Workflows automate machine learning tasks, combining GPU instances with an expressive syntax to generate production-ready machine learning pipelines with a few lines of code.
Gradient Workflows provides a way to automate machine learning tasks using CI/CD and build machine learning applications. Workflows utilize GitHub-action style syntax via YAML files to create automation. You can install the GitHub app on any repo and connect to a Gradient project to train models directly from pull requests or commits.
Workflows is based on the Argo runtime engine, which is a container-native continuous delivery tool for Kubernetes. With Workflows, you can build complex and scalable projects with an arbitrary number of discrete steps and create continuously updated machine learning models. You can deploy your model with a RESTful API.
Workflows is used by ML engineers to build deterministic machine learning pipelines.