File size: 9,519 Bytes
3bbb319 |
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 |
import pytest
import torch
from mmdet.models.dense_heads import (DecoupledSOLOHead,
DecoupledSOLOLightHead, SOLOHead)
def test_solo_head_loss():
"""Tests solo head loss when truth is empty and non-empty."""
s = 256
img_metas = [{
'img_shape': (s, s, 3),
'scale_factor': 1,
'pad_shape': (s, s, 3)
}]
self = SOLOHead(
num_classes=4,
in_channels=1,
num_grids=[40, 36, 24, 16, 12],
loss_mask=dict(type='DiceLoss', use_sigmoid=True, loss_weight=3.0),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0))
feat = [
torch.rand(1, 1, s // feat_size, s // feat_size)
for feat_size in [4, 8, 16, 32, 64]
]
mask_preds, cls_preds = self.forward(feat)
# Test that empty ground truth encourages the network to
# predict background.
gt_bboxes = [torch.empty((0, 4))]
gt_labels = [torch.LongTensor([])]
gt_masks = [torch.empty((0, 550, 550))]
gt_bboxes_ignore = None
empty_gt_losses = self.loss(
mask_preds,
cls_preds,
gt_labels,
gt_masks,
img_metas,
gt_bboxes,
gt_bboxes_ignore=gt_bboxes_ignore)
# When there is no truth, the cls loss should be nonzero but there should
# be no box loss.
empty_mask_loss = empty_gt_losses['loss_mask']
empty_cls_loss = empty_gt_losses['loss_cls']
assert empty_cls_loss.item() > 0, 'cls loss should be non-zero'
assert empty_mask_loss.item() == 0, (
'there should be no mask loss when there are no true masks')
# When truth is non-empty then both cls and box loss should be nonzero for
# random inputs.
gt_bboxes = [
torch.Tensor([[23.6667, 23.8757, 238.6326, 151.8874]]),
]
gt_labels = [torch.LongTensor([2])]
gt_masks = [(torch.rand((1, 256, 256)) > 0.5).float()]
one_gt_losses = self.loss(
mask_preds,
cls_preds,
gt_labels,
gt_masks,
img_metas,
gt_bboxes,
gt_bboxes_ignore=gt_bboxes_ignore)
onegt_mask_loss = one_gt_losses['loss_mask']
onegt_cls_loss = one_gt_losses['loss_cls']
assert onegt_cls_loss.item() > 0, 'cls loss should be non-zero'
assert onegt_mask_loss.item() > 0, 'mask loss should be non-zero'
# When the length of num_grids, scale_ranges, and num_levels are not equal.
with pytest.raises(AssertionError):
SOLOHead(
num_classes=4,
in_channels=1,
num_grids=[36, 24, 16, 12],
loss_mask=dict(type='DiceLoss', use_sigmoid=True, loss_weight=3.0),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0))
# When input feature length is not equal to num_levels.
with pytest.raises(AssertionError):
feat = [
torch.rand(1, 1, s // feat_size, s // feat_size)
for feat_size in [4, 8, 16, 32]
]
self.forward(feat)
def test_desolo_head_loss():
"""Tests solo head loss when truth is empty and non-empty."""
s = 256
img_metas = [{
'img_shape': (s, s, 3),
'scale_factor': 1,
'pad_shape': (s, s, 3)
}]
self = DecoupledSOLOHead(
num_classes=4,
in_channels=1,
num_grids=[40, 36, 24, 16, 12],
loss_mask=dict(
type='DiceLoss', use_sigmoid=True, activate=False,
loss_weight=3.0),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0))
feat = [
torch.rand(1, 1, s // feat_size, s // feat_size)
for feat_size in [4, 8, 16, 32, 64]
]
mask_preds_x, mask_preds_y, cls_preds = self.forward(feat)
# Test that empty ground truth encourages the network to
# predict background.
gt_bboxes = [torch.empty((0, 4))]
gt_labels = [torch.LongTensor([])]
gt_masks = [torch.empty((0, 550, 550))]
gt_bboxes_ignore = None
empty_gt_losses = self.loss(
mask_preds_x,
mask_preds_y,
cls_preds,
gt_labels,
gt_masks,
img_metas,
gt_bboxes,
gt_bboxes_ignore=gt_bboxes_ignore)
# When there is no truth, the cls loss should be nonzero but there should
# be no box loss.
empty_mask_loss = empty_gt_losses['loss_mask']
empty_cls_loss = empty_gt_losses['loss_cls']
assert empty_cls_loss.item() > 0, 'cls loss should be non-zero'
assert empty_mask_loss.item() == 0, (
'there should be no mask loss when there are no true masks')
