COCO

class streaming.vision.COCO(local, remote=None, split=None, shuffle=True, transform=None, prefetch=100000, keep_zip=None, retry=2, timeout=60, hash=None, batch_size=None)

Implementation of the COCO dataset using streaming Dataset.

Parameters

• local (str) – Local filesystem directory where dataset is cached during operation.

• remote (str, optional) – Remote directory (S3 or local filesystem) where dataset is stored. Defaults to None.

• split (str, optional) – The dataset split to use, either ‘train’ or ‘val’. Defaults to None.

• shuffle (bool, optional) – Whether to iterate over the samples in randomized order. Defaults to True.

• transform (callable, optional) – A function/transform that takes in an image and bboxes and returns a transformed version. Defaults to None.

• prefetch (int, optional) – Target number of samples remaining to prefetch while iterating. Defaults to 100_000.

• keep_zip (bool, optional) – Whether to keep or delete the compressed file when decompressing downloaded shards. If set to None, keep iff remote is local. Defaults to None.

• retry (int, optional) – Number of download re-attempts before giving up. Defaults to 2.

• timeout (float, optional) – Number of seconds to wait for a shard to download before raising an exception. Defaults to 60.

• hash (str, optional) – Hash or checksum algorithm to use to validate shards. Defaults to None.

• batch_size (int, optional) – Hint the batch size that will be used on each device’s DataLoader. Defaults to None.