Henry Gouk

Results 2 issues of Henry Gouk

**Describe the bug** I've noticed that the validation accuracy logged by dl4j is almost an exact inversion of the test set accuray reported by weka for some binary classification datasets...

I've noticed a few oddities in the multi-label prediction API: - There are multiple `MultiLabelInstance` classes: `moa.core.MultiLabelInstance` and `com.yahoo.labs.samoa.instances.MultiLabelInstance`. Should one of these be removed? - `BasicMultiLabelLearner` and `BasicMultiLabelClassifier` are...