search

dotnet / machinelearning

ML.NET is an open source and cross-platform machine learning framework for .NET.
https://dot.net/ml
MIT License
8.91k stars 1.86k forks source link

Specify Categorical Features in LightGBM #7137

Open matnatx opened 2 months ago

matnatx commented 2 months ago

I'm trying to use LightGBM algorithm in ML.NET to train a model using a dataset.

The dataset has a number of categorical features such as "gender", "race", "country_of_birth".

I can convert/encode these categorical data as integers.

My question is how I can specify that these features are categorical and not (continuous variables), so that LightGBM can handle them as categorical data.

I can use OneHotEncoding/OneHotHashEncoding, but I believe it's not the same.

TIA