Machine Learning Studio provides a visual experiment and development environment for machine learning. It allows you to develop AI services without the need to write code. In addition, Machine Learning Studio provides multiple proven machine learning algorithms for you to meet your business requirements in different scenarios, such as product recommendation, financial risk management, and advertising prediction.
You can log on to Machine Learning Studio by using an Alibaba Cloud account or a Resource Access Management (RAM) user. If you use a RAM user, you must first use your Alibaba Cloud account to authorize the RAM user. For more information, see Authorization.
Machine Learning Studio allows you to use a template to create an experiment or manually create an experiment. Template-based experiment creation is simple. After you create an experiment by using a template and run the experiment, you can deploy models. If you manually create an experiment, the system provides more than one hundred algorithm components and supports access to various data sources, such as MaxCompute table data and Object Storage Service (OSS) data.
When you train models, the system allows you to use Automated Machine Learning (AutoML) to automatically tune parameters and export models in the Predictive Model Markup Language (PMML) format. This helps you obtain the optimal models.
- Traditional machine learning components
These components include data preprocessing, feature engineering, statistical analysis, time series, text analysis, and network analysis.
- Deep learning framework components
These components include TensorFlow, Caffe, MXNet, and PyTorch.