lenskit.data.sources.ambar#

Support for the AMBAR music recommendation dataset.

Functions#

load_ambar(path)

Load the AMBAR music dataset.

Module Contents#

lenskit.data.sources.ambar.load_ambar(path)#

Load the AMBAR music dataset.

AMBAR consists of roughly 3.3M ratings from around 31,013 users for 443,921 tracks and 30,667 artists. Ratings are on a 1-5 Likert scale, derived from quintiles of each user’s listening-history playcounts.

The dataset directory is expected to contain four files: users_info.csv, tracks_info.csv, artists_info.csv, and ratings_info.csv.

Parameters:

path (pathlib.Path | str | os.PathLike[str]) – The directory containing the AMBAR data files.

Returns:

The loaded AMBAR dataset.

Return type:

lenskit.data._dataset.Dataset

Exported Aliases#

lenskit.data.sources.ambar.get_logger()#

Re-exported alias for lenskit.logging.get_logger().

class lenskit.data.sources.ambar.DatasetBuilder#

Re-exported alias for lenskit.data._builder.DatasetBuilder.

class lenskit.data.sources.ambar.Dataset#

Re-exported alias for lenskit.data._dataset.Dataset.