[ENH] Sort out clustering base class #2251
Conversation
aeon-actions-bot
bot
added
the
enhancement
Thank you for contributing to
|
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
There was a problem hiding this comment.
@hadifawaz1999 is this the one you wanted delayed?
Approving for if thats fine. Left some comments.
| if hasattr(self, "n_clusters"): | ||
| n_clusters = self.n_clusters | ||
| else: | ||
| n_clusters = len(np.unique(preds)) |
There was a problem hiding this comment.
Think this could be risky down the line if there are non-int methods of generating n_clusters, but no solution for now other than always using unique.
| @@ -173,10 +173,10 @@ def __init__( | |||
| self.save_last_model = save_last_model | |||
| self.best_file_name = best_file_name | |||
| self.random_state = random_state | |||
| self.n_clusters = n_clusters | |||
There was a problem hiding this comment.
Can keep this for now, but I think @hadifawaz1999 said these would be better removed (n_clusters should be input with estimator)
There was a problem hiding this comment.
i think it should not be added as we're gonna remove it in a few days
| if hasattr(self, "labels_"): | ||
| return self.labels_ |
There was a problem hiding this comment.
Ideally, everything should have this.
no the one am delaying is Chris's pr of deprecation, but the additions here to deep clustering i think should not be made @TonyBagnall as they're gonna be removed with the purging in a few days once all Aadya's prs are in |
|
could also do this at the same time @chrisholder ? |
Fixes #1530