from sklearn_benchmarks.reporting.hp_match import HpMatchReporting
from sklearn_benchmarks.utils import default_run_dir, default_report_config
from pathlib import Path
import pandas as pd
pd.set_option('display.max_colwidth', None)
pd.set_option('display.max_columns', None)
pd.set_option('display.max_rows', None)

scikit-learn-intelex (Intel® oneAPI) vs. scikit-learn

run_dir = default_run_dir()
report_config = default_report_config("scikit_learn_intelex_vs_scikit_learn")

# Parameters
report_config = {
    "parameters": {"n_cols": 2},
    "log_scale": True,
    "estimators": [
        {
            "title": "KNeighborsClassifier (brute force)",
            "name": "KNeighborsClassifier_brute_force",
            "split_bars_by_parameters": ["n_jobs"],
        },
        {
            "title": "KNeighborsClassifier (KD tree)",
            "name": "KNeighborsClassifier_kd_tree",
            "split_bars_by_parameters": ["n_jobs"],
        },
        {"title": "KMeans (tall)", "name": "KMeans_tall"},
        {"title": "KMeans (short)", "name": "KMeans_short"},
        {"title": "Logistic Regression", "name": "LogisticRegression"},
        {"title": "LinearRegression", "name": "LinearRegression"},
        {"title": "Ridge", "name": "Ridge"},
        {"title": "TSNE", "name": "TSNE"},
        {"title": "PCA", "name": "PCA"},
    ],
    "benchmarking_estimator_parameters": {
        "sklearn_KNeighborsClassifier_brute_force": [
            "n_neighbors",
            "algorithm",
            "n_jobs",
        ],
        "sklearnex_KNeighborsClassifier_brute_force": [],
        "sklearn_KNeighborsClassifier_kd_tree": ["n_neighbors", "algorithm", "n_jobs"],
        "sklearnex_KNeighborsClassifier_kd_tree": [],
        "sklearn_KMeans_tall": [
            "init",
            "algorithm",
            "n_clusters",
            "max_iter",
            "n_init",
            "tol",
        ],
        "sklearnex_KMeans_tall": [],
        "sklearn_KMeans_short": [
            "init",
            "algorithm",
            "n_clusters",
            "max_iter",
            "n_init",
            "tol",
        ],
        "sklearnex_KMeans_short": [],
        "sklearn_LogisticRegression": [
            "penalty",
            "dual",
            "tol",
            "C",
            "fit_intercept",
            "intercept_scaling",
            "class_weight",
            "random_state",
            "solver",
            "max_iter",
            "multi_class",
            "verbose",
            "warm_start",
            "n_jobs",
            "l1_ratio",
        ],
        "sklearnex_LogisticRegression": [],
        "sklearn_LinearRegression": [],
        "sklearnex_LinearRegression": [],
        "sklearn_Ridge": ["alpha"],
        "sklearnex_Ridge": [],
        "sklearn_TSNE": [],
        "sklearnex_TSNE": [],
        "sklearn_PCA": ["n_components"],
        "sklearnex_PCA": [],
        "sklearn_HistGradientBoostingClassifier_best": [
            "learning_rate",
            "n_iter_no_change",
            "max_leaf_nodes",
            "max_bins",
            "min_samples_leaf",
            "max_iter",
        ],
        "sklearn_HistGradientBoostingClassifier": [
            "learning_rate",
            "n_iter_no_change",
            "max_leaf_nodes",
            "max_bins",
            "min_samples_leaf",
            "max_iter",
            "early_stopping",
        ],
        "lightgbm": [
            "learning_rate",
            "num_leaves",
            "max_bin",
            "min_data_in_leaf",
            "bagging_fraction",
            "feature_fraction_bynode",
            "n_estimators",
        ],
        "xgboost": [
            "validate_parameters",
            "tree_method",
            "use_label_encoder",
            "learning_rate",
            "max_leaves",
            "max_bin",
            "min_child_weight",
            "subsample",
            "colsample_bynode",
            "n_estimators",
        ],
        "catboost_lossguide": [
            "allow_writing_files",
            "grow_policy",
            "bootstrap_type",
            "learning_rate",
            "max_leaves",
            "border_count",
            "min_data_in_leaf",
            "subsample",
            "rsm",
            "max_depth",
            "iterations",
        ],
        "catboost_symmetric": [
            "allow_writing_files",
            "grow_policy",
            "bootstrap_type",
            "learning_rate",
            "border_count",
            "min_data_in_leaf",
            "subsample",
            "rsm",
            "max_depth",
            "iterations",
        ],
    },
    "other_library": "sklearnex",
}
run_dir = "./results/local/20220322T182455/"

run_dir = Path(run_dir)

reporting = HpMatchReporting(**report_config, run_dir=run_dir)
reporting.make_report()

