str, default='modelnet40', metavar='N',
choices=['modelnet40'])
parser.add_argument('--batch_size', type=int, default=32, metavar='batch_size',
help='Size of batch)')
parser.add_argument('--test_batch_size', type=int, default=16, metavar='batch_size',
help='Size of batch)')
parser.add_argument('--epochs', type=int, default=250, metavar='N',
help='number of episode to train ')
parser.add_argument('--use_sgd', type=bool, default=True,
help='Use SGD')
parser.add_argument('--lr', type=float, default=0.001, metavar='LR',
help='learning rate (default: 0.001, 0.1 if using sgd)')
parser.add_argument('--momentum', type=float, default=0.9, metavar='M',
help='SGD momentum (default: 0.9)')
parser.add_argument('--scheduler', type=str, default='cos', metavar='N',
choices=['cos', 'step'],
help='Scheduler to use, [cos, step]')
parser.add_argument('--no_cuda', type=bool, default=False,
help='enables CUDA training')
parser.add_argument('--seed', type=int, default=1, metavar='S',
help='random seed (default: 1)')
parser.add_argument('--eval', type=bool, default=False,
help='evaluate the model')
parser.add_argument('--num_points', type=int, default=1024,
help='num of points to use')
parser.add_argument('--dropout', type=float, default=0.5,
help='initial dropout rate')
parser.add_argument('--emb_dims', type=int, default=1024, metavar='N',
help='Dimension of embeddings')
parser.add_argument('--k', type=int, default=20, metavar='N',
help='Num of nearest neighbors to use')
parser.add_argument('--model_path', type=str, default='', metavar='N',
help='Pretrained model path')
parser.add_argument('--visu', type=bool, default=False,
help='visualize atp by saving nodes')
parser.add_argument('--file_name', type=str, default='bg', metavar='N',
choices=['bg', 't25', 't25r', 't50r', 'rs75'],
help='file name for scan object dataset')
parser.add_argument('--cd_weights', type=float, default=1.0, metavar='LR',
help='weights of cd_loss')
args = parser.parse_args()
BASE_DIR = args.base_dir
_init_()
io = IOStream(BASE_DIR + '/ckpt/scan/' + args.exp_name + '/run.log')
io.cprint(str(args))
name_lut = {'bg': 'training_objectdataset.h5', 't25': 'training_objectdataset_augmented25_norot.h5',
't25r': 'training_objectdataset_augmented25rot.h5', 't50r': 'training_objectdataset_augmentedrot.h5',
'rs75': 'training_objectdataset_augmentedrot_scale75.h5'}
args.file_name = name_lut[args.file_name]
args.cuda = not args.no_cuda and torch.cuda.is_available()
torch.manual_seed(args.seed)
if args.cuda:
io.cprint(
'Using GPU : ' + str(torch.cuda.current_device()) + ' from ' + str(torch.cuda.device_count()) + ' devices')
torch.cuda.manual_seed(args.seed)
else:
io.cprint('Using CPU')
if not args.eval:
train(args, io)
else:
test(args, io)