A bug in elm.pipeline.steps resulted in DropNaRows(X, y) dropping NaN rows from the X matrix without dropping the corresponding rows from y. The outcome was a dimension error on fitting. I did not add a test for this fix. I'll make a separate issue to better test:
supervised learning (this bug was discovered at this time because we have lower test/docs/examples coverage of supervised learning than unsupervised)
elm.pipeline.steps.DropNaRows
I am deferring the testing on those items above, rather than adding them to this PR, because we are about to start a 4 to 8 week major refactor of elm.sample_util, moving the Pipeline of filter functions like DropNaRows to earthio.filters (a new subpackage). See also #149 for details on that refactor.
A bug in
elm.pipeline.stepsresulted inDropNaRows(X, y)dropping NaN rows from theXmatrix without dropping the corresponding rows fromy. The outcome was a dimension error on fitting. I did not add a test for this fix. I'll make a separate issue to better test:elm.pipeline.steps.DropNaRowsI am deferring the testing on those items above, rather than adding them to this PR, because we are about to start a 4 to 8 week major refactor of
elm.sample_util, moving thePipelineof filter functions likeDropNaRowstoearthio.filters(a new subpackage). See also #149 for details on that refactor.