This content refers to an unreleased development version of PyMVPA
mvpa.datasets.splittersΒΆ
Collection of dataset splitters.
Splitters are destined to split the provided dataset various 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.
Classes
| CustomSplitter(splitrule, **kwargs) | Split a dataset using an arbitrary custom rule. |
| HalfSplitter(**kwargs) | Split a dataset into two halves of the sample attribute. |
| NFoldSplitter(**kwargs[, cvtype]) | Generic N-fold data splitter. |
| NGroupSplitter(**kwargs[, ngroups]) | Split a dataset into N-groups of the sample attribute. |
| NoneSplitter(**kwargs) | Non-splitting Splitter for resampling purposes. |
| OddEvenSplitter(**kwargs[, usevalues]) | Split a dataset into odd and even values of the sample attribute. |
| Splitter([npertarget, nrunspersplit, ...]) | Base class of dataset splitters. |
