This topic describes how to create an experiment in Machine Learning Platform for AI.

To create an experiment in Machine Learning Platform for AI, perform the following operations:

  1. Create a project

    Machine Learning Studio uses projects to manage resources, permissions, and experiments. You can create one or more experiments for each project.

  2. Data preparation

    Upload raw data to MaxCompute or Object Storage Service (OSS), and configure the data source for the experiment.

  3. Data preprocessing

    Preprocess the raw data to generate model training sets and model prediction sets.

  4. Data visualization

    Visualize the experiment data to obtain data information.

  5. Algorithm modeling

    Use model training sets to perform algorithm modeling and generate a training model.

  6. Model evaluation

    Use the trained model to make predictions of model prediction sets to evaluate the precision of the model.