Trusted-AI / Trusted-AI/AIX360

CEM for Multi-label Classification

Open
#29 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.8k
Forks
327
Avg merge
1h 13m
Merged PRs (30d)
1

Description

Inspecting the KerasClassifier class for CEM, I could see that it is made specifically for single label classification:
predicted_class = np.argmax(prob)

It may be important to prepare this class to handle Multi Label Multi Class classification, maybe allowing the programmer to select which of the N target classes to take into account when explaining with CEMExplainer.

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 with aix360/algorithms/contrastive/classifiers.py, especially KerasClassifier around the referenced prediction logic, then trace how CEMExplainer uses it. Determine the expected behavior for multi-label, multi-class models and how a target class is selected. Done means CEM can explain a chosen target class without assuming single-label output, with suitable coverage in the existing test structure if one is found.

Written by the indexing model from the issue text.

Assessment

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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.