Development Process

Table 1 describes each phase in developing the machine learning algorithm library.

Figure 1 Development process of the machine learning algorithm library

**Table 1** Development process of the graph analysis algorithm library
Phase	Description
Setting up the development environment	The machine learning algorithm library can be developed by using Scala, Java, Python, and R. This document uses Scala as an example to describe how to use IDEA to set up the development environment.
Creating a project	You can create a machine learning algorithm library project according to related guide.
Developing an application	You can select a machine learning algorithm for development based on your service scenarios. This chapter uses the GBDT algorithm as an example to describe how to develop machine learning algorithms.
Executing the task and viewing the result	This phase describes how to submit a developed application to a cluster for running. The application running result will be printed. You can view the application status on the UI.

Before application development, you need to understand the basic concepts of machine learning algorithm library, Spark Core, and the Spark algorithm library. For details, see Machine Learning Algorithm Libraries.

Parent topic: Example Projects