from sklearn_benchmarks.reporting.hp_match import HpMatchReporting
from sklearn_benchmarks.utils import default_results_directory
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)
results_dir = default_results_directory()
# Parameters
results_dir = "./results/local/20220316T120802/"
results_dir = Path(results_dir)
reporting = HpMatchReporting(other_library="sklearnex", config="config.yml", log_scale=True, results_dir=results_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.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters: algorithm=brute.
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 100000 | 100000 | 100 | 1.187 | 0.017 | 0.067 | 0.0 | -1 | 1 | 0.049 | 0.005 | 24.448 | 24.571 | Download | Download |
| 3 | fit | 100000 | 100000 | 100 | 1.152 | 0.025 | 0.069 | 0.0 | -1 | 5 | 0.047 | 0.000 | 24.474 | 24.474 | Download | Download |
| 6 | fit | 100000 | 100000 | 100 | 1.160 | 0.019 | 0.069 | 0.0 | 1 | 100 | 0.047 | 0.000 | 24.840 | 24.840 | Download | Download |
| 9 | fit | 100000 | 100000 | 100 | 1.185 | 0.056 | 0.068 | 0.0 | -1 | 100 | 0.047 | 0.000 | 25.152 | 25.152 | Download | Download |
| 12 | fit | 100000 | 100000 | 100 | 1.161 | 0.014 | 0.069 | 0.0 | 1 | 5 | 0.047 | 0.000 | 24.720 | 24.720 | Download | Download |
| 15 | fit | 100000 | 100000 | 100 | 1.133 | 0.015 | 0.071 | 0.0 | 1 | 1 | 0.047 | 0.000 | 24.211 | 24.211 | Download | Download |
| 18 | fit | 100000 | 100000 | 2 | 0.047 | 0.001 | 0.034 | 0.0 | -1 | 1 | 0.009 | 0.000 | 5.044 | 5.044 | Download | Download |
| 21 | fit | 100000 | 100000 | 2 | 0.046 | 0.000 | 0.035 | 0.0 | -1 | 5 | 0.009 | 0.000 | 5.044 | 5.045 | Download | Download |
| 24 | fit | 100000 | 100000 | 2 | 0.046 | 0.000 | 0.035 | 0.0 | 1 | 100 | 0.009 | 0.000 | 4.927 | 4.927 | Download | Download |
| 27 | fit | 100000 | 100000 | 2 | 0.046 | 0.000 | 0.035 | 0.0 | -1 | 100 | 0.009 | 0.000 | 5.011 | 5.011 | Download | Download |
| 30 | fit | 100000 | 100000 | 2 | 0.051 | 0.000 | 0.031 | 0.0 | 1 | 5 | 0.009 | 0.000 | 5.550 | 5.550 | Download | Download |
| 33 | fit | 100000 | 100000 | 2 | 0.052 | 0.000 | 0.031 | 0.0 | 1 | 1 | 0.009 | 0.000 | 5.593 | 5.593 | Download | Download |
predict
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | predict | 100000 | 1000 | 100 | 2.773 | 0.379 | 0.000 | 0.003 | -1 | 1 | 0.189 | 0.008 | 14.656 | 14.669 | Download | Download |
| 2 | predict | 100000 | 1 | 100 | 0.022 | 0.002 | 0.000 | 0.022 | -1 | 1 | 0.009 | 0.000 | 2.436 | 2.437 | Download | Download |
| 4 | predict | 100000 | 1000 | 100 | 3.143 | 0.073 | 0.000 | 0.003 | -1 | 5 | 0.186 | 0.003 | 16.892 | 16.894 | Download | Download |
| 5 | predict | 100000 | 1 | 100 | 0.024 | 0.002 | 0.000 | 0.024 | -1 | 5 | 0.009 | 0.000 | 2.654 | 2.655 | Download | Download |
