MLBazaar / MLBazaar/piex

Make MLBlocks an optional dependency

Open
#9 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Jupyter Notebook
Stars
17
Forks
6
PR merge metrics
No merged PRs in 30d

Description

MLBlocks is a pretty heavy-weight dependency as it requires tensorflow, xgboost, etc. to be installed. To just "explore" pipelines from S3, users shoudn't need MLBlocks.

Desired functionality:

  • for all methods except PipelineExplorer.score_pipeline and Pipeline.score_template, MLBlocks should not need to be installed
  • if the user wants to score pipelines or templates using PipelineExplorer.score_pipeline and Pipeline.score_template, they can enable this by pip install piex[mlblocks] (or piex[scoring] or piex[all] or some other variant)

Note that this would require a workaround for using MLPipeline to load pipelines/templates for the express purpose of extracting metadata from them

Contributor guide

Open the contributing guide

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 tracing PipelineExplorer.score_pipeline, Pipeline.score_template, and MLPipeline loading to see how MLBlocks is currently imported for exploration and scoring. Check the package installation options mentioned in the issue and verify that S3 exploration works without MLBlocks while both scoring entry points remain available with the scoring extra.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, tensorflow
Domain
build-system, machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.