lincc-frameworks / lincc-frameworks/hyrax

Mechanism for user-defined artifacts in MLFlow

Open
#393 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Python
Stars
41
Forks
7
Avg merge
5d 1h
Merged PRs (30d)
8

Description

Feature request
MLFlow does a good job of displaying any sort of artifact related to a training or inference run, however, Hyrax doesn't provide any sort of mechanism for users to define additional artifacts to be saved along side.

We've introduced a little bit of help for TensorFlow that allows users to save metrics related to training that are returned from train_step, but this request is more broad than that.

During the most recent Hyrax tech & science bi-weekly meeting, Peter Ferguson mentioned that there was a lot of effort put into saving sample images associated with training and inference. MLFlow can support this in a much more organized way, so we should investigate a friendly way to allow Hyrax users to exploit this.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing the existing TensorFlow support for metrics returned from train_step and how MLFlow organizes artifacts. Investigate a user-facing mechanism for saving additional artifacts, including sample images from training or inference. Done means Hyrax users can define and persist these artifacts in an organized MLFlow-compatible way.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, tensorflow
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.