Sparse Matrix Support in R

catboost / catboost

A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification, regression and other machine learning tasks for Python, R, Java, C++. Supports computation on CPU and GPU.

https://catboost.ai

Apache License 2.0

8.09k stars 1.19k forks source link

Sparse Matrix Support in R #1407

Open ThomasWolf0701 opened 4 years ago

ThomasWolf0701 commented 4 years ago

Problem: catboost version: R package 0.24.1 Operating System: Windows CPU: Intel Core i7

The R version does not seem to support sparse matrices Matrix::sparseMatrix(). Would be really great if this would be enabled, especially as this would be fully compatible with mlr3

Or maybe it is available and I just missed it in the package ? But contrary to Xgboost and Lightgbm Catboost does not seem to support dgCMatrix in R. For Xgboost this works really well.

Evgueni-Petrov-aka-espetrov commented 4 years ago

Many thanks! Sounds interesting. Will look into it soon. You are correct, we do not support sparse matrices in R yet.