25 Xgboost model reference
As discussed in the Machine Learning chapter, this book contains reference chapters for machine learning models that can be registered in metamorph.ml.
This specific chapter focuses on the XGBoost algorithm provided by scicloj.ml.xgboost.
In the following we have a list of all model keys of Xgboost models including parameters. They can be used like this:
(comment
(ml/train df
{:model-type <model-key>
:param-1 0
:param-2 1}))25.1 :xgboost/binary-hinge-loss
25.2 :xgboost/classification
25.3 :xgboost/count-poisson
25.4 :xgboost/gamma-regression
25.5 :xgboost/gpu-binary-logistic-classification
25.6 :xgboost/gpu-binary-logistic-raw-classification
25.7 :xgboost/gpu-linear-regression
25.8 :xgboost/gpu-logistic-regression
25.9 :xgboost/linear-regression
| name | type | default | description |
|---|---|---|---|
| gamma | |||
| max-depth | |||
| min-child-weight | |||
| max-delta-step | |||
| subsample | |||
| sampling-method | |||
| colsample-bytree | |||
| colsample-bylevel | |||
| colsample-bynode | |||
| lambda | |||
| alpha | |||
| tree-method | |||
| sketch-eps | |||
| scale-pos-weight | |||
| updater | |||
| refresh-leaf | |||
| process-type | |||
| grow-policy | |||
| max-leaves | |||
| max-bin | |||
| predictor | |||
| num-parallel-tree | |||
| monotone-constraints | |||
| interaction-constraints |
25.10 :xgboost/logistic-binary-classification
25.11 :xgboost/logistic-binary-raw-classification
25.12 :xgboost/logistic-regression
25.13 :xgboost/multiclass-softmax
25.14 :xgboost/multiclass-softprob
25.15 :xgboost/rank-map
25.16 :xgboost/rank-ndcg
25.17 :xgboost/rank-pairwise
25.18 :xgboost/regression
25.19 :xgboost/squared-error-regression
25.20 :xgboost/survival-cox
25.21 :xgboost/tweedie-regression
source: notebooks/noj_book/xgboost.clj