Overview

This chapter describes the functions, input, output, and usage examples of all algorithm interfaces in the Kunpeng BoostKit machine learning algorithm library.

For details about the API parameters of native algorithms, see the official website. The default parameter values may vary according to Spark versions.

For example, for Spark 2.3.2, see https://archive.apache.org/dist/spark/docs/2.3.2/api/scala.
For Spark 2.4.6, see https://archive.apache.org/dist/spark/docs/2.4.6/api/scala.
For Spark 3.1.1, see https://archive.apache.org/dist/spark/docs/3.1.1/api/scala/org/apache/spark/.

The value ranges of API parameters and parameters of all existing machine learning algorithms follow Spark conventions. If you set parameter values beyond the defined ranges, the corresponding Spark task may exit unexpectedly.
API applicability:
- The algorithm APIs apply to the Kunpeng-based HDP platform. For the Java development environment, Java 1.8 or later is required. For the Spark development environment, Spark 2.3.2, 2.4.5, 2.4.6, or 3.1.1 is required. HDP 3.1.0 or later is required.
- To run the algorithm on the HDP platform, you need to deploy components HDFS, Spark, Yarn, and ZooKeeper.
- The Kunpeng BoostKit machine learning algorithm library for Spark 2.4.6 uses the same core code as the native algorithm library for Spark 2.3.2, so their algorithm execution results are the same. The algorithm execution results may be different from the native algorithm library for Spark 2.4.6 (for example, DTB), depending on whether there are functional changes between algorithm libraries for open source Spark 2.3.2 and Spark 2.4.6.

Parent topic: Algorithm APIs