test_heads.py 9.55 KB
Newer Older
wuyuefeng's avatar
wuyuefeng committed
1
2
3
4
5
6
import copy
from os.path import dirname, exists, join

import pytest
import torch

7
8
from mmdet3d.core.bbox import Box3DMode, LiDARInstance3DBoxes

wuyuefeng's avatar
wuyuefeng committed
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

def _get_config_directory():
    """ Find the predefined detector config directory """
    try:
        # Assume we are running in the source mmdetection repo
        repo_dpath = dirname(dirname(__file__))
    except NameError:
        # For IPython development when this __file__ is not defined
        import mmdet
        repo_dpath = dirname(dirname(mmdet.__file__))
    config_dpath = join(repo_dpath, 'configs')
    if not exists(config_dpath):
        raise Exception('Cannot find config path')
    return config_dpath


def _get_config_module(fname):
    """
    Load a configuration as a python module
    """
    from mmcv import Config
    config_dpath = _get_config_directory()
    config_fpath = join(config_dpath, fname)
    config_mod = Config.fromfile(config_fpath)
    return config_mod


def _get_head_cfg(fname):
    """
    Grab configs necessary to create a bbox_head. These are deep copied to
    allow for safe modification of parameters without influencing other tests.
    """
    import mmcv
    config = _get_config_module(fname)
    model = copy.deepcopy(config.model)
    train_cfg = mmcv.Config(copy.deepcopy(config.train_cfg))
    test_cfg = mmcv.Config(copy.deepcopy(config.test_cfg))

    bbox_head = model.bbox_head
    bbox_head.update(train_cfg=train_cfg)
    bbox_head.update(test_cfg=test_cfg)
    return bbox_head


def _get_rpn_head_cfg(fname):
    """
    Grab configs necessary to create a rpn_head. These are deep copied to allow
    for safe modification of parameters without influencing other tests.
    """
    import mmcv
    config = _get_config_module(fname)
    model = copy.deepcopy(config.model)
    train_cfg = mmcv.Config(copy.deepcopy(config.train_cfg))
    test_cfg = mmcv.Config(copy.deepcopy(config.test_cfg))

    rpn_head = model.rpn_head
    rpn_head.update(train_cfg=train_cfg.rpn)
    rpn_head.update(test_cfg=test_cfg.rpn)
    return rpn_head, train_cfg.rpn_proposal


zhangwenwei's avatar
zhangwenwei committed
70
def test_anchor3d_head_loss():
wuyuefeng's avatar
wuyuefeng committed
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
    if not torch.cuda.is_available():
        pytest.skip('test requires GPU and torch+cuda')
    bbox_head_cfg = _get_head_cfg(
        'kitti/dv_second_secfpn_2x8_cosine_80e_kitti-3d-3class.py')

    from mmdet3d.models.builder import build_head
    self = build_head(bbox_head_cfg)
    self.cuda()
    assert isinstance(self.conv_cls, torch.nn.modules.conv.Conv2d)
    assert self.conv_cls.in_channels == 512
    assert self.conv_cls.out_channels == 18
    assert self.conv_reg.out_channels == 42
    assert self.conv_dir_cls.out_channels == 12

    # test forward
    feats = list()
    feats.append(torch.rand([2, 512, 200, 176], dtype=torch.float32).cuda())
    (cls_score, bbox_pred, dir_cls_preds) = self.forward(feats)
    assert cls_score[0].shape == torch.Size([2, 18, 200, 176])
    assert bbox_pred[0].shape == torch.Size([2, 42, 200, 176])
    assert dir_cls_preds[0].shape == torch.Size([2, 12, 200, 176])

    # test loss
    gt_bboxes = list(
        torch.tensor(
            [[[6.4118, -3.4305, -1.7291, 1.7033, 3.4693, 1.6197, -0.9091]],
             [[16.9107, 9.7925, -1.9201, 1.6097, 3.2786, 1.5307, -2.4056]]],
            dtype=torch.float32).cuda())
    gt_labels = list(torch.tensor([[0], [1]], dtype=torch.int64).cuda())
    input_metas = [{
        'sample_idx': 1234
    }, {
        'sample_idx': 2345
    }]  # fake input_metas

    losses = self.loss(cls_score, bbox_pred, dir_cls_preds, gt_bboxes,
                       gt_labels, input_metas)
wuyuefeng's avatar
wuyuefeng committed
108
109
110
    assert losses['loss_rpn_cls'][0] > 0
    assert losses['loss_rpn_bbox'][0] > 0
    assert losses['loss_rpn_dir'][0] > 0
wuyuefeng's avatar
wuyuefeng committed
111
112
113
114
115
116

