Load benchmark datasetΒΆ

Currently the following datasets based on TREC 2009 legal collection are supported:

  • treclegal09_2k_subset : 2 400 documents, 2 MB
  • treclegal09_20k_subset : 20 000 documents, 30 MB
  • treclegal09_37k_subset : 37 000 documents, 55 MB
  • treclegal09 : 700 000 documents, 1.2 GB The ground truth files for categorization are adapted from TAR Toolkit.

If you encounter any issues for downloads with this function, you can also manually download and extract the required dataset to cache_dir (the download url is http://r0h.eu/d/<name>.tar.gz), then re-run this function to get the required metadata.

  • URL: /api/v0/dataset/<dataset-name>

  • Method: GET URL Params: None

  • Data Params: None

  • Success Response: HTTP 200

     {"data_dir": <str>, "base_dir": <str>,
  "seed_non_relevant_files": <list[str]>, "seed_relevant_files": <list[str]>,
  "ground_truth_file": <str>}