We assume here there is a perfect match between the hyperparameters of both librairies. For a given set of parameters and a given dataset, we compute the speed-up time scikit-learn / time sklearnex. For instance, a speed-up of 2 means that sklearnex is twice as fast as scikit-learn for a given set of parameters and a given dataset.

KNeighborsClassifier (brute force) ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters: algorithm=brute.

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	100000	100000	100	1.159	0.021	0.069	0.0	-1	1	0.050	0.000	23.325	23.326
3	Download	Download	fit	100000	100000	100	1.102	0.013	0.073	0.0	-1	5	0.044	0.001	25.259	25.261
6	Download	Download	fit	100000	100000	100	1.149	0.026	0.070	0.0	1	100	0.044	0.001	25.848	25.850
9	Download	Download	fit	100000	100000	100	1.153	0.016	0.069	0.0	-1	100	0.050	0.000	23.278	23.278
12	Download	Download	fit	100000	100000	100	1.102	0.011	0.073	0.0	1	5	0.049	0.000	22.297	22.297
15	Download	Download	fit	100000	100000	100	1.110	0.019	0.072	0.0	1	1	0.047	0.003	23.822	23.859
18	Download	Download	fit	100000	100000	2	0.044	0.001	0.036	0.0	-1	1	0.008	0.000	5.280	5.281
21	Download	Download	fit	100000	100000	2	0.044	0.001	0.036	0.0	-1	5	0.008	0.000	5.260	5.261
24	Download	Download	fit	100000	100000	2	0.044	0.001	0.036	0.0	1	100	0.008	0.000	5.269	5.270
27	Download	Download	fit	100000	100000	2	0.044	0.001	0.037	0.0	-1	100	0.008	0.000	5.209	5.210
30	Download	Download	fit	100000	100000	2	0.043	0.001	0.037	0.0	1	5	0.008	0.000	5.188	5.189
33	Download	Download	fit	100000	100000	2	0.043	0.001	0.037	0.0	1	1	0.008	0.000	5.164	5.164

predict

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
1	Download	Download	predict	100000	1000	100	2.243	0.096	0.000	0.002	-1	1	0.166	0.000	13.535	13.535
2	Download	Download	predict	100000	1	100	0.021	0.003	0.000	0.021	-1	1	0.008	0.000	2.647	2.648
4	Download	Download	predict	100000	1000	100	2.754	0.072	0.000	0.003	-1	5	0.167	0.000	16.490	16.490
5	Download	Download	predict	100000	1	100	0.021	0.001	0.000	0.021	-1	5	0.008	0.000	2.715	2.716
7	Download	Download	predict	100000	1000	100	1.859	0.003	0.000	0.002	1	100	0.205	0.001	9.083	9.083
8	Download	Download	predict	100000	1	100	0.018	0.000	0.000	0.018	1	100	0.008	0.000	2.299	2.300
10	Download	Download	predict	100000	1000	100	2.561	0.029	0.000	0.003	-1	100	0.205	0.001	12.503	12.503
11	Download	Download	predict	100000	1	100	0.023	0.003	0.000	0.023	-1	100	0.008	0.000	2.848	2.849
13	Download	Download	predict	100000	1000	100	1.862	0.033	0.000	0.002	1	5	0.167	0.001	11.127	11.127
14	Download	Download	predict	100000	1	100	0.018	0.000	0.000	0.018	1	5	0.008	0.000	2.263	2.263
16	Download	Download	predict	100000	1000	100	1.143	0.006	0.001	0.001	1	1	0.166	0.000	6.894	6.894
17	Download	Download	predict	100000	1	100	0.017	0.000	0.000	0.017	1	1	0.008	0.000	2.218	2.218
19	Download	Download	predict	100000	1000	2	1.682	0.020	0.000	0.002	-1	1	0.026	0.001	65.129	65.169
20	Download	Download	predict	100000	1	2	0.004	0.002	0.000	0.004	-1	1	0.001	0.000	4.817	4.985
22	Download	Download	predict	100000	1000	2	2.435	0.025	0.000	0.002	-1	5	0.027	0.000	91.643	91.644
23	Download	Download	predict	100000	1	2	0.006	0.003	0.000	0.006	-1	5	0.001	0.000	7.412	7.442
25	Download	Download	predict	100000	1000	2	1.804	0.002	0.000	0.002	1	100	0.060	0.000	30.184	30.184
26	Download	Download	predict	100000	1	2	0.003	0.000	0.000	0.003	1	100	0.001	0.000	3.171	3.193
28	Download	Download	predict	100000	1000	2	2.475	0.020	0.000	0.002	-1	100	0.060	0.000	41.410	41.410
29	Download	Download	predict	100000	1	2	0.005	0.002	0.000	0.005	-1	100	0.001	0.000	6.594	6.611
31	Download	Download	predict	100000	1000	2	1.800	0.005	0.000	0.002	1	5	0.027	0.000	67.629	67.629
32	Download	Download	predict	100000	1	2	0.003	0.000	0.000	0.003	1	5	0.001	0.000	3.452	3.477
34	Download	Download	predict	100000	1000	2	1.042	0.007	0.000	0.001	1	1	0.026	0.000	40.842	40.843
35	Download	Download	predict	100000	1	2	0.002	0.000	0.000	0.002	1	1	0.001	0.000	2.321	2.325