    # test empty ground truth case
    gt_bboxes = list(torch.empty((2, 0, 7)).cuda())
    gt_labels = list(torch.empty((2, 0)).cuda())
    empty_gt_losses = self.loss(cls_score, bbox_pred, dir_cls_preds, gt_bboxes,
                                gt_labels, input_metas)
wuyuefeng's avatar
wuyuefeng committed
117
118
119
    assert empty_gt_losses['loss_rpn_cls'][0] > 0
    assert empty_gt_losses['loss_rpn_bbox'][0] == 0
    assert empty_gt_losses['loss_rpn_dir'][0] == 0
wuyuefeng's avatar
wuyuefeng committed
120
121


zhangwenwei's avatar
zhangwenwei committed
122
def test_anchor3d_head_getboxes():
wuyuefeng's avatar
wuyuefeng committed
123
124
125
126
127
128
129
130
131
132
133
    if not torch.cuda.is_available():
        pytest.skip('test requires GPU and torch+cuda')
    bbox_head_cfg = _get_head_cfg(
        'kitti/dv_second_secfpn_2x8_cosine_80e_kitti-3d-3class.py')

    from mmdet3d.models.builder import build_head
    self = build_head(bbox_head_cfg)
    self.cuda()

    feats = list()
    feats.append(torch.rand([2, 512, 200, 176], dtype=torch.float32).cuda())
134
    # fake input_metas
wuyuefeng's avatar
wuyuefeng committed
135
    input_metas = [{
136
137
138
        'sample_idx': 1234,
        'box_type_3d': LiDARInstance3DBoxes,
        'box_mode_3d': Box3DMode.LIDAR
wuyuefeng's avatar
wuyuefeng committed
139
    }, {
140
141
142
143
        'sample_idx': 2345,
        'box_type_3d': LiDARInstance3DBoxes,
        'box_mode_3d': Box3DMode.LIDAR
    }]
wuyuefeng's avatar
wuyuefeng committed
144
145
146
147
148
149
    (cls_score, bbox_pred, dir_cls_preds) = self.forward(feats)

    # test get_boxes
    cls_score[0] -= 1.5  # too many positive samples may cause cuda oom
    result_list = self.get_bboxes(cls_score, bbox_pred, dir_cls_preds,
                                  input_metas)
zhangwenwei's avatar
zhangwenwei committed
150
    assert (result_list[0][1] > 0.3).all()
wuyuefeng's avatar
wuyuefeng committed
151
152
153
154
155
156


def test_parta2_rpnhead_getboxes():
    if not torch.cuda.is_available():
        pytest.skip('test requires GPU and torch+cuda')
    rpn_head_cfg, proposal_cfg = _get_rpn_head_cfg(
wuyuefeng's avatar
wuyuefeng committed
157
        'kitti/hv_PartA2_secfpn_4x8_cyclic_80e_kitti-3d-3class.py')
wuyuefeng's avatar
wuyuefeng committed
158
159
160
161
162
163
164

    from mmdet3d.models.builder import build_head
    self = build_head(rpn_head_cfg)
    self.cuda()

    feats = list()
    feats.append(torch.rand([2, 512, 200, 176], dtype=torch.float32).cuda())
165
    # fake input_metas
wuyuefeng's avatar
wuyuefeng committed
166
    input_metas = [{
167
168
169
        'sample_idx': 1234,
        'box_type_3d': LiDARInstance3DBoxes,
        'box_mode_3d': Box3DMode.LIDAR
wuyuefeng's avatar
wuyuefeng committed
170
    }, {
171
172
173
174
        'sample_idx': 2345,
        'box_type_3d': LiDARInstance3DBoxes,
        'box_mode_3d': Box3DMode.LIDAR
    }]
wuyuefeng's avatar
wuyuefeng committed
175
176
177
178
179
180
    (cls_score, bbox_pred, dir_cls_preds) = self.forward(feats)

