Welcome to use Baidu Machine Learning(BML). This article aims to explain the operation of BML's management console and help you quickly apply the machine learning model to your own business. BML consists of three core modules:

• Model Training: Two models training methods are provided. You can choose the appropriate model development method according to your needs.

  • Notebook: A fully managed interactive programming environment Jupyter Lab is built in to implement data processing and code debugging.
  • Job modeling: Support multiple deep machine learning frameworks, launch large-scale training jobs with one key, and maximize training efficiency and effect. Including four types of job: deep learning job, machine learning job, AutoDL job, AutoML job.
• Model warehouse: store and manage the trained models in order according to different model categories, properties, classifications and versions.
• Prediction service: Quickly deploy the trained model as a highly available online service, flexibly select a variety of computing components to accelerate prediction execution, and complete model test iteration and service operation and maintenance management through A / B test, Beta test upgrade, service monitoring, etc.

Machine learning is a continuous cycle process. Model development-Model management-Release prediction services are used for production deployment. Then, you can combine more business data and retrain the model according to actual usage to improve prediction accuracy.

In addition to these three core modules, the navigation bar on the left side of the console has two modules:

• Item List: Display the item list under the user name. You can add, delete, modify and check items.
• Package: Purchase prepaid package.