Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Improve vector leaf performance #311

Merged
merged 5 commits into from
Sep 17, 2021
Merged

Conversation

RAMitchell
Copy link
Member

On a large random forest problem, this improves Treelite model generation time from ~650s to 1s.

include/treelite/tree_impl.h Outdated Show resolved Hide resolved
include/treelite/tree_impl.h Show resolved Hide resolved
@hcho3
Copy link
Collaborator

hcho3 commented Sep 16, 2021

There is some problem with Conda environment. Let me sort that out.

@codecov
Copy link

codecov bot commented Sep 17, 2021

Codecov Report

Merging #311 (54694c1) into mainline (193445d) will decrease coverage by 0.90%.
The diff coverage is 100.00%.

Impacted file tree graph

@@              Coverage Diff               @@
##             mainline     #311      +/-   ##
==============================================
- Coverage       85.20%   84.29%   -0.91%     
  Complexity         46       46              
==============================================
  Files             108      108              
  Lines            8157     8126      -31     
  Branches          466      466              
==============================================
- Hits             6950     6850     -100     
- Misses           1182     1251      +69     
  Partials           25       25              
Impacted Files Coverage Δ
include/treelite/tree.h 97.77% <100.00%> (ø)
include/treelite/tree_impl.h 90.63% <100.00%> (+0.59%) ⬆️
src/json_serializer.cc 92.17% <100.00%> (ø)
python/treelite/contrib/msvc.py 0.00% <0.00%> (-67.15%) ⬇️
python/treelite/libpath.py 67.85% <0.00%> (-14.29%) ⬇️
runtime/python/treelite_runtime/libpath.py 67.85% <0.00%> (-14.29%) ⬇️
python/treelite/contrib/__init__.py 69.67% <0.00%> (-5.74%) ⬇️
python/treelite/contrib/util.py 86.17% <0.00%> (-4.26%) ⬇️
python/treelite/core.py 78.57% <0.00%> (-3.58%) ⬇️
runtime/python/treelite_runtime/util.py 92.10% <0.00%> (-2.64%) ⬇️
... and 14 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 193445d...54694c1. Read the comment docs.

@hcho3
Copy link
Collaborator

hcho3 commented Sep 17, 2021

@RAMitchell I've fixed the CI. We should be able to merge this PR soon.

@hcho3 hcho3 merged commit 1147cea into dmlc:mainline Sep 17, 2021
This was referenced Sep 20, 2021
rapids-bot bot pushed a commit to rapidsai/cuml that referenced this pull request Sep 23, 2021
The 2.1.0 version of Treelite incorporates the following major improvements:

* dmlc/treelite#311
* dmlc/treelite#302
* dmlc/treelite#303
* dmlc/treelite#296

In particular, dmlc/treelite#311 is a critical follow-up to #4191 and addresses a performance regression.

Requires rapidsai/integration#353

Authors:
  - Philip Hyunsu Cho (https://github.com/hcho3)

Approvers:
  - Jordan Jacobelli (https://github.com/Ethyling)
  - Dante Gama Dessavre (https://github.com/dantegd)

URL: #4220
vimarsh6739 pushed a commit to vimarsh6739/cuml that referenced this pull request Oct 9, 2023
The 2.1.0 version of Treelite incorporates the following major improvements:

* dmlc/treelite#311
* dmlc/treelite#302
* dmlc/treelite#303
* dmlc/treelite#296

In particular, dmlc/treelite#311 is a critical follow-up to rapidsai#4191 and addresses a performance regression.

Requires rapidsai/integration#353

Authors:
  - Philip Hyunsu Cho (https://github.com/hcho3)

Approvers:
  - Jordan Jacobelli (https://github.com/Ethyling)
  - Dante Gama Dessavre (https://github.com/dantegd)

URL: rapidsai#4220
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants