mvpa2.featsel.base.FeatureSelection

Inheritance diagram of FeatureSelection

class mvpa2.featsel.base.FeatureSelection(filler=0, **kwargs)

Mapper to select a subset of features.

Depending on the actual slicing two FeatureSelections can be merged in a number of ways: incremental selection (+=), union (&=) and intersection (|=). Were the former assumes that two feature selections are applied subsequently, and the latter two assume that both slicings operate on the set of input features.

Notes

Available conditional attributes:

  • calling_time+: Time (in seconds) it took to call the node
  • raw_results: Computed results before invoking postproc. Stored only if postproc is not None.
  • training_time+: Time (in seconds) it took to train the learner

(Conditional attributes enabled by default suffixed with +)

Examples

>>> from mvpa2.datasets import *
>>> ds = Dataset([[1,2,3,4,5]])
>>> fs0 = StaticFeatureSelection([0,1,2,3])
>>> fs0(ds).samples
array([[1, 2, 3, 4]])

Merge two incremental selections: the resulting mapper performs a selection that is equivalent to first applying one slicing and subsequently the next slicing. In this scenario the slicing argument of the second mapper is relative to the output feature space of the first mapper.

>>> fs1 = StaticFeatureSelection([0,2])
>>> fs0 += fs1
>>> fs0(ds).samples
array([[1, 3]])

Methods

forward(data) Map data from input to output space.
forward1(data) Wrapper method to map single samples.
generate(ds) Yield processing results.
get_postproc() Returns the post-processing node or None.
get_space() Query the processing space name of this node.
reset()
reverse(data) Reverse-map data from output back into input space.
reverse1(data)
set_postproc(node) Assigns a post-processing node
set_space(name) Set the processing space name of this node.
train(ds) The default implementation calls _pretrain(), _train(), and finally _posttrain().
untrain() Reverts changes in the state of this node caused by previous training
Parameters :

filler : optional

Value to fill empty entries upon reverse operation

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

auto_train : bool

Flag whether the learner will automatically train itself on the input dataset when called untrained.

force_train : bool

Flag whether the learner will enforce training on the input dataset upon every call.

space: str, optional :

Name of the ‘processing space’. The actual meaning of this argument heavily depends on the sub-class implementation. In general, this is a trigger that tells the node to compute and store information about the input data that is “interesting” in the context of the corresponding processing in the output dataset.

postproc : Node instance, optional

Node to perform post-processing of results. This node is applied in __call__() to perform a final processing step on the to be result dataset. If None, nothing is done.

descr : str

Description of the instance

Methods

forward(data) Map data from input to output space.
forward1(data) Wrapper method to map single samples.
generate(ds) Yield processing results.
get_postproc() Returns the post-processing node or None.
get_space() Query the processing space name of this node.
reset()
reverse(data) Reverse-map data from output back into input space.
reverse1(data)
set_postproc(node) Assigns a post-processing node
set_space(name) Set the processing space name of this node.
train(ds) The default implementation calls _pretrain(), _train(), and finally _posttrain().
untrain() Reverts changes in the state of this node caused by previous training
reverse1(data)

NeuroDebian

NITRC-listed