| |
| from unittest import TestCase |
|
|
| import pytest |
| import torch |
|
|
| from mmengine.structures import LabelData |
|
|
|
|
| class TestLabelData(TestCase): |
|
|
| def test_label_to_onehot(self): |
| item = torch.tensor([1], dtype=torch.int64) |
| num_classes = 10 |
| onehot = LabelData.label_to_onehot(label=item, num_classes=num_classes) |
| assert tuple(onehot.shape) == (num_classes, ) |
| assert onehot.device == item.device |
| |
| with self.assertRaises(AssertionError): |
| LabelData.label_to_onehot(label='item', num_classes=num_classes) |
|
|
| |
| with self.assertRaises(AssertionError): |
| LabelData.label_to_onehot( |
| torch.tensor([11], dtype=torch.int64), num_classes) |
| onehot = LabelData.label_to_onehot( |
| label=torch.tensor([], dtype=torch.int64), num_classes=num_classes) |
| assert (onehot == torch.zeros((num_classes, ), |
| dtype=torch.int64)).all() |
|
|
| def test_onehot_to_label(self): |
| |
| with self.assertRaisesRegex( |
| ValueError, |
| 'input is not one-hot and can not convert to label'): |
| LabelData.onehot_to_label( |
| onehot=torch.tensor([2], dtype=torch.int64)) |
|
|
| with self.assertRaises(AssertionError): |
| LabelData.onehot_to_label(onehot='item') |
|
|
| item = torch.arange(0, 9) |
| onehot = LabelData.label_to_onehot(item, num_classes=10) |
| label = LabelData.onehot_to_label(onehot) |
| assert (label == item).all() |
| assert label.device == item.device |
| item = torch.tensor([2]) |
| onehot = LabelData.label_to_onehot(item, num_classes=10) |
| label = LabelData.onehot_to_label(onehot) |
| assert label == item |
| assert label.device == item.device |
|
|
| @pytest.mark.skipif( |
| not torch.cuda.is_available(), reason='GPU is required!') |
| def test_cuda(self): |
| item = torch.arange(0, 9).cuda() |
| onehot = LabelData.label_to_onehot(item, num_classes=10) |
| assert item.device == onehot.device |
| label = LabelData.onehot_to_label(onehot) |
| assert label.device == onehot.device |
|
|
