NVIDIA-Merlin / NVIDIA-Merlin/Merlin
[RMP] Support individual Vertex AI tools (like feature store) for serving recs with Merlin
Open
@mikemckiernan is already working on this.
Since Aug 3, 2022.
roadmap
- Dominant language
- Python
- Stars
- 907
- Forks
- 129
- PR merge metrics
- No merged PRs in 30d
Description
Problem:
As a current Vertex AI user, I want to use the managed tools provided by Vertex AI instead of running my own real-time services for model serving, such as a feature store, ANN engine, model registry, or Triton server.
Goal:
- Produce a notebook with an example of building a Systems ensemble using Vertex AI operators and deploy to Vertex AI Prediction.
Constraints:
- Getting access to a GCP project.
More research is needed in order to identify further constraints.
Starting Point:
-
Provide operators to integrate with the Vertex AI tools for serving:
- Feature Store
- Matching Engine (ANN service)
- Model Registry
-
Create an example demonstrating how Merlin works within the Vertex ecosystem
-
Add documentation for best practices of Vertex+Merlin
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.