tensorflow / tensorflow/model-optimization

[clustering] Possible wrong call of the centroids initializer from the ClusterWeights wrapper

Open
#939 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
Python
Stars
1.6k
Forks
349
Avg merge
3d 2h
Merged PRs (30d)
1

Description

An instance of the centroids initializer expects to be initialized with data_format.
However, instead of data_format it gets num_channels here: https://github.com/tensorflow/model-optimization/blob/v0.7.1/tensorflow_model_optimization/python/core/clustering/keras/cluster_wrapper.py#L218 I suppose that line is erroneous. This may cause looping over the the wrong axis during clusterization when data_format is channels_first.

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

Inspect tensorflow_model_optimization/python/core/clustering/keras/cluster_wrapper.py around line 218 and compare the call with clustering_centroids.py line 47. Check the channels_first path and relevant clustering behavior; done means the initializer receives the intended data-format value and axis handling is correct.

Written by the indexing model from the issue text.

Assessment

Tech stack
keras, python, tensorflow
Domain
machine-learning
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Clearly specified
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.