LensKit CLI#

LensKit comes with a command-line tool, lenskit, that provides a growing set of operations for LensKit data and pipelines.

The LensKit CLI works through subcommands (like git).

Base Options#

These options can be provided before subcommand to control output.

The following are the supported commands. Run any with --help to print its options and usage.

lenskit data convert: Convert data into LensKit native format.
lenskit data describe: Describe a Dataset.
lenskit data fetch: Fetch a public dataset.
lenskit data split: Perform a global train-test split of a dataset.
lenskit doctor: Print information about the installed Python and LensKit environment.
lenskit config: Print the LensKit configuration.
lenskit version: Display the LensKit version.
lenskit train: Train a recommendation pipeline (experimental).
lenskit recommend: Generate recommendations from a trained pipeline (experimental).
lenskit tune: Tune hyperparameters (experimental).