Shortcuts

S3FileLister

class torchdata.datapipes.iter.S3FileLister(source_datapipe: IterDataPipe[str], length: int = - 1, request_timeout_ms=- 1, region='', masks: Union[str, List[str]] = '')

Iterable DataPipe that lists Amazon S3 file URLs with the given prefixes (functional name: list_files_by_s3). Acceptable prefixes include s3://bucket-name, s3://bucket-name/, s3://bucket-name/folder.

Note

  1. source_datapipe must contain a list of valid S3 URLs

  2. length is -1 by default, and any call to __len__() is invalid, because the length is unknown until all files are iterated.

  3. request_timeout_ms and region will overwrite settings in the configuration file or environment variables.

  4. The lack of AWS proper configuration can lead empty response. For more details related to S3 IO DataPipe setup and AWS config, please see the README file.

Parameters:
  • source_datapipe – a DataPipe that contains URLs/URL prefixes to s3 files

  • length – Nominal length of the datapipe

  • request_timeout_ms – timeout setting for each reqeust (3,000ms by default)

  • region – region for access files (inferred from credentials by default)

Example:

from torchdata.datapipes.iter import IterableWrapper, S3FileLister

s3_prefixes = IterableWrapper(['s3://bucket-name/folder/', ...])

dp_s3_urls = S3FileLister(s3_prefixes)
for d in dp_s3_urls:
    pass

# Functional API
dp_s3_urls = s3_prefixes.list_files_by_s3(request_timeout_ms=100)
for d in dp_s3_urls:
    pass

Docs

Access comprehensive developer documentation for PyTorch

View Docs

Tutorials

Get in-depth tutorials for beginners and advanced developers

View Tutorials

Resources

Find development resources and get your questions answered

View Resources