From 42d03873b01983ea327c437510b70bbd955f9e84 Mon Sep 17 00:00:00 2001 From: David Cortes Date: Wed, 14 Jan 2026 16:11:53 +0100 Subject: [PATCH 1/6] add CV version of logistic regression --- configs/regular/logreg_cv.json | 35 +++++++++++++++++++++ envs/conda-env-sklearn.yml | 2 +- envs/requirements-sklearn.txt | 2 +- sklbench/benchmarks/estimator_task_map.json | 1 + 4 files changed, 38 insertions(+), 2 deletions(-) create mode 100644 configs/regular/logreg_cv.json diff --git a/configs/regular/logreg_cv.json b/configs/regular/logreg_cv.json new file mode 100644 index 00000000..51cb257b --- /dev/null +++ b/configs/regular/logreg_cv.json @@ -0,0 +1,35 @@ +{ + "PARAMETERS_SETS": { + "common logreg_cv parameters": { + "algorithm": { + "estimator": "LogisticRegressionCV", + "estimator_methods": { "inference": "predict" }, + "estimator_params": { + "Cs": 10, + "l1_ratios": [0.0], + "solver": "newton-cg", + "random_state": 123 + } + } + }, + "logreg_cv datasets": [ + { + "data": { + "source": "make_classification", + "generation_kwargs": [ + { + "n_samples": 250000, + "n_features": 1000, + "n_informative": 800, + "n_classes": [2] + } + ], + "split_kwargs": { + "train_size": 0.05, + "test_size": 0.95 + } + } + } + ] + } +} diff --git a/envs/conda-env-sklearn.yml b/envs/conda-env-sklearn.yml index d8f0571f..3dcbabc6 100644 --- a/envs/conda-env-sklearn.yml +++ b/envs/conda-env-sklearn.yml @@ -10,7 +10,7 @@ dependencies: - modin-all - scikit-learn-intelex # sklbench dependencies - - scikit-learn + - scikit-learn>=1.8 - pandas - tabulate - fastparquet diff --git a/envs/requirements-sklearn.txt b/envs/requirements-sklearn.txt index 064c9e90..6067dce2 100644 --- a/envs/requirements-sklearn.txt +++ b/envs/requirements-sklearn.txt @@ -8,7 +8,7 @@ scikit-learn-intelex dpctl dpnp # sklbench dependencies -scikit-learn +scikit-learn>=1.8 pandas tabulate fastparquet diff --git a/sklbench/benchmarks/estimator_task_map.json b/sklbench/benchmarks/estimator_task_map.json index 09eb59b5..d5a120f2 100644 --- a/sklbench/benchmarks/estimator_task_map.json +++ b/sklbench/benchmarks/estimator_task_map.json @@ -2,6 +2,7 @@ "classification": [ "Classifier", "LogisticRegression", + "LogisticRegressionCV", "SVC" ], "regression": [ From 9007a51e301880bb4d39f1163611f89b9197bbdd Mon Sep 17 00:00:00 2001 From: David Cortes Date: Wed, 14 Jan 2026 16:21:33 +0100 Subject: [PATCH 2/6] fix --- configs/regular/logreg_cv.json | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/configs/regular/logreg_cv.json b/configs/regular/logreg_cv.json index 51cb257b..41111d8b 100644 --- a/configs/regular/logreg_cv.json +++ b/configs/regular/logreg_cv.json @@ -1,6 +1,6 @@ { "PARAMETERS_SETS": { - "common logreg_cv parameters": { + "sklearn logreg_cv parameters": { "algorithm": { "estimator": "LogisticRegressionCV", "estimator_methods": { "inference": "predict" }, @@ -31,5 +31,14 @@ } } ] + }, + "TEMPLATES": { + "sklearn logreg_cv": { + "SETS": [ + "sklearn-ex[cpu] implementations", + "sklearn logreg_cv parameters", + "logreg_cv datasets" + ] + } } } From b31186866b3978168ed84eea176868463e78d689 Mon Sep 17 00:00:00 2001 From: David Cortes Date: Wed, 14 Jan 2026 16:23:08 +0100 Subject: [PATCH 3/6] fix --- configs/regular/logreg_cv.json | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configs/regular/logreg_cv.json b/configs/regular/logreg_cv.json index 41111d8b..0b54726e 100644 --- a/configs/regular/logreg_cv.json +++ b/configs/regular/logreg_cv.json @@ -35,7 +35,7 @@ "TEMPLATES": { "sklearn logreg_cv": { "SETS": [ - "sklearn-ex[cpu] implementations", + "sklearn-ex[cpu,gpu] implementations", "sklearn logreg_cv parameters", "logreg_cv datasets" ] From 0f6d2cad31355006260f5d4114ca75a29ebbd70f Mon Sep 17 00:00:00 2001 From: david-cortes-intel Date: Wed, 14 Jan 2026 07:34:47 -0800 Subject: [PATCH 4/6] more fixes --- configs/regular/logreg_cv.json | 3 +++ 1 file changed, 3 insertions(+) diff --git a/configs/regular/logreg_cv.json b/configs/regular/logreg_cv.json index 0b54726e..f8d00f46 100644 --- a/configs/regular/logreg_cv.json +++ b/configs/regular/logreg_cv.json @@ -1,8 +1,11 @@ { + "INCLUDE": ["../common/sklearn.json"], "PARAMETERS_SETS": { "sklearn logreg_cv parameters": { "algorithm": { "estimator": "LogisticRegressionCV", + "library": ["sklearn", "sklearnex"], + "device": "cpu", "estimator_methods": { "inference": "predict" }, "estimator_params": { "Cs": 10, From 8c33a6081d9f25f4e307bee0ed66e01b90544cc8 Mon Sep 17 00:00:00 2001 From: david-cortes-intel Date: Wed, 14 Jan 2026 07:38:51 -0800 Subject: [PATCH 5/6] yet more fixes --- configs/regular/logreg_cv.json | 1 - 1 file changed, 1 deletion(-) diff --git a/configs/regular/logreg_cv.json b/configs/regular/logreg_cv.json index f8d00f46..ae0ee07f 100644 --- a/configs/regular/logreg_cv.json +++ b/configs/regular/logreg_cv.json @@ -9,7 +9,6 @@ "estimator_methods": { "inference": "predict" }, "estimator_params": { "Cs": 10, - "l1_ratios": [0.0], "solver": "newton-cg", "random_state": 123 } From 7839432265b0ca6d83a0b9e584245d2f967f1f89 Mon Sep 17 00:00:00 2001 From: david-cortes-intel Date: Wed, 14 Jan 2026 07:53:46 -0800 Subject: [PATCH 6/6] change params --- configs/regular/logreg_cv.json | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/configs/regular/logreg_cv.json b/configs/regular/logreg_cv.json index ae0ee07f..ec7a6a6d 100644 --- a/configs/regular/logreg_cv.json +++ b/configs/regular/logreg_cv.json @@ -9,7 +9,8 @@ "estimator_methods": { "inference": "predict" }, "estimator_params": { "Cs": 10, - "solver": "newton-cg", + "solver": "lbfgs", + "max_iter": 1000, "random_state": 123 } } @@ -20,7 +21,7 @@ "source": "make_classification", "generation_kwargs": [ { - "n_samples": 250000, + "n_samples": 100000, "n_features": 1000, "n_informative": 800, "n_classes": [2]