From 4ff7cbedd3b0f1ce9bf4c9887de61b53d7f3288c Mon Sep 17 00:00:00 2001 From: Essam Date: Tue, 6 Aug 2024 22:46:04 -0500 Subject: [PATCH] Update src/types.jl Co-authored-by: Anthony Blaom, PhD --- src/types.jl | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/types.jl b/src/types.jl index d89a082..44f1241 100644 --- a/src/types.jl +++ b/src/types.jl @@ -199,7 +199,9 @@ as described in Entity Embeddings of Categorical Variables by Cheng Guo, Felix B const XDOC = """ - `X` is either a `Matrix` or any table of input features (eg, a `DataFrame`) whose columns are of scitype - `Finite`; check column scitypes with `schema(X)`. If `X` is a `Matrix`, it is assumed to have columns corresponding to features and rows corresponding to observations. + `Multiclass` or `OrderedFactor`; check column scitypes with `schema(X)`. + If `X` is a `Matrix`, it is assumed to have columns corresponding to features and rows + corresponding to observations. In case any `Multiclass` or `OrderedFactor` columns appear in `X`, the constructed network will use an `EntityEmbedder` layer to transform them into dense vectors. """