# When truth is non-empty then both cls and box loss should be nonzero for
# random inputs.
gt_bboxes = [
torch.Tensor([[23.6667, 23.8757, 238.6326, 151.8874]]),
]
gt_labels = [torch.LongTensor([2])]
gt_masks = [(torch.rand((1, 256, 256)) > 0.5).float()]
one_gt_losses = self.loss(
mask_preds_x,
mask_preds_y,
cls_preds,
gt_labels,
gt_masks,
img_metas,
gt_bboxes,
gt_bboxes_ignore=gt_bboxes_ignore)
onegt_mask_loss = one_gt_losses['loss_mask']
onegt_cls_loss = one_gt_losses['loss_cls']
assert onegt_cls_loss.item() > 0, 'cls loss should be non-zero'
assert onegt_mask_loss.item() > 0, 'mask loss should be non-zero'
# When the length of num_grids, scale_ranges, and num_levels are not equal.
with pytest.raises(AssertionError):
DecoupledSOLOHead(
num_classes=4,
in_channels=1,
num_grids=[36, 24, 16, 12],
loss_mask=dict(
type='DiceLoss',
use_sigmoid=True,
activate=False,
loss_weight=3.0),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0))
# When input feature length is not equal to num_levels.
with pytest.raises(AssertionError):
feat = [
torch.rand(1, 1, s // feat_size, s // feat_size)
for feat_size in [4, 8, 16, 32]
]
self.forward(feat)
def test_desolo_light_head_loss():
"""Tests solo head loss when truth is empty and non-empty."""
s = 256
img_metas = [{
'img_shape': (s, s, 3),
'scale_factor': 1,
'pad_shape': (s, s, 3)
}]
self = DecoupledSOLOLightHead(
num_classes=4,
in_channels=1,
num_grids=[40, 36, 24, 16, 12],
loss_mask=dict(
type='DiceLoss', use_sigmoid=True, activate=False,
loss_weight=3.0),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0))
feat = [
torch.rand(1, 1, s // feat_size, s // feat_size)
for feat_size in [4, 8, 16, 32, 64]
]
mask_preds_x, mask_preds_y, cls_preds = self.forward(feat)
# Test that empty ground truth encourages the network to
# predict background.
gt_bboxes = [torch.empty((0, 4))]
gt_labels = [torch.LongTensor([])]
gt_masks = [torch.empty((0, 550, 550))]
gt_bboxes_ignore = None
empty_gt_losses = self.loss(
mask_preds_x,
mask_preds_y,
cls_preds,
gt_labels,
gt_masks,
img_metas,
gt_bboxes,
gt_bboxes_ignore=gt_bboxes_ignore)
# When there is no truth, the cls loss should be nonzero but there should
# be no box loss.
empty_mask_loss = empty_gt_losses['loss_mask']
empty_cls_loss = empty_gt_losses['loss_cls']
assert empty_cls_loss.item() > 0, 'cls loss should be non-zero'
assert empty_mask_loss.item() == 0, (
'there should be no mask loss when there are no true masks')
# When truth is non-empty then both cls and box loss should be nonzero for
# random inputs.
gt_bboxes = [
torch.Tensor([[23.6667, 23.8757, 238.6326, 151.8874]]),
]
gt_labels = [torch.LongTensor([2])]
gt_masks = [(torch.rand((1, 256, 256)) > 0.5).float()]
one_gt_losses = self.loss(
mask_preds_x,
mask_preds_y,
cls_preds,
gt_labels,
gt_masks,
img_metas,
gt_bboxes,
gt_bboxes_ignore=gt_bboxes_ignore)
onegt_mask_loss = one_gt_losses['loss_mask']
onegt_cls_loss = one_gt_losses['loss_cls']
assert onegt_cls_loss.item() > 0, 'cls loss should be non-zero'
assert onegt_mask_loss.item() > 0, 'mask loss should be non-zero'
# When the length of num_grids, scale_ranges, and num_levels are not equal.
with pytest.raises(AssertionError):
DecoupledSOLOLightHead(
num_classes=4,
in_channels=1,
num_grids=[36, 24, 16, 12],
loss_mask=dict(type='DiceLoss', use_sigmoid=True, loss_weight=3.0),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0))
# When input feature length is not equal to num_levels.
with pytest.raises(AssertionError):
feat = [
torch.rand(1, 1, s // feat_size, s // feat_size)
for feat_size in [4, 8, 16, 32]
]
self.forward(feat)
|