Overview
Overview of machine learning Models in Gradient
Last updated
Overview of machine learning Models in Gradient
Last updated
This section of the documentation covers our previous generation of Gradient. For the current version go to Gradient Next.
The Gradient model repository is a hub for importing, managing, and deploying ML models. Gradient offers the ability to create your own custom models and in addition to launching pre-trained public models.
Gradient Models can be created in two ways: they can be 1) generated by running machine learning Experiments, or 2) imported into Gradient by uploading them directly from the Web UI or CLI. Learn more here.
Gradient has a repository of Models per team, as well as a Public Models repository. If you want a Generated Model to appear in your team's Model Repository, be sure to use the appropriate Model Path. Uploaded Models will automatically be placed in your team's Model Repository.
The Model Repository holds references to all uploaded Models and any Experiment-generated Models. This includes model and checkpoint files generated during an Experiment's training period as well as summary metrics associated with the model's performance, such as accuracy and loss.
Supported Models:
ONNX (Open Neural Network Exchange)
Custom