Profiling traces can be visualized using Perfetto UI.

KNeighborsClassifier (KD tree) ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters: algorithm=kd_tree.

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	1000000	1000000	10	2.573	0.011	0.031	0.0	-1	1	0.659	0.003	3.904	3.904
3	Download	Download	fit	1000000	1000000	10	2.561	0.016	0.031	0.0	-1	5	0.669	0.014	3.829	3.830
6	Download	Download	fit	1000000	1000000	10	2.569	0.007	0.031	0.0	1	100	0.662	0.005	3.882	3.882
9	Download	Download	fit	1000000	1000000	10	2.588	0.020	0.031	0.0	-1	100	0.668	0.009	3.873	3.873
12	Download	Download	fit	1000000	1000000	10	2.576	0.016	0.031	0.0	1	5	0.677	0.016	3.807	3.808
15	Download	Download	fit	1000000	1000000	10	2.561	0.012	0.031	0.0	1	1	0.672	0.012	3.813	3.814
18	Download	Download	fit	1000	1000	2	0.001	0.000	0.030	0.0	-1	1	0.001	0.000	0.569	0.574
21	Download	Download	fit	1000	1000	2	0.001	0.000	0.031	0.0	-1	5	0.001	0.000	0.578	0.583
24	Download	Download	fit	1000	1000	2	0.001	0.000	0.031	0.0	1	100	0.001	0.000	0.568	0.573
27	Download	Download	fit	1000	1000	2	0.001	0.000	0.031	0.0	-1	100	0.001	0.000	0.583	0.585
30	Download	Download	fit	1000	1000	2	0.001	0.000	0.030	0.0	1	5	0.001	0.000	0.579	0.584
33	Download	Download	fit	1000	1000	2	0.001	0.000	0.031	0.0	1	1	0.001	0.000	0.578	0.583

