fastdev.datasets.s3dis

Module Contents

fastdev.datasets.s3dis.logger[source]
fastdev.datasets.s3dis.collect_point_label(anno_path, out_filename, class_names_path, file_format='txt')[source]
Convert original dataset files to data_label file (each line is XYZRGBL).

We aggregated all the points from each instance in the room.

Parameters:

  • anno_path – path to annotations. e.g. Area_1/office_2/Annotations/

  • out_filename – path to save collected points and labels (each line is XYZRGBL)

  • file_format – txt or numpy, determines what file format to save.

Returns:

None

Note

the points are shifted before save, the most negative point is now at origin.

fastdev.datasets.s3dis.sample_points(points: numpy.ndarray, labels: numpy.ndarray, room_coord_max: numpy.ndarray, num_points: int = 4096, block_size: float = 1.0)[source]
Parameters:

  • points (numpy.ndarray)

  • labels (numpy.ndarray)

  • room_coord_max (numpy.ndarray)

  • num_points (int)

  • block_size (float)

class fastdev.datasets.s3dis.S3DISDatasetConfig[source]

Configuration for S3DISDataset.

data_root: str[source]
download_if_not_exist: bool = False[source]
num_points: int = 4096[source]
test_area: int = 5[source]
block_size: float = 1.0[source]
sample_rate: float = 1.0[source]
class fastdev.datasets.s3dis.S3DISDataset(config: S3DISDatasetConfig, split: Literal['train', 'test'] = 'train')[source]

Bases: torch.utils.data.Dataset

S3DIS dataset.

Parameters:
config[source]
num_points[source]
block_size[source]
labelweights[source]
room_idxs[source]
__getitem__(idx)[source]
__len__()[source]
static download_data(data_root: str)[source]
Parameters:

data_root (str)