    # test get_boxes
    cls_score[0] -= 1.5  # too many positive samples may cause cuda oom
    result_list = self.get_bboxes(cls_score, bbox_pred, dir_cls_preds,
                                  input_metas, proposal_cfg)
zhangwenwei's avatar
zhangwenwei committed
181
182
    assert result_list[0]['scores_3d'].shape == torch.Size([512])
    assert result_list[0]['labels_3d'].shape == torch.Size([512])
wuyuefeng's avatar
wuyuefeng committed
183
    assert result_list[0]['cls_preds'].shape == torch.Size([512, 3])
184
    assert result_list[0]['boxes_3d'].tensor.shape == torch.Size([512, 7])
wuyuefeng's avatar
Votenet  
wuyuefeng committed
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


def test_vote_head():
    if not torch.cuda.is_available():
        pytest.skip('test requires GPU and torch+cuda')
    from mmdet3d.models.dense_heads import VoteHead
    bbox_head_cfg = dict(
        num_classes=10,
        bbox_coder=dict(
            type='PartialBinBasedBBoxCoder',
            num_sizes=10,
            num_dir_bins=5,
            with_rot=True,
            mean_sizes=[[2.114256, 1.620300, 0.927272],
                        [0.791118, 1.279516, 0.718182],
                        [0.923508, 1.867419, 0.845495],
                        [0.591958, 0.552978, 0.827272],
                        [0.699104, 0.454178, 0.75625],
                        [0.69519, 1.346299, 0.736364],
                        [0.528526, 1.002642, 1.172878],
                        [0.500618, 0.632163, 0.683424],
                        [0.404671, 1.071108, 1.688889],
                        [0.76584, 1.398258, 0.472728]]),
        vote_moudule_cfg=dict(
            in_channels=64,
            vote_per_seed=1,
            gt_per_seed=3,
            conv_channels=(64, 64),
            conv_cfg=dict(type='Conv1d'),
            norm_cfg=dict(type='BN1d'),
            norm_feats=True,
            vote_loss=dict(
                type='ChamferDistance',
                mode='l1',
                reduction='none',
                loss_dst_weight=10.0)),
        vote_aggregation_cfg=dict(
            num_point=256,
            radius=0.3,
            num_sample=16,
            mlp_channels=[64, 32, 32, 32],
            use_xyz=True,
            normalize_xyz=True),
        feat_channels=(64, 64),
        conv_cfg=dict(type='Conv1d'),
        norm_cfg=dict(type='BN1d'),
        objectness_loss=dict(
            type='CrossEntropyLoss',
            class_weight=[0.2, 0.8],
            reduction='sum',
            loss_weight=5.0),
        center_loss=dict(
            type='ChamferDistance',
            mode='l2',
            reduction='sum',
            loss_src_weight=10.0,
            loss_dst_weight=10.0),
        dir_class_loss=dict(
            type='CrossEntropyLoss', reduction='sum', loss_weight=1.0),
        dir_res_loss=dict(
            type='SmoothL1Loss', reduction='sum', loss_weight=10.0),
        size_class_loss=dict(
            type='CrossEntropyLoss', reduction='sum', loss_weight=1.0),
        size_res_loss=dict(
            type='SmoothL1Loss', reduction='sum', loss_weight=10.0 / 3.0),
        semantic_loss=dict(
            type='CrossEntropyLoss', reduction='sum', loss_weight=1.0))

    train_cfg = dict(
        pos_distance_thr=0.3, neg_distance_thr=0.6, sample_mod='vote')

    self = VoteHead(train_cfg=train_cfg, **bbox_head_cfg).cuda()
    fp_xyz = [torch.rand([2, 64, 3], dtype=torch.float32).cuda()]
    fp_features = [torch.rand([2, 64, 64], dtype=torch.float32).cuda()]
    fp_indices = [torch.randint(0, 128, [2, 64]).cuda()]

    input_dict = dict(
        fp_xyz=fp_xyz, fp_features=fp_features, fp_indices=fp_indices)
    # test forward
    ret_dict = self(input_dict, 'vote')
    assert ret_dict['center'].shape == torch.Size([2, 256, 3])
    assert ret_dict['obj_scores'].shape == torch.Size([2, 256, 2])
    assert ret_dict['size_res'].shape == torch.Size([2, 256, 10, 3])
    assert ret_dict['dir_res'].shape == torch.Size([2, 256, 5])