predict

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
1	Download	Download	predict	1000000	1000	10	0.378	0.008	0.000	0.000	-1	1	0.090	0.001	4.185	4.185
2	Download	Download	predict	1000000	1	10	0.003	0.000	0.000	0.003	-1	1	0.000	0.000	9.663	10.326
4	Download	Download	predict	1000000	1000	10	0.686	0.009	0.000	0.001	-1	5	0.165	0.003	4.150	4.151
5	Download	Download	predict	1000000	1	10	0.003	0.000	0.000	0.003	-1	5	0.000	0.000	8.988	9.526
7	Download	Download	predict	1000000	1000	10	4.093	0.012	0.000	0.004	1	100	0.484	0.003	8.457	8.457
8	Download	Download	predict	1000000	1	10	0.002	0.001	0.000	0.002	1	100	0.001	0.000	4.227	4.424
10	Download	Download	predict	1000000	1000	10	2.438	0.009	0.000	0.002	-1	100	0.493	0.004	4.950	4.950
11	Download	Download	predict	1000000	1	10	0.004	0.001	0.000	0.004	-1	100	0.001	0.000	7.220	7.634
13	Download	Download	predict	1000000	1000	10	1.237	0.007	0.000	0.001	1	5	0.164	0.001	7.546	7.546
14	Download	Download	predict	1000000	1	10	0.001	0.000	0.000	0.001	1	5	0.000	0.000	3.750	4.009
16	Download	Download	predict	1000000	1000	10	0.661	0.006	0.000	0.001	1	1	0.092	0.001	7.217	7.217
17	Download	Download	predict	1000000	1	10	0.001	0.000	0.000	0.001	1	1	0.000	0.000	3.666	3.902
19	Download	Download	predict	1000	1000	2	0.022	0.000	0.001	0.000	-1	1	0.001	0.000	42.753	42.944
20	Download	Download	predict	1000	1	2	0.002	0.000	0.000	0.002	-1	1	0.000	0.000	17.202	17.864
22	Download	Download	predict	1000	1000	2	0.023	0.000	0.001	0.000	-1	5	0.001	0.000	31.008	31.390
23	Download	Download	predict	1000	1	2	0.002	0.000	0.000	0.002	-1	5	0.000	0.000	17.071	17.908
25	Download	Download	predict	1000	1000	2	0.037	0.011	0.000	0.000	1	100	0.005	0.000	8.039	8.056
26	Download	Download	predict	1000	1	2	0.001	0.000	0.000	0.001	1	100	0.000	0.000	3.854	4.367
28	Download	Download	predict	1000	1000	2	0.034	0.000	0.000	0.000	-1	100	0.005	0.000	7.391	7.401
29	Download	Download	predict	1000	1	2	0.002	0.000	0.000	0.002	-1	100	0.000	0.000	15.388	16.688
31	Download	Download	predict	1000	1000	2	0.021	0.000	0.001	0.000	1	5	0.001	0.000	25.476	25.732
32	Download	Download	predict	1000	1	2	0.001	0.000	0.000	0.001	1	5	0.000	0.000	4.338	4.582
34	Download	Download	predict	1000	1000	2	0.019	0.000	0.001	0.000	1	1	0.001	0.000	37.930	38.228
35	Download	Download	predict	1000	1	2	0.001	0.000	0.000	0.001	1	1	0.000	0.000	4.301	4.561

Profiling traces can be visualized using Perfetto UI.

KMeans (tall) ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters: algorithm=full, n_clusters=3, max_iter=30, n_init=1, tol=1e-16.

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_iter	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	1000000	1000000	2	0.473	0.002	0.034	0.0	30	random	0.350	0.017	1.350	1.352
3	Download	Download	fit	1000000	1000000	2	0.528	0.007	0.030	0.0	30	k-means++	0.391	0.012	1.351	1.351
6	Download	Download	fit	1000000	1000000	100	4.495	0.262	0.178	0.0	30	random	2.431	0.020	1.849	1.850
9	Download	Download	fit	1000000	1000000	100	4.595	0.034	0.174	0.0	30	k-means++	2.627	0.006	1.749	1.749

predict

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_iter	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
1	Download	Download	predict	1000000	1000	2	0.0	0.0	0.073	0.0	30	random	0.0	0.0	1.127	1.189
2	Download	Download	predict	1000000	1	2	0.0	0.0	0.000	0.0	30	random	0.0	0.0	1.317	1.375
4	Download	Download	predict	1000000	1000	2	0.0	0.0	0.072	0.0	30	k-means++	0.0	0.0	1.102	1.175
5	Download	Download	predict	1000000	1	2	0.0	0.0	0.000	0.0	30	k-means++	0.0	0.0	1.151	1.286
7	Download	Download	predict	1000000	1000	100	0.0	0.0	1.936	0.0	30	random	0.0	0.0	1.338	1.402
8	Download	Download	predict	1000000	1	100	0.0	0.0	0.004	0.0	30	random	0.0	0.0	1.258	1.341
10	Download	Download	predict	1000000	1000	100	0.0	0.0	1.982	0.0	30	k-means++	0.0	0.0	1.365	1.433
11	Download	Download	predict	1000000	1	100	0.0	0.0	0.004	0.0	30	k-means++	0.0	0.0	1.278	1.334

Profiling traces can be visualized using Perfetto UI.

KMeans (short) ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters: algorithm=full, n_clusters=300, max_iter=20, n_init=1, tol=1e-16.

Mismatches between validation scores

WARNING! Mismatch between validation scores for 2 predictions (25.0%).
The observed differences can be found in the diff_adjusted_rand_scores column.
The chosen difference threshold is 0.01 for adjusted_rand_score.
The biggest difference is 0.077 for adjusted_rand_score.
See details in the dataframe below.

	estimator	library	diff_adjusted_rand_scores	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_iter	algorithm	init	max_iter	n_clusters	n_init	tol	adjusted_rand_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	adjusted_rand_score_sklearnex	speedup	std_speedup
7	sklearn_KMeans_short	sklearn	0.074768	predict	10000	1000	100	0.001315	0.000210	0.608421	0.000001	20	full	random	20	300	1	1.000000e-16	0.222701	0.001131	0.000108	0.297469	1.162543	1.167836
10	sklearn_KMeans_short	sklearn	0.076616	predict	10000	1000	100	0.001232	0.000114	0.649283	0.000001	20	full	k-means++	20	300	1	1.000000e-16	0.408030	0.001115	0.000098	0.331414	1.104843	1.109090

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_iter	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	10000	10000	2	0.077	0.000	0.002	0.0	20	random	0.031	0.000	2.495	2.495
3	Download	Download	fit	10000	10000	2	0.193	0.001	0.001	0.0	20	k-means++	0.086	0.002	2.248	2.249
6	Download	Download	fit	10000	10000	100	0.193	0.010	0.042	0.0	20	random	0.125	0.002	1.536	1.536
9	Download	Download	fit	10000	10000	100	0.537	0.006	0.015	0.0	20	k-means++	0.308	0.001	1.741	1.741

predict

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_iter	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
1	Download	Download	predict	10000	1000	2	0.001	0.0	0.024	0.0	20	random	0.001	0.0	1.070	1.071
2	Download	Download	predict	10000	1	2	0.000	0.0	0.000	0.0	20	random	0.000	0.0	1.191	1.255
4	Download	Download	predict	10000	1000	2	0.001	0.0	0.025	0.0	20	k-means++	0.001	0.0	1.040	1.043
5	Download	Download	predict	10000	1	2	0.000	0.0	0.000	0.0	20	k-means++	0.000	0.0	1.339	1.392
7	Download	Download	predict	10000	1000	100	0.001	0.0	0.608	0.0	20	random	0.001	0.0	1.163	1.168
8	Download	Download	predict	10000	1	100	0.000	0.0	0.003	0.0	20	random	0.000	0.0	1.264	1.303
10	Download	Download	predict	10000	1000	100	0.001	0.0	0.649	0.0	20	k-means++	0.001	0.0	1.105	1.109
11	Download	Download	predict	10000	1	100	0.000	0.0	0.003	0.0	20	k-means++	0.000	0.0	1.407	1.450

Profiling traces can be visualized using Perfetto UI.

Logistic Regression ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters: penalty=l2, dual=False, tol=0.0001, C=1.0, fit_intercept=True, intercept_scaling=1.0, class_weight=nan, random_state=nan, solver=lbfgs, max_iter=100, multi_class=auto, verbose=0, warm_start=False, n_jobs=nan, l1_ratio=nan.

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_iter	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	1000000	1000000	100	9.720	0.021	0.082	0.000	[20]	1.677	0.004	5.797	5.797
3	Download	Download	fit	1000	1000	10000	0.701	0.042	0.114	0.001	[27]	0.676	0.056	1.037	1.040

predict

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_iter	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
1	Download	Download	predict	1000000	1000	100	0.000	0.0	3.051	0.0	[20]	0.000	0.0	0.710	0.737
2	Download	Download	predict	1000000	1	100	0.000	0.0	0.014	0.0	[20]	0.000	0.0	0.260	0.284
4	Download	Download	predict	1000	100	10000	0.002	0.0	5.204	0.0	[27]	0.003	0.0	0.543	0.547
5	Download	Download	predict	1000	1	10000	0.000	0.0	0.989	0.0	[27]	0.001	0.0	0.111	0.115

Profiling traces can be visualized using Perfetto UI.

