Skip to content

Brush Up On Benchmark - #16

Merged
KIwabuchi merged 38 commits into
llnl:developfrom
KIwabuchi:feature/brush_up_bench
Sep 17, 2026
Merged

Brush Up On Benchmark#16
KIwabuchi merged 38 commits into
llnl:developfrom
KIwabuchi:feature/brush_up_bench

Conversation

@KIwabuchi

@KIwabuchi KIwabuchi commented Sep 16, 2026 •
edited
Loading

Copy link
Copy Markdown
Member

Key changes:

  • Added a --cluster_noise_points flag and related logic to run_clams_bench.py to optionally cluster noise points after the main clustering process.

  • Removed the --distributed_hdbscan option and always use the distributed clustering executable by default, simplifying the clustering logic.

KIwabuchi added 30 commits August 7, 2026 16:47
@KIwabuchi
Change default AMST approx bound value to 1.1
8b09aa4
@KIwabuchi
Use MFC by default
a8e6204
@KIwabuchi
Brush up scripts
805652b
@KIwabuchi
Update bench scripts
6932cc6
@KIwabuchi
Update bench scripts
3ad72d6
@KIwabuchi
Minor code brush ups
e10e513
@KIwabuchi
Update bench script
5c17de2
@KIwabuchi
Assign clusters to noise points
8189ff9
@KIwabuchi
Add copy_pm_datastore
495a903
@KIwabuchi
Fix build error in cluster_noise_points.cpp
a7f510d
@KIwabuchi
Improve run_clams_bench.py
726511f
@KIwabuchi
Enhance clustering evaluation and correlation analysis
756c207 
- Updated eval_clusters function to include an option to ignore true noise points.
- Added new tool for evaluating correlation between kNNG and clusters.
@KIwabuchi
Merge remote-tracking branch 'upstream/develop' into feature/brush_up…
cc8f3e2 
…_bench
@KIwabuchi
Enhance documentation for kNNG functions and correlation evaluation
0c57721
@KIwabuchi
Refactor clustering evaluation output and streamline echo statements …
b4f47fd 
…in job script
@KIwabuchi
Refactor clustering execution logic into a dedicated function for imp…
c200068 
…roved readability and maintainability
@KIwabuchi
Minor bugfix in cluster_noise_points.cpp
2f02c26
@KIwabuchi
Minor bugfix in cluster_noise_points.cpp
c030037
@KIwabuchi
Brush up benchmark scripts
4dda3ca
@KIwabuchi
Brush up benchmark scripts
15acaa4
@KIwabuchi
Brush up benchmark scripts
b0b78d9
@KIwabuchi
Update output header in extract_clams_clustering_time.sh
17ec574
@KIwabuchi
Merge remote-tracking branch 'upstream/develop' into feature/brush_up…
c66d607 
…_bench

# Conflicts:
#	script/benchmark/extract_clams_clustering_time.sh
@KIwabuchi
Bugfix: fix to detect noise points correctly.
3ab6ad5
@KIwabuchi
Brush up on the HDBSCAN script
1ca4060
@KIwabuchi
Add knng and clusters correlation evaluation script
13db1e7
@KIwabuchi
Add MST-based clusetring complement v2
0164955
@KIwabuchi
Merge remote-tracking branch 'upstream/develop' into feature/brush_up…
be532ad 
…_bench
@KIwabuchi
Change to use parallel PM datastore copy program when back up KNNG
b97dde9
@KIwabuchi
Bugfix: Boost component linking in CMakeLists.txt
f489b55
@KIwabuchi
KIwabuchi requested a review from graceli24 September 17, 2026 04:43
@KIwabuchi

KIwabuchi commented Sep 17, 2026

Copy link
Copy Markdown
Member Author

@graceli24

I made a few changes that are related to your distributed HDBSCAN code.

  1. Removed the --distributed_hdbscan option from run_clams_bench.py and always use the distributed clustering executable by default.
  2. In the clustering output file, the header row starts from '#' : https://github.com/llnl/ClaMS/pull/16/changes#diff-38f0836fd8d9ed5a34e78765ae6accc7cf2f87f955b9ef6a16fbcee2ef3e9fffR1198

Could you check if it's okay for you?

@graceli24

graceli24 commented Sep 17, 2026

Copy link
Copy Markdown
Collaborator

@graceli24

I made a few changes that are related to your distributed HDBSCAN code.

This all looks good to me!

graceli24
graceli24 approved these changes Sep 17, 2026
View reviewed changes

@graceli24 graceli24 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Distributed HDBSCAN changes and use in bench script look good.

@KIwabuchi
KIwabuchi merged commit e375b50 into llnl:develop Sep 17, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@graceli24 graceli24 graceli24 approved these changes

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@KIwabuchi @graceli24