[BUG] Cluster predict proba edge cases

chrisholder commented 2 days ago

Reference Issues/PRs

What does this implement/fix? Explain your changes.

There were two edge cases with the clustering predict_proba function that would cause crashes.

1) If there wasn't at least 1 prediction in every cluster 2) If the prediction didn't use cluster values of 0-n_clusters. For example I used a k-medoids package that when predict was called returned [22, 22, 32, 3, 32, 22] where instead of the cluster values being a value from 0-n_clusters it was referring to the index of the cluster centre that each value was assigned to.

This PR makes two main changes to address the above. 1) Instead of using .shape[0] to create the dist array n_clusters is used (when available isn't required):

            n_clusters = self.n_clusters
            if n_clusters is None:
                n_clusters = int(max(preds)) + 1
            dists = np.zeros((X.shape[0], n_clusters))

2) A additional sanitisation of the predict result is done to convert predictions value to 0-n_clusters. This allows the other logic to work if the predictions doesn't return 0-n_clusters:

  unique = np.unique(preds)
  for i, u in enumerate(unique):
      preds[preds == u] = i

Does your contribution introduce a new dependency? If yes, which one?

Any other comments?

PR checklist

For all contributions

[ ] I've added myself to the list of contributors. Alternatively, you can use the @all-contributors bot to do this for you.
[ ] The PR title starts with either [ENH], [MNT], [DOC], [BUG], [REF], [DEP] or [GOV] indicating whether the PR topic is related to enhancement, maintenance, documentation, bugs, refactoring, deprecation or governance.

For new estimators and functions

[ ] I've added the estimator to the online API documentation.
[ ] (OPTIONAL) I've added myself as a __maintainer__ at the top of relevant files and want to be contacted regarding its maintenance. Unmaintained files may be removed. This is for the full file, and you should not add yourself if you are just making minor changes or do not want to help maintain its contents.

For developers with write access

[ ] (OPTIONAL) I've updated aeon's CODEOWNERS to receive notifications about future changes to these files.

aeon-actions-bot[bot] commented 2 days ago

Thank you for contributing to `aeon`

I have added the following labels to this PR based on the title: [ $\color{#d73a4a}{\textsf{bug}}$ ]. I have added the following labels to this PR based on the changes made: [ $\color{#4011F3}{\textsf{clustering}}$ ]. Feel free to change these if they do not properly represent the PR.

The Checks tab will show the status of our automated tests. You can click on individual test runs in the tab or "Details" in the panel below to see more information if there is a failure.

If our pre-commit code quality check fails, any trivial fixes will automatically be pushed to your PR unless it is a draft.

Don't hesitate to ask questions on the aeon Slack channel if you have any.

PR CI actions

These checkboxes will add labels to enable/disable CI functionality for this PR. This may not take effect immediately, and a new commit may be required to run the new configuration.

[ ] Run pre-commit checks for all files
[ ] Run all pytest tests and configurations
[ ] Run all notebook example tests
[ ] Run numba-disabled codecov tests
[ ] Stop automatic pre-commit fixes (always disabled for drafts)

TonyBagnall commented 2 days ago

chrisholder commented 1 day ago

No this is just a general bug fix

aeon-toolkit / aeon