lablup / lablup/backend.ai

Integrate Model Store feature with Model Reservoir service

Open
#7,694 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
670
Forks
183
Avg merge
17h 7m
Merged PRs (30d)
358

Description

## **Client Workflow Overview**

1. On the model reservoir page, the user clicks the **“Model Import to Model Store”** button.
(The user can download the model to their own VFS, object storage, or the **_model store_**.)
1. A page for selecting a vfolder is displayed.
On this page, the user can either create a **model-type vfolder** or select an existing one.
1. The name of the selected vfolder is passed to the `import_artifacts` mutation, so that the model is downloaded to the path of that vfolder instead of the preconfigured storage.
(This requires a feature extension.)

## **Additional Considerations**

1. We need to consider providing a UI for creating and editing `service-definition.toml`.
1. It would be great if there were a way to automatically generate `service-definition.toml` when a vfolder is created directly.
1. It would help with automation if some metadata from `model-definition.yaml` could be automatically extracted and populated at model import time.

JIRA Issue: BA-3637

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.