| 7 | predict | 100000 | 1000 | 100 | 1.792 | 0.002 | 0.000 | 0.002 | 1 | 100 | 0.231 | 0.005 | 7.772 | 7.774 | Download | Download |
| 8 | predict | 100000 | 1 | 100 | 0.020 | 0.000 | 0.000 | 0.020 | 1 | 100 | 0.009 | 0.000 | 2.239 | 2.240 | Download | Download |
| 10 | predict | 100000 | 1000 | 100 | 2.868 | 0.060 | 0.000 | 0.003 | -1 | 100 | 0.241 | 0.012 | 11.919 | 11.935 | Download | Download |
| 11 | predict | 100000 | 1 | 100 | 0.022 | 0.003 | 0.000 | 0.022 | -1 | 100 | 0.009 | 0.000 | 2.463 | 2.464 | Download | Download |
| 13 | predict | 100000 | 1000 | 100 | 1.779 | 0.005 | 0.000 | 0.002 | 1 | 5 | 0.189 | 0.004 | 9.434 | 9.436 | Download | Download |
| 14 | predict | 100000 | 1 | 100 | 0.020 | 0.000 | 0.000 | 0.020 | 1 | 5 | 0.009 | 0.000 | 2.263 | 2.263 | Download | Download |
| 16 | predict | 100000 | 1000 | 100 | 1.159 | 0.007 | 0.001 | 0.001 | 1 | 1 | 0.191 | 0.005 | 6.064 | 6.067 | Download | Download |
| 17 | predict | 100000 | 1 | 100 | 0.020 | 0.001 | 0.000 | 0.020 | 1 | 1 | 0.009 | 0.000 | 2.219 | 2.220 | Download | Download |
| 19 | predict | 100000 | 1000 | 2 | 1.880 | 0.049 | 0.000 | 0.002 | -1 | 1 | 0.032 | 0.001 | 59.549 | 59.557 | Download | Download |
| 20 | predict | 100000 | 1 | 2 | 0.004 | 0.000 | 0.000 | 0.004 | -1 | 1 | 0.001 | 0.000 | 4.650 | 4.684 | Download | Download |
| 22 | predict | 100000 | 1000 | 2 | 2.701 | 0.050 | 0.000 | 0.003 | -1 | 5 | 0.032 | 0.001 | 84.799 | 84.837 | Download | Download |
| 23 | predict | 100000 | 1 | 2 | 0.005 | 0.000 | 0.000 | 0.005 | -1 | 5 | 0.001 | 0.000 | 6.353 | 6.389 | Download | Download |
| 25 | predict | 100000 | 1000 | 2 | 1.732 | 0.007 | 0.000 | 0.002 | 1 | 100 | 0.072 | 0.001 | 24.029 | 24.032 | Download | Download |
| 26 | predict | 100000 | 1 | 2 | 0.003 | 0.000 | 0.000 | 0.003 | 1 | 100 | 0.001 | 0.000 | 2.924 | 2.961 | Download | Download |
| 28 | predict | 100000 | 1000 | 2 | 2.726 | 0.094 | 0.000 | 0.003 | -1 | 100 | 0.071 | 0.002 | 38.169 | 38.183 | Download | Download |
| 29 | predict | 100000 | 1 | 2 | 0.008 | 0.006 | 0.000 | 0.008 | -1 | 100 | 0.001 | 0.000 | 9.574 | 9.674 | Download | Download |
| 31 | predict | 100000 | 1000 | 2 | 1.842 | 0.004 | 0.000 | 0.002 | 1 | 5 | 0.031 | 0.001 | 59.026 | 59.091 | Download | Download |
| 32 | predict | 100000 | 1 | 2 | 0.003 | 0.000 | 0.000 | 0.003 | 1 | 5 | 0.001 | 0.000 | 3.366 | 3.394 | Download | Download |
| 34 | predict | 100000 | 1000 | 2 | 1.081 | 0.002 | 0.000 | 0.001 | 1 | 1 | 0.031 | 0.001 | 35.273 | 35.301 | Download | Download |
| 35 | predict | 100000 | 1 | 2 | 0.002 | 0.000 | 0.000 | 0.002 | 1 | 1 | 0.001 | 0.000 | 2.124 | 2.134 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters: algorithm=kd_tree.
