vis4d.zoo.base.datasets.bdd100k.detect

BDD100K dataset config for object detection.

Functions

get_bdd100k_detection_config([data_root, ...])

Get the default config for BDD100K detection.

get_test_dataloader(data_root, anno_path[, ...])

Get the default test dataloader for BDD100K segmentation.

get_train_dataloader(data_root, anno_path[, ...])

Get the default train dataloader for BDD100K segmentation.

get_bdd100k_detection_config(data_root='data/bdd100k/images/100k', train_split='train', train_keys_to_load=('images', 'boxes2d'), test_split='val', test_keys_to_load=('images', 'original_images'), ins_seg=False, data_backend=None, image_size=(720, 1280), multi_scale=False, samples_per_gpu=2, workers_per_gpu=2)[source]

Get the default config for BDD100K detection.

Return type:

DataConfig

get_test_dataloader(data_root, anno_path, keys_to_load=('images', 'original_images'), ins_seg=False, data_backend=None, image_size=(720, 1280), samples_per_gpu=1, workers_per_gpu=1)[source]

Get the default test dataloader for BDD100K segmentation.

Return type:

ConfigDict

get_train_dataloader(data_root, anno_path, keys_to_load=('images', 'boxes2d'), ins_seg=False, data_backend=None, image_size=(720, 1280), multi_scale=False, samples_per_gpu=2, workers_per_gpu=2)[source]

Get the default train dataloader for BDD100K segmentation.

Return type:

ConfigDict