With VESSL, machine learning researchers can run experiments and deploy models on Kubernetes clusters without any background in software engineering or DevOps. A typical workflow on VESSL is composed of 5 steps:
- 1.Allocate machine resource according to the needs of the Project.
- 2.Import project source code from GitHub.
- 3.Upload a dataset from local disk or cloud vendors.
- 4.Run experiments and use Sweep to find the optimal hyperparameter.
- 5.Deploy models into production as REST APIs.