LinearRegression ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters:

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	copy_X	fit_intercept	normalize	positive	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	1000	1000	10000	1.230	0.011	0.065	0.001	True	True	deprecated	False	1.305	0.013	0.943	0.943
3	Download	Download	fit	1000000	1000000	100	7.166	0.110	0.112	0.000	True	True	deprecated	False	0.177	0.000	40.421	40.421

predict

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	copy_X	fit_intercept	normalize	positive	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
1	Download	Download	predict	1000	1000	10000	0.01	0.0	8.254	0.0	True	True	deprecated	False	0.016	0.0	0.589	0.589
2	Download	Download	predict	1000	10	10000	0.00	0.0	6.442	0.0	True	True	deprecated	False	0.000	0.0	0.431	0.450
4	Download	Download	predict	1000000	1000	100	0.00	0.0	5.713	0.0	True	True	deprecated	False	0.000	0.0	0.358	0.368
5	Download	Download	predict	1000000	10	100	0.00	0.0	0.145	0.0	True	True	deprecated	False	0.000	0.0	0.377	0.394

Profiling traces can be visualized using Perfetto UI.

Ridge ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters: alpha=1e-06.

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	1000	1000	10000	0.159	0.000	0.505	0.0	0.167	0.001	0.951	0.951
3	Download	Download	fit	1000000	1000000	100	0.992	0.004	0.806	0.0	0.210	0.001	4.736	4.736

predict

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
1	Download	Download	predict	1000	1000	10000	0.01	0.0	8.283	0.0	0.016	0.0	0.590	0.590
2	Download	Download	predict	1000	10	10000	0.00	0.0	6.528	0.0	0.000	0.0	0.437	0.454
4	Download	Download	predict	1000000	1000	100	0.00	0.0	6.029	0.0	0.000	0.0	0.477	0.495
5	Download	Download	predict	1000000	10	100	0.00	0.0	0.151	0.0	0.000	0.0	0.370	0.381

Profiling traces can be visualized using Perfetto UI.

TSNE ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters:

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_iter	angle	early_exaggeration	init	learning_rate	method	metric	min_grad_norm	n_components	n_iter_without_progress	perplexity	square_distances	verbose	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	1000	1000	3	3.439	0.167	0.0	0.003	1000	0.5	12.0	warn	warn	barnes_hut	euclidean	0.0	2	300	30.0	legacy	0	3.514	0.23	0.979	0.981

Profiling traces can be visualized using Perfetto UI.

PCA ¶

scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)

Speedup barplots ¶

All estimators share the following parameters: n_components=10.

Raw results ¶

fit

	sklearn_profiling	sklearnex_profiling	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup
0	Download	Download	fit	100000	100000	100	0.411	0.001	0.195	0.0	0.025	0.000	16.729	16.730
1	Download	Download	fit	10000	10000	1000	0.360	0.002	0.222	0.0	0.176	0.001	2.042	2.042

Profiling traces can be visualized using Perfetto UI.

Benchmark environment information

System

python	3.8.12 | packaged by conda-forge | (default, Jan 30 2022, 23:42:07) [GCC 9.4.0]
executable	/usr/share/miniconda/envs/sklbench/bin/python
machine	Linux-5.11.0-1028-azure-x86_64-with-glibc2.10

Dependencies

pip	22.0.4
setuptools	60.10.0
sklearn	1.0.2
numpy	1.22.3
scipy	1.8.0
Cython	None
pandas	1.4.1
matplotlib	3.5.1
joblib	1.1.0
threadpoolctl	3.1.0

Threadpool

	user_api	internal_api	prefix	filepath	version	threading_layer	architecture	num_threads
0	blas	openblas	libopenblas	/usr/share/miniconda/envs/sklbench/lib/libopenblasp-r0.3.18.so	0.3.18	pthreads	SkylakeX	2
1	openmp	openmp	libgomp	/usr/share/miniconda/envs/sklbench/lib/libgomp.so.1.0.0	None	NaN	NaN	2

Cpu_count

cpu_count	2
physical_cpu_count	2