Initializes the dataset with the filename lists. The transform `img_transform` is applied to the images and `seg_transform` to the segmentations. Args: img: sequence of images. img_transform: transform to apply to each element in `img`. s
(
self,
img: Sequence,
img_transform: Callable | None = None,
seg: Sequence | None = None,
seg_transform: Callable | None = None,
labels: Sequence | None = None,
label_transform: Callable | None = None,
)
| 1370 | """ |
| 1371 | |
| 1372 | def __init__( |
| 1373 | self, |
| 1374 | img: Sequence, |
| 1375 | img_transform: Callable | None = None, |
| 1376 | seg: Sequence | None = None, |
| 1377 | seg_transform: Callable | None = None, |
| 1378 | labels: Sequence | None = None, |
| 1379 | label_transform: Callable | None = None, |
| 1380 | ) -> None: |
| 1381 | """ |
| 1382 | Initializes the dataset with the filename lists. The transform `img_transform` is applied |
| 1383 | to the images and `seg_transform` to the segmentations. |
| 1384 | |
| 1385 | Args: |
| 1386 | img: sequence of images. |
| 1387 | img_transform: transform to apply to each element in `img`. |
| 1388 | seg: sequence of segmentations. |
| 1389 | seg_transform: transform to apply to each element in `seg`. |
| 1390 | labels: sequence of labels. |
| 1391 | label_transform: transform to apply to each element in `labels`. |
| 1392 | |
| 1393 | """ |
| 1394 | items = [(img, img_transform), (seg, seg_transform), (labels, label_transform)] |
| 1395 | self.set_random_state(seed=get_seed()) |
| 1396 | datasets = [Dataset(x[0], x[1]) for x in items if x[0] is not None] |
| 1397 | self.dataset = datasets[0] if len(datasets) == 1 else ZipDataset(datasets) |
| 1398 | |
| 1399 | self._seed = 0 # transform synchronization seed |
| 1400 | |
| 1401 | def __len__(self) -> int: |
| 1402 | return len(self.dataset) |
nothing calls this directly
no test coverage detected