Open ablaom opened 1 week ago
Thanks for catching this. I believe it's highly likely because I use recode
from categorical arrays during transform. This was intended to preserve the categorical type (which is useful in MLJTransforms where the indices are kept as integers and not floats).
I will confirm this and try to implement ordinal_encoder_transform
differently to fix. But maybe just not today.
In stepping through
fit
for NeuraNetworkRegressor, using the data at the top of the test fileregressors.jl
, I am getting some unexpected behaviour.Here is a minimal version of that data giving the same behaviour:
And the model:
Okay, now the following lines are copied from
fit
, as given in "src/mlj_model_iinterface.jl" on the dev branch:At this point I expect
X
to haveContinuous
scitype - no more categoricals. However:The raw element type is
Float32
but these are getting wrapped as categorical vectors.