Inheritance diagram of LLEMapper

class mvpa2.mappers.lle.LLEMapper(k, alg='LLE', nodeargs=None, **kwargs)

Locally linear embbeding Mapper.

This mapper performs dimensionality reduction. It wraps two algorithms provided by the Modular Data Processing (MDP) framework.

Locally linear embedding (LLE) approximates the input data with a low-dimensional surface and reduces its dimensionality by learning a mapping to the surface.

This wrapper class provides access to two different LLE algorithms (i.e. the corresponding MDP processing nodes). 1) An algorithm outlined in An Introduction to Locally Linear Embedding by L. Saul and S. Roweis, using improvements suggested in Locally Linear Embedding for Classification by D. deRidder and Duin (aka LLENode) and 2) Hessian Locally Linear Embedding analysis based on algorithm outlined in Hessian Eigenmaps: new locally linear embedding techniques for high-dimensional data by C. Grimes and D. Donoho, 2003.

For more information see the MDP website at


This mapper only provides forward-mapping functionality – no reverse mapping is available.

Available conditional attributes:

  • calling_time+: None
  • raw_results: None
  • trained_dataset: None
  • trained_nsamples+: None
  • trained_targets+: None
  • training_time+: None

(Conditional attributes enabled by default suffixed with +)



k : int

Number of nearest neighbors to be used by the algorithm.

algorithm : {‘LLE’, ‘HLLE’}

Either use the standard LLE algorithm or Hessian Linear Local Embedding (HLLE).

nodeargs : None or dict

Arguments passed to the MDP node in various stages of its lifetime. See the baseclass for more details.

enable_ca : None or list of str

Names of the conditional attributes which should be enabled in addition to the default ones

disable_ca : None or list of str

Names of the conditional attributes which should be disabled

node : mdp.Node instance

This node instance is taken as the pristine source of which a copy is made for actual processing upon each training attempt.