mvpa2.measures.gnbsearchlight.GNBSearchlight

Inheritance diagram of GNBSearchlight

class mvpa2.measures.gnbsearchlight.GNBSearchlight(gnb, generator, qe, errorfx=<function mean_mismatch_error at 0x6918578>, indexsum=None, **kwargs)

Efficient implementation of Gaussian Naive Bayes Searchlight.

This implementation takes advantage that GNB is “naive” in its reliance on massive univariate conditional probabilities of each feature given a target class. Plain Searchlight analysis approach asks for the same information over again and over again for the same feature in multiple “lights”. So it becomes possible to drastically cut running time of a Searchlight by pre-computing basic statistics necessary used by GNB beforehand and then doing their subselection for a given split/feature set.

Kudos for the idea and showing that it indeed might be beneficial over generic Searchlight with GNB go to Francisco Pereira.

Notes

Available conditional attributes:

  • calling_time+: Time (in seconds) it took to call the node
  • null_prob+: None
  • null_t: None
  • raw_results: Computed results before invoking postproc. Stored only if postproc is not None.
  • roi_feature_ids: Feature IDs for all generated ROIs.
  • roi_sizes: Number of features in each ROI.
  • training_time+: Time (in seconds) it took to train the learner

(Conditional attributes enabled by default suffixed with +)

Initialize a GNBSearchlight

Parameters :

gnb : GNB

GNB classifier as the specification of what GNB parameters to use. Instance itself isn’t used.

generator : Generator

Some Generator to prepare partitions for cross-validation.

errorfx : func, optional

Functor that computes a scalar error value from the vectors of desired and predicted values (e.g. subclass of ErrorFunction).

indexsum : (‘sparse’, ‘fancy’), optional

What use to compute sums over arbitrary columns. ‘fancy’ corresponds to regular fancy indexing over columns, whenever in ‘sparse’, produce of sparse matrices is used (usually faster, so is default if scipy is available.

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

queryengine : QueryEngine

Engine to use to discover the “neighborhood” of each feature. See QueryEngine.

roi_ids : None or list(int) or str

List of feature ids (not coordinates) the shall serve as ROI seeds (e.g. sphere centers). Alternatively, this can be the name of a feature attribute of the input dataset, whose non-zero values determine the feature ids. By default all features will be used.

null_dist : instance of distribution estimator

The estimated distribution is used to assign a probability for a certain value of the computed measure.

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

errorfx
generator
gnb
indexsum

NeuroDebian

NITRC-listed