Skip to content

Model Cards Endpoint#126

Merged
Mesh-ach merged 13 commits into
stagingfrom
develop
Jul 10, 2025
Merged

Model Cards Endpoint#126
Mesh-ach merged 13 commits into
stagingfrom
develop

Conversation

@Mesh-ach
Copy link
Copy Markdown
Contributor

@Mesh-ach Mesh-ach commented Jul 9, 2025

changes

  • Calls mlflow.set_tracking_uri("databricks") and downloads the specified runs:/<run_id>/<artifact_path> PDF into a secure TemporaryDirectory.
  • Streams the PDF back as a FileResponse with application/pdf and Content-Disposition: attachment.

context

  • Part of a list of endpoints needed for front-end work.

questions

No questions.


@Mesh-ach Mesh-ach requested a review from vishpillai123 July 9, 2025 17:48
Comment thread pyproject.toml
"fuzzywuzzy",
"databricks-sql-connector",
"pandera~=0.13"
"pandera~=0.13",
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

eventually we may need to specify versions for the rest like you've done here for pandera

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, to be honest, I wasn’t too sure what the other versions should be but I totally agree.

Copy link
Copy Markdown
Collaborator

@vishpillai123 vishpillai123 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

hey @Mesh-ach, the get_model_cards(...) endpoint looks good to me. I'm not familiar with dynamic endpoints like this but overall the logic looks good.

Also how do we test/ensure that these endpoints work? I realize it's challenging since that connects to the other repo. Is the best way to test to launch the api locally?

@Mesh-ach
Copy link
Copy Markdown
Contributor Author

hey @Mesh-ach, the get_model_cards(...) endpoint looks good to me. I'm not familiar with dynamic endpoints like this but overall the logic looks good.

Also how do we test/ensure that these endpoints work? I realize it's challenging since that connects to the other repo. Is the best way to test to launch the api locally?

Right now, testing locally is complicated. Since we connect to GCP and then Databricks, some components can be tested locally, but others must be deployed and tested in staging. So we run all tests in staging to ensure everything works.

@Mesh-ach Mesh-ach merged commit 50b2ac7 into staging Jul 10, 2025
9 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants