xpresso Version Control
Get Visibility Into Four Quadrants Of AI/ML Applications – Code, Datasets, Workflow And Models
Easy to use version control tools that helps arrive at optimum results with the ability to time-travel to previous versions as needed
Data Version Control
- Store and version all types of data and metadata (structured or unstructured)
- Bring together datasets from multiple sources into a repository
- Pull datasets in your workflows directly using libraries, APIs or pre-built components
Code Version Control
- Integrated with code repositories such as GitLab, Bitbucket and interactive notebooks such as JupyterHub
- Keep track of all changes and bridge communication between teams
- Leverage the language agnostic capabilities to use multiple programming languages within your solution
- Store dependencies along with the code
Experiment & Model Version Control
- Seamless model tracking and model versioning with a central repository to store and visualize every experiment and model along with datasets and parameters
- Version models and rollback to previous experimentation and results when needed
- Compare experiments and deploy the most optimum model that suits the business needs
- Easily reproduce and transfer models into any environment
