HGB speed-up breakdown (branch comparison)

Hardware

CPU

Intel(R) Xeon(R) 6787P

X86_64, 172 physical cores, 344 logical CPUs

499 GB RAM

GPU(s)

No GPU detected.

sklearn-dev@cakedev0:main

Python 3.12.13

Packages

scikit-learn @ 6aa2f57c2
numpy 2.5.0 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.33 (128 threads)
libgomp (344 threads)

OpenMP

GNU libgomp (OpenMP 4.5)
GOMP_SPINCOUNT: 300000

Full environment

view pixi env JSON

sklearn-dev@cakedev0:hgb/adapt_to_active_wait

Python 3.12.13

Packages

scikit-learn @ 27d3e13ba
numpy 2.5.0 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.33 (128 threads)
libgomp (344 threads)

OpenMP

GNU libgomp (OpenMP 4.5)
GOMP_SPINCOUNT: 300000

Full environment

view pixi env JSON

sklearn-dev@cakedev0:hgb/both_threads_optim

Python 3.12.13

Packages

scikit-learn @ 44f1cbbde
numpy 2.5.0 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.33 (128 threads)
libgomp (344 threads)

OpenMP

GNU libgomp (OpenMP 4.5)
GOMP_SPINCOUNT: 300000

Full environment

view pixi env JSON

1 threads

2 threads

4 threads

8 threads

16 threads

32 threads

64 threads

128 threads

Hardware

CPU

Intel(R) Xeon(R) 6787P

X86_64, 172 physical cores, 344 logical CPUs

499 GB RAM

GPU(s)

No GPU detected.

sklearn-dev-libomp@cakedev0:main

Python 3.12.13

Packages

scikit-learn @ 6aa2f57c2
numpy 2.5.2 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.34 (128 threads)
libomp (344 threads)

OpenMP

Intel/LLVM OpenMP (spec 201611)
KMP_BLOCKTIME: 200ms

Full environment

view pixi env JSON

sklearn-dev-libomp@cakedev0:hgb/adapt_to_active_wait

Python 3.12.13

Packages

scikit-learn @ 27d3e13ba
numpy 2.5.2 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.34 (128 threads)
libomp (344 threads)

OpenMP

Intel/LLVM OpenMP (spec 201611)
KMP_BLOCKTIME: 200ms

Full environment

view pixi env JSON

1 threads

2 threads

4 threads

8 threads

16 threads

32 threads

64 threads

128 threads

Hardware

CPU

Intel(R) Core(TM) Ultra X7 358H

X86_64, 16 physical cores, 16 logical CPUs

31 GB RAM

GPU(s)

level_zero:gpu:0: Intel(R) Arc(TM) B390 GPU (29 GB)

sklearn-dev@cakedev0:main

Python 3.12.13

Packages

scikit-learn @ 6aa2f57c2
numpy 2.5.0 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.33 (16 threads)
libgomp (16 threads)

OpenMP

GNU libgomp (OpenMP 4.5)
GOMP_SPINCOUNT: 300000
KMP_BLOCKTIME: 200ms

Full environment

view pixi env JSON

sklearn-dev@cakedev0:hgb/adapt_to_active_wait

Python 3.12.13

Packages

scikit-learn @ 27d3e13ba
numpy 2.5.0 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.33 (16 threads)
libgomp (16 threads)

OpenMP

GNU libgomp (OpenMP 4.5)
GOMP_SPINCOUNT: 300000
KMP_BLOCKTIME: 200ms

Full environment

view pixi env JSON

1 threads

2 threads

4 threads

8 threads

16 threads

Hardware

CPU

Intel(R) Core(TM) Ultra X7 358H

X86_64, 16 physical cores, 16 logical CPUs

31 GB RAM

GPU(s)

level_zero:gpu:0: Intel(R) Arc(TM) B390 GPU (29 GB)

sklearn-dev@cakedev0:main

Python 3.12.13

Packages

scikit-learn @ 6aa2f57c2
numpy 2.5.0 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.33 (16 threads)
libgomp (16 threads)

OpenMP

GNU libgomp (OpenMP 4.5)
GOMP_SPINCOUNT: 1

Full environment

view pixi env JSON

sklearn-dev@cakedev0:hgb/adapt_to_active_wait

Python 3.12.13

Packages

scikit-learn @ 27d3e13ba
numpy 2.5.0 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.33 (16 threads)
libgomp (16 threads)

OpenMP

GNU libgomp (OpenMP 4.5)
GOMP_SPINCOUNT: 1

Full environment

view pixi env JSON

1 threads

2 threads

4 threads

8 threads

16 threads

Hardware

CPU

Intel(R) Core(TM) Ultra X7 358H

X86_64, 16 physical cores, 16 logical CPUs

31 GB RAM

GPU(s)

level_zero:gpu:0: Intel(R) Arc(TM) B390 GPU (29 GB)

sklearn-dev-libomp@cakedev0:main

Python 3.12.13

Packages

scikit-learn @ 6aa2f57c2
numpy 2.5.2 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.34 (16 threads)
libomp (16 threads)

OpenMP

Intel/LLVM OpenMP (spec 201611)
GOMP_SPINCOUNT: 300000
KMP_BLOCKTIME: 200ms

Full environment

view pixi env JSON

sklearn-dev-libomp@cakedev0:hgb/adapt_to_active_wait

Python 3.12.13

Packages

scikit-learn @ 27d3e13ba
numpy 2.5.2 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.34 (16 threads)
libomp (16 threads)

OpenMP

Intel/LLVM OpenMP (spec 201611)
GOMP_SPINCOUNT: 300000
KMP_BLOCKTIME: 200ms

Full environment

view pixi env JSON

1 threads

2 threads

4 threads

8 threads

16 threads

Hardware

CPU

Intel(R) Core(TM) Ultra X7 358H

X86_64, 16 physical cores, 16 logical CPUs

31 GB RAM

GPU(s)

level_zero:gpu:0: Intel(R) Arc(TM) B390 GPU (29 GB)

sklearn-dev-libomp@cakedev0:main

Python 3.12.13

Packages

scikit-learn @ 6aa2f57c2
numpy 2.5.2 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.34 (16 threads)
libomp (16 threads)

OpenMP

Intel/LLVM OpenMP (spec 201611)
KMP_BLOCKTIME: 0ms

Full environment

view pixi env JSON

sklearn-dev-libomp@cakedev0:hgb/adapt_to_active_wait

Python 3.12.13

Packages

scikit-learn @ 27d3e13ba
numpy 2.5.2 (conda-forge)
scipy 1.18.0 (conda-forge)
pandas 2.3.3 (conda-forge)

Threadpools

libopenblas 0.3.34 (16 threads)
libomp (16 threads)

OpenMP

Intel/LLVM OpenMP (spec 201611)
KMP_BLOCKTIME: 0ms

Full environment

view pixi env JSON