Package mvpa :: Package datasets :: Module splitters
[hide private]
[frames] | no frames]

Module splitters

source code

Collection of dataset splitters.

Module Description

Splitters are destined to split the provided dataset varous ways to simplify cross-validation analysis, implement boosting of the estimates, or sample null-space via permutation testing.

Most of the splitters at the moment split 2-ways -- conventionally first part is used for training, and 2nd part for testing by CrossValidatedTransferError and SplitClassifier.

Brief Description of Available Splitters

Module Organization

Package Tree for splitters
Package Tree for splitters

Classes [hide private]
  Splitter
Base class of dataset splitters.
  NoneSplitter
This is a dataset splitter that does not split. It simply returns the full dataset that it is called with.
  OddEvenSplitter
Split a dataset into odd and even values of the sample attribute.
  HalfSplitter
Split a dataset into two halves of the sample attribute.
  NGroupSplitter
Split a dataset into N-groups of the sample attribute.
  NFoldSplitter
Generic N-fold data splitter.
  CustomSplitter
Split a dataset using an arbitrary custom rule.

Imports: operator, N, support, enhancedDocString, coarsenChunks, debug