https://github.com/lantanacamara/lightgbmExplainer

An R package that makes lightgbm models fully interpretable (take reference from https://github.com/AppliedDataSciencePartners/xgboostExplainer)
https://github.com/lantanacamara/lightgbmExplainer

lightgbm

Last synced: about 1 month ago
JSON representation

An R package that makes lightgbm models fully interpretable (take reference from https://github.com/AppliedDataSciencePartners/xgboostExplainer)

• Host: GitHub
• URL: https://github.com/lantanacamara/lightgbmExplainer
• Owner: lantanacamara
• Created: 2017-10-09T08:16:14.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2019-08-08T05:15:42.000Z (over 6 years ago)
• Last Synced: 2025-05-30T12:18:09.656Z (10 months ago)
• Topics: lightgbm
• Language: R
• Homepage:
• Size: 25.4 KB
• Stars: 23
• Watchers: 3
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-machine-learning-interpretability - lightgbmExplainer -  | "An R package that makes LightGBM models fully interpretable."| (Technical Resources / Open Source/Access Responsible AI Software Packages)

README

          
[![Build Status](https://travis-ci.org/lantanacamara/lightgbmExplainer.svg?branch=master)](https://travis-ci.org/lantanacamara/lightgbmExplainer)

## lightgbmExplainer

An R package that makes LightGBM models fully interpretable

### Example 

```

library(lightgbm) # v2.1.0 or above

library(lightgbmExplainer)

# Load Data

data(agaricus.train, package = "lightgbm")

# Train a model

lgb.dtrain <- lgb.Dataset(agaricus.train$data, label = agaricus.train$label)

lgb.params <- list(objective = "binary")

lgb.model <- lgb.train(lgb.params, lgb.dtrain, 5)

# Build Explainer

lgb.trees <- lgb.model.dt.tree(lgb.model) # First get a lgb tree

explainer <- buildExplainer(lgb.trees)

# compute contribution for each data point

pred.breakdown <- explainPredictions(lgb.model, explainer, agaricus.train$data)

# Show waterfall for the 8th observation

showWaterfall(lgb.model, explainer, lgb.dtrain, agaricus.train$data,  8, type = "binary")

```

Take reference from [xgboostExplainer](https://github.com/AppliedDataSciencePartners/xgboostExplainer) and credit to David Foster.

Note: LightGBM provides similar function *lgb.interprete* and *lgb.plot.interpretation*. *lgb.interprete* could be faster if you only want to interprete a few data point, but it could be much slower if you want to interprete many data point.