yahoo_ltrc

• Description:

The Yahoo Learning to Rank Challenge dataset (also called "C14") is a Learning-to-Rank dataset released by Yahoo. The dataset consists of query-document pairs represented as feature vectors and corresponding relevance judgment labels.

The dataset contains two versions:

• set1: Containing 709,877 query-document pairs.
• set2: Containing 172,870 query-document pairs.

You can specify whether to use the set1 or set2 version of the dataset as follows:

ds = tfds.load("yahoo_ltrc/set1")
ds = tfds.load("yahoo_ltrc/set2")

If only yahoo_ltrc is specified, the yahoo_ltrc/set1 option is selected by default:

# This is the same as `tfds.load("yahoo_ltrc/set1")`
ds = tfds.load("yahoo_ltrc")

• Homepage: https://research.yahoo.com/datasets

• Source code: tfds.ranking.yahoo_ltrc.YahooLTRC

• Versions:

  • 1.0.0: Initial release.
  • 1.1.0 (default): Add query and document identifiers.

• Download size: Unknown size

• Manual download instructions: This dataset requires you to download the source data manually into download_config.manual_dir (defaults to ~/tensorflow_datasets/downloads/manual/):
  Request access for the C14 Yahoo Learning To Rank Challenge dataset on https://research.yahoo.com/datasets Extract the downloaded dataset.tgz file and place the ltrc_yahoo.tar.bz2 file in manual_dir/.

• Supervised keys (See as_supervised doc): None

• Figure (tfds.show_examples): Not supported.

• Citation:

@inproceedings{chapelle2011yahoo,
  title={Yahoo! learning to rank challenge overview},
  author={Chapelle, Olivier and Chang, Yi},
  booktitle={Proceedings of the learning to rank challenge},
  pages={1--24},
  year={2011},
  organization={PMLR}
}

yahoo_ltrc/set1 (default config)

• Dataset size: 795.39 MiB

• Auto-cached (documentation): No

• Splits:

• Feature structure:

FeaturesDict({
    'doc_id': Tensor(shape=(None,), dtype=int64),
    'float_features': Tensor(shape=(None, 699), dtype=float64),
    'label': Tensor(shape=(None,), dtype=float64),
    'query_id': Text(shape=(), dtype=string),
})

• Feature documentation:

• Examples (tfds.as_dataframe):

yahoo_ltrc/set2

• Dataset size: 194.92 MiB

• Auto-cached (documentation): Yes

• Splits:

• Feature structure:

FeaturesDict({
    'doc_id': Tensor(shape=(None,), dtype=int64),
    'float_features': Tensor(shape=(None, 700), dtype=float64),
    'label': Tensor(shape=(None,), dtype=float64),
    'query_id': Text(shape=(), dtype=string),
})

• Feature documentation:

• Examples (tfds.as_dataframe):