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 1000000 | 1000000 | 10 | 3.387 | 0.021 | 0.024 | 0.0 | -1 | 1 | 0.767 | 0.008 | 4.415 | 4.415 | Download | Download |
| 3 | fit | 1000000 | 1000000 | 10 | 3.359 | 0.040 | 0.024 | 0.0 | -1 | 5 | 0.785 | 0.019 | 4.282 | 4.283 | Download | Download |
| 6 | fit | 1000000 | 1000000 | 10 | 3.348 | 0.018 | 0.024 | 0.0 | 1 | 100 | 0.797 | 0.019 | 4.203 | 4.204 | Download | Download |
| 9 | fit | 1000000 | 1000000 | 10 | 3.325 | 0.031 | 0.024 | 0.0 | -1 | 100 | 0.781 | 0.008 | 4.258 | 4.258 | Download | Download |
| 12 | fit | 1000000 | 1000000 | 10 | 3.300 | 0.011 | 0.024 | 0.0 | 1 | 5 | 0.770 | 0.023 | 4.287 | 4.289 | Download | Download |
| 15 | fit | 1000000 | 1000000 | 10 | 3.525 | 0.038 | 0.023 | 0.0 | 1 | 1 | 0.753 | 0.004 | 4.682 | 4.682 | Download | Download |
| 18 | fit | 1000 | 1000 | 2 | 0.001 | 0.000 | 0.031 | 0.0 | -1 | 1 | 0.001 | 0.000 | 0.564 | 0.565 | Download | Download |
| 21 | fit | 1000 | 1000 | 2 | 0.000 | 0.000 | 0.032 | 0.0 | -1 | 5 | 0.001 | 0.000 | 0.566 | 0.567 | Download | Download |
| 24 | fit | 1000 | 1000 | 2 | 0.001 | 0.000 | 0.032 | 0.0 | 1 | 100 | 0.001 | 0.000 | 0.542 | 0.542 | Download | Download |
| 27 | fit | 1000 | 1000 | 2 | 0.001 | 0.000 | 0.032 | 0.0 | -1 | 100 | 0.001 | 0.000 | 0.512 | 0.513 | Download | Download |
| 30 | fit | 1000 | 1000 | 2 | 0.001 | 0.000 | 0.031 | 0.0 | 1 | 5 | 0.001 | 0.000 | 0.506 | 0.507 | Download | Download |
| 33 | fit | 1000 | 1000 | 2 | 0.000 | 0.000 | 0.033 | 0.0 | 1 | 1 | 0.001 | 0.000 | 0.513 | 0.519 | Download | Download |
predict
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | predict | 1000000 | 1000 | 10 | 0.443 | 0.005 | 0.000 | 0.000 | -1 | 1 | 0.116 | 0.001 | 3.832 | 3.832 | Download | Download |
| 2 | predict | 1000000 | 1 | 10 | 0.003 | 0.000 | 0.000 | 0.003 | -1 | 1 | 0.000 | 0.000 | 9.813 | 10.325 | Download | Download |
| 4 | predict | 1000000 | 1000 | 10 | 0.822 | 0.004 | 0.000 | 0.001 | -1 | 5 | 0.208 | 0.002 | 3.954 | 3.954 | Download | Download |
| 5 | predict | 1000000 | 1 | 10 | 0.004 | 0.001 | 0.000 | 0.004 | -1 | 5 | 0.000 | 0.000 | 11.415 | 12.103 | Download | Download |
| 7 | predict | 1000000 | 1000 | 10 | 4.683 | 0.013 | 0.000 | 0.005 | 1 | 100 | 0.626 | 0.007 | 7.482 | 7.483 | Download | Download |
| 8 | predict | 1000000 | 1 | 10 | 0.003 | 0.000 | 0.000 | 0.003 | 1 | 100 | 0.001 | 0.000 | 4.649 | 4.814 | Download | Download |
| 10 | predict | 1000000 | 1000 | 10 | 2.676 | 0.017 | 0.000 | 0.003 | -1 | 100 | 0.622 | 0.008 | 4.301 | 4.301 | Download | Download |
| 11 | predict | 1000000 | 1 | 10 | 0.006 | 0.001 | 0.000 | 0.006 | -1 | 100 | 0.001 | 0.000 | 8.950 | 9.283 | Download | Download |
| 13 | predict | 1000000 | 1000 | 10 | 1.433 | 0.018 | 0.000 | 0.001 | 1 | 5 | 0.205 | 0.001 | 7.007 | 7.007 | Download | Download |
| 14 | predict | 1000000 | 1 | 10 | 0.001 | 0.000 | 0.000 | 0.001 | 1 | 5 | 0.000 | 0.000 | 4.005 | 4.232 | Download | Download |
| 16 | predict | 1000000 | 1000 | 10 | 0.754 | 0.002 | 0.000 | 0.001 | 1 | 1 | 0.116 | 0.002 | 6.481 | 6.481 | Download | Download |
| 17 | predict | 1000000 | 1 | 10 | 0.001 | 0.000 | 0.000 | 0.001 | 1 | 1 | 0.000 | 0.000 | 3.644 | 3.884 | Download | Download |
| 19 | predict | 1000 | 1000 | 2 | 0.021 | 0.001 | 0.001 | 0.000 | -1 | 1 | 0.001 | 0.001 | 21.381 | 34.023 | Download | Download |
