lenskit.sklearn.svd#

Singular value decomposition for explicit feedback.

This module contains a truncated SVD explicit-feedback scorer built on sklearn.decomposition.TruncatedSVD.

Classes#

BiasedSVDConfig

Mixin for configuring embedding sizes (# of latent dimensions).

BiasedSVDScorer

Biased matrix factorization for explicit feedback using SciKit-Learn's

Module Contents#

class lenskit.sklearn.svd.BiasedSVDConfig#

Bases: lenskit.config.common.EmbeddingSizeMixin, pydantic.BaseModel

Mixin for configuring embedding sizes (# of latent dimensions).

Component configuration classes can extend this class to inherit a standardized definition of an embedding size, along with useful behavior like configuring with base-2 logs.

Example usage:

class SVDConfig(EmbeddingSizeMixin, BaseModel):
    pass

cfg = SVDConfig(embedding_size=32)
embedding_size: int#

The dimension of user and item embeddings (number of latent features to learn).

damping: lenskit.basic.Damping = 5#
algorithm: Literal['arpack', 'randomized'] = 'randomized'#
n_iter: int = 5#
class lenskit.sklearn.svd.BiasedSVDScorer(config=None, **kwargs)#

Bases: lenskit.pipeline.Component[lenskit.data.ItemList], lenskit.training.Trainable

Biased matrix factorization for explicit feedback using SciKit-Learn’s TruncatedSVD. It operates by first computing the bias, then computing the SVD of the bias residuals.

You’ll generally want one of the iterative SVD implementations such as lenskit.als.BiasedMFScorer; this is here primarily as an example and for cases where you want to evaluate a pure SVD implementation.

Stability:
Caller (see Stability Levels).
Parameters:

  • config (object | None)

  • kwargs (Any)

config: BiasedSVDConfig#

The component configuration object. Component classes that support configuration must redefine this attribute with their specific configuration class type, which can be a Python dataclass or a Pydantic model class.

bias: lenskit.basic.BiasModel#
factorization: sklearn.decomposition.TruncatedSVD#
users: lenskit.data.Vocabulary#
items: lenskit.data.Vocabulary#
user_components: numpy.typing.NDArray[numpy.float64]#
is_trained()#

Query if this component has already been trained.

train(data, options=TrainingOptions())#

Train the model to learn its parameters from a training dataset.

Parameters:
__call__(query, items)#

Run the pipeline’s operation and produce a result. This is the key method for components to implement.

Parameters:
Return type:

lenskit.data.ItemList

Exported Aliases#

class lenskit.sklearn.svd.BiasModel#

Re-exported alias for lenskit.basic.BiasModel.

lenskit.sklearn.svd.Damping#

Re-exported alias for lenskit.basic.Damping.

class lenskit.sklearn.svd.EmbeddingSizeMixin#

Re-exported alias for lenskit.config.common.EmbeddingSizeMixin.

class lenskit.sklearn.svd.Dataset#

Re-exported alias for lenskit.data.Dataset.

class lenskit.sklearn.svd.ItemList#

Re-exported alias for lenskit.data.ItemList.

lenskit.sklearn.svd.QueryInput#

Re-exported alias for lenskit.data.QueryInput.

class lenskit.sklearn.svd.RecQuery#

Re-exported alias for lenskit.data.RecQuery.

class lenskit.sklearn.svd.Vocabulary#

Re-exported alias for lenskit.data.Vocabulary.

class lenskit.sklearn.svd.Stopwatch#

Re-exported alias for lenskit.logging.Stopwatch.

lenskit.sklearn.svd.get_logger()#

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

class lenskit.sklearn.svd.Component#

Re-exported alias for lenskit.pipeline.Component.

class lenskit.sklearn.svd.Trainable#

Re-exported alias for lenskit.training.Trainable.

class lenskit.sklearn.svd.TrainingOptions#

Re-exported alias for lenskit.training.TrainingOptions.