catsnip
is a package that wraps catboost
models in parsnip
format.
This package is based on treesnip
package and provides following
functions.
Attach catboost
model to boost_tree()
function
Install catboost
package through install_catboost()
function
Avoid conflict between bonsai
package and catsnip
package during
stacking model
For a detailed explanation of catboost
and treesnip
, see links
below.
Since catsnip
is not listed on CRAN, so use install_github()
function.
devtools::install_github(repo="csh01470/catsnip")
You can also install catboost
released version with,
catsnip::install_catboost()
To install another version(ex: 0.16.5
), Use version
parameter.
parsnip | catboost |
---|---|
mtry | rsm |
trees | iterations |
min_n | min_data_in_leaf |
tree_depth | depth |
learn_rate | learning_rate |
loss_reduction | ??? |
sample_size | subsample |
stop_iter | early_stopping_rounds |
[x] Support GPU
processing
[x] Optimization of parameters for catboost
model and boost_tree()
function
[ ] Extract SHAP value by interworking with fastshap
(or shapviz
?)
package
Please note that the catsnip
project is released with a Contributor
Code of
Conduct.
By contributing to this project, you agree to abide by its terms.