| 20 | predict | 1000 | 1 | 2 | 0.002 | 0.000 | 0.000 | 0.002 | -1 | 1 | 0.000 | 0.000 | 20.055 | 21.333 | Download | Download |
| 22 | predict | 1000 | 1000 | 2 | 0.021 | 0.000 | 0.001 | 0.000 | -1 | 5 | 0.001 | 0.000 | 27.835 | 27.916 | Download | Download |
| 23 | predict | 1000 | 1 | 2 | 0.002 | 0.000 | 0.000 | 0.002 | -1 | 5 | 0.000 | 0.000 | 17.885 | 19.098 | Download | Download |
| 25 | predict | 1000 | 1000 | 2 | 0.030 | 0.000 | 0.001 | 0.000 | 1 | 100 | 0.004 | 0.000 | 6.861 | 6.868 | Download | Download |
| 26 | predict | 1000 | 1 | 2 | 0.001 | 0.000 | 0.000 | 0.001 | 1 | 100 | 0.000 | 0.000 | 4.374 | 4.772 | Download | Download |
| 28 | predict | 1000 | 1000 | 2 | 0.035 | 0.001 | 0.000 | 0.000 | -1 | 100 | 0.004 | 0.000 | 7.830 | 7.837 | Download | Download |
| 29 | predict | 1000 | 1 | 2 | 0.002 | 0.001 | 0.000 | 0.002 | -1 | 100 | 0.000 | 0.000 | 16.760 | 18.719 | Download | Download |
| 31 | predict | 1000 | 1000 | 2 | 0.019 | 0.000 | 0.001 | 0.000 | 1 | 5 | 0.001 | 0.000 | 27.899 | 28.001 | Download | Download |
| 32 | predict | 1000 | 1 | 2 | 0.001 | 0.000 | 0.000 | 0.001 | 1 | 5 | 0.000 | 0.000 | 4.745 | 5.079 | Download | Download |
| 34 | predict | 1000 | 1000 | 2 | 0.017 | 0.000 | 0.001 | 0.000 | 1 | 1 | 0.000 | 0.000 | 41.923 | 42.282 | Download | Download |
| 35 | predict | 1000 | 1 | 2 | 0.001 | 0.000 | 0.000 | 0.001 | 1 | 1 | 0.000 | 0.000 | 4.638 | 4.968 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters: algorithm=full, n_clusters=3, max_iter=30, n_init=1, tol=1e-16.
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 1000000 | 1000000 | 2 | 0.482 | 0.006 | 0.033 | 0.0 | 30 | random | 0.418 | 0.022 | 1.151 | 1.153 | Download | Download |
| 3 | fit | 1000000 | 1000000 | 2 | 0.559 | 0.010 | 0.029 | 0.0 | 30 | k-means++ | 0.449 | 0.025 | 1.246 | 1.248 | Download | Download |
| 6 | fit | 1000000 | 1000000 | 100 | 5.181 | 0.415 | 0.154 | 0.0 | 30 | random | 2.858 | 0.027 | 1.813 | 1.813 | Download | Download |
| 9 | fit | 1000000 | 1000000 | 100 | 5.106 | 0.024 | 0.157 | 0.0 | 30 | k-means++ | 3.045 | 0.039 | 1.677 | 1.677 | Download | Download |
predict
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | predict | 1000000 | 1000 | 2 | 0.0 | 0.000 | 0.074 | 0.0 | 30 | random | 0.0 | 0.0 | 1.103 | 1.192 | Download | Download |
| 2 | predict | 1000000 | 1 | 2 | 0.0 | 0.000 | 0.000 | 0.0 | 30 | random | 0.0 | 0.0 | 1.391 | 1.466 | Download | Download |
| 4 | predict | 1000000 | 1000 | 2 | 0.0 | 0.000 | 0.070 | 0.0 | 30 | k-means++ | 0.0 | 0.0 | 1.037 | 1.151 | Download | Download |
| 5 | predict | 1000000 | 1 | 2 | 0.0 | 0.001 | 0.000 | 0.0 | 30 | k-means++ | 0.0 | 0.0 | 3.176 | 3.457 | Download | Download |
| 7 | predict | 1000000 | 1000 | 100 | 0.0 | 0.000 | 1.698 | 0.0 | 30 | random | 0.0 | 0.0 | 1.312 | 1.382 | Download | Download |
| 8 | predict | 1000000 | 1 | 100 | 0.0 | 0.000 | 0.005 | 0.0 | 30 | random | 0.0 | 0.0 | 1.099 | 1.149 | Download | Download |
| 10 | predict | 1000000 | 1000 | 100 | 0.0 | 0.000 | 1.685 | 0.0 | 30 | k-means++ | 0.0 | 0.0 | 1.434 | 1.528 | Download | Download |
| 11 | predict | 1000000 | 1 | 100 | 0.0 | 0.000 | 0.004 | 0.0 | 30 | k-means++ | 0.0 | 0.0 | 1.153 | 1.199 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters: algorithm=full, n_clusters=300, max_iter=20, n_init=1, tol=1e-16.
