eXtreme Gradient Boosting
What is this?
XGBoost is an optimized distributed gradient boosting library designed to be highly efficient, flexible and portable — it implements machine learning algorithms under the Gradient Boosting framework.
XGBoost is an optimized distributed gradient boosting library designed to be highly efficient, flexible and portable — it implements machine learning algorithms under the Gradient Boosting framework.Is it a program I install and run, or something I call from my own code?It is a library you call from your own code — it ships for Python, R, Java, Scala, C and more, with the core written in C++.
It is a library you call from your own code — it ships for Python, R, Java, Scala, C and more, with the core written in C++.Do I need a Spark or Hadoop cluster to use it?No — it runs on a single machine, and the same code also runs on major distributed environments (Kubernetes, Hadoop, SGE, Dask, Spark, PySpark), handling problems beyond billions of examples.
No — it runs on a single machine, and the same code also runs on major distributed environments (Kubernetes, Hadoop, SGE, Dask, Spark, PySpark), handling problems beyond billions of examples.Is it free to use, including commercially?Yes — XGBoost is licensed under Apache-2.0, so it is free to use in commercial projects.
Yes — XGBoost is licensed under Apache-2.0, so it is free to use in commercial projects.What does the algorithm actually do?It provides parallel tree boosting — gradient boosted decision trees, also known as GBDT or GBM — which solves many data science problems in a fast and accurate way.
It provides parallel tree boosting — gradient boosted decision trees, also known as GBDT or GBM — which solves many data science problems in a fast and accurate way.Where did XGBoost come from?It originated as a research project at the University of Washington and was published by Tianqi Chen and Carlos Guestrin as the paper 'XGBoost: A Scalable Tree Boosting System' at the 2016 SIGKDD conference.
It originated as a research project at the University of Washington and was published by Tianqi Chen and Carlos Guestrin as the paper 'XGBoost: A Scalable Tree Boosting System' at the 2016 SIGKDD conference.What are similar projects?LightGBM, spark, deeplearning4j, doris, flink-learning, dask, polars, mxnet
LightGBM, spark, deeplearning4j, doris, flink-learning, dask, polars, mxnetWhich topics is it in?flink, r, parallel, spark, jvm, big-data, data-analysis, high-performance, algorithms, data-science, distributed-systems, kubernetes
flink, r, parallel, spark, jvm, big-data, data-analysis, high-performance, algorithms, data-science, distributed-systems, kubernetes