Data Abstractions

The lenskit.data module provides the core data abstractions LensKit uses to represent recommender system inputs and outputs.

• Basic Data Types

Data Sets

Dataset	Representation of a data set for LensKit training, evaluation, etc.
EntitySet	Representation of a set of entities from the dataset.
AttributeSet	Base class for attributes associated with entities.
RelationshipSet	Representation for a set of relationship records.
MatrixRelationshipSet	Two-entity relationships without duplicates, accessible in matrix form.
CSRStructure	Representation of the compressed sparse row structure of a sparse matrix, without any data values.

Building Data Sets

DatasetBuilder	Construct data sets from data and tables.
from_interactions_df	Create a dataset from a data frame of ratings or other user-item interactions.
load_movielens	Load a MovieLens dataset.
load_movielens_df	Load the ratings from a MovieLens dataset as a raw data frame.

Item Data

ItemList	Representation of a (usually ordered) list of items, possibly with scores and other associated data; many components take and return item lists.
ItemListCollection	A collection of item lists.
ItemListCollector	Collect item lists with associated keys, as in ItemListCollection.
ListILC	Mutable item list collection backed by a Python list.
UserIDKey	Key type for user IDs.
GenericKey	Built-in immutable sequence.
MutableItemListCollection	Intersection type of ItemListCollection and ItemListCollector.

Recommendation Queries

RecQuery	Representation of a the data available for a recommendation query.
QueryInput	Represent a PEP 604 union type

Schemas and Identifiers

lenskit.data.schema	Pydantic models for LensKit data schemas.
Vocabulary	Vocabularies of entity identifiers for the LensKit data model.

See also:

• lenskit.data.types.EntityId