Add new argument for limiting the maximum epsilon#529
Merged
Conversation
Collaborator
|
Sorry for taking so long to get to this. It looks like a useful addition. Any chance you could add a test to the test suite to check that it works as intended? |
Contributor
Author
I totally missed your comment:s I'll do that yes. |
# Conflicts: # hdbscan/hdbscan_.py
… cluster_label_map
Contributor
|
Hi @lmcinnes! :D It has been a while since the last update in this PR. Could you take another look? Thanks! |
Contributor
|
Gentle reminder to revisit this PR @lmcinnes |
Collaborator
|
Thanks. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR aims to introduce to HDBSCAN an argument for a max threshold to the epsilon used when picking the best clusters. With this PR we allow for this new argument,
cluster_selection_epsilon_max, to be used in the EOM search method.This is very useful for cases where you know from the get go that your samples should not be very far from each other, because you have some domain knowledge.
For this implementation, we use
cluster_selection_epsilon_maxin a very similar way tomax_cluster_size. This way the clusters with an epsilon bigger thancluster_selection_epsilon_maxcan still appear if there are no valid clusters bellow that epsilon. This is, in fact, the exact same behavior asmax_cluster_size.