| 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 | 0.032069 | predict | 10000 | 1000 | 100 | 0.001505 | 0.000121 | 0.531682 | 0.000002 | 20 | full | random | 20 | 300 | 1 | 1.000000e-16 | 0.293414 | 0.001367 | 0.000108 | 0.325483 | 1.100712 | 1.104112 |
| 10 | sklearn_KMeans_short | sklearn | 0.032656 | predict | 10000 | 1000 | 100 | 0.002034 | 0.001902 | 0.393260 | 0.000002 | 20 | full | k-means++ | 20 | 300 | 1 | 1.000000e-16 | 0.283247 | 0.001393 | 0.000126 | 0.315903 | 1.460296 | 1.466269 |
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 10000 | 10000 | 2 | 0.094 | 0.012 | 0.002 | 0.0 | 20 | random | 0.038 | 0.002 | 2.498 | 2.502 | Download | Download |
| 3 | fit | 10000 | 10000 | 2 | 0.224 | 0.002 | 0.001 | 0.0 | 20 | k-means++ | 0.091 | 0.002 | 2.460 | 2.461 | Download | Download |
| 6 | fit | 10000 | 10000 | 100 | 0.233 | 0.003 | 0.034 | 0.0 | 20 | random | 0.157 | 0.002 | 1.477 | 1.477 | Download | Download |
| 9 | fit | 10000 | 10000 | 100 | 0.647 | 0.017 | 0.012 | 0.0 | 20 | k-means++ | 0.377 | 0.003 | 1.717 | 1.717 | Download | Download |
predict
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | predict | 10000 | 1000 | 2 | 0.001 | 0.000 | 0.024 | 0.0 | 20 | random | 0.001 | 0.0 | 0.931 | 0.934 | Download | Download |
| 2 | predict | 10000 | 1 | 2 | 0.000 | 0.000 | 0.000 | 0.0 | 20 | random | 0.000 | 0.0 | 1.379 | 1.430 | Download | Download |
| 4 | predict | 10000 | 1000 | 2 | 0.001 | 0.000 | 0.023 | 0.0 | 20 | k-means++ | 0.001 | 0.0 | 0.956 | 0.963 | Download | Download |
| 5 | predict | 10000 | 1 | 2 | 0.000 | 0.000 | 0.000 | 0.0 | 20 | k-means++ | 0.000 | 0.0 | 1.404 | 1.504 | Download | Download |
| 7 | predict | 10000 | 1000 | 100 | 0.002 | 0.000 | 0.532 | 0.0 | 20 | random | 0.001 | 0.0 | 1.101 | 1.104 | Download | Download |
| 8 | predict | 10000 | 1 | 100 | 0.000 | 0.000 | 0.003 | 0.0 | 20 | random | 0.000 | 0.0 | 1.219 | 1.292 | Download | Download |
| 10 | predict | 10000 | 1000 | 100 | 0.002 | 0.002 | 0.393 | 0.0 | 20 | k-means++ | 0.001 | 0.0 | 1.460 | 1.466 | Download | Download |
| 11 | predict | 10000 | 1 | 100 | 0.000 | 0.000 | 0.002 | 0.0 | 20 | k-means++ | 0.000 | 0.0 | 1.419 | 1.509 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
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.
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 1000000 | 1000000 | 100 | 11.149 | 0.054 | 0.072 | 0.000 | [20] | 1.995 | 0.018 | 5.587 | 5.587 | Download | Download |
| 3 | fit | 1000 | 1000 | 10000 | 0.794 | 0.049 | 0.101 | 0.001 | [27] | 0.769 | 0.044 | 1.032 | 1.034 | Download | Download |
predict
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | predict | 1000000 | 1000 | 100 | 0.000 | 0.0 | 2.651 | 0.0 | [20] | 0.001 | 0.002 | 0.299 | 0.615 | Download | Download |
| 2 | predict | 1000000 | 1 | 100 | 0.000 | 0.0 | 0.012 | 0.0 | [20] | 0.000 | 0.000 | 0.312 | 0.330 | Download | Download |
| 4 | predict | 1000 | 100 | 10000 | 0.002 | 0.0 | 4.382 | 0.0 | [27] | 0.004 | 0.000 | 0.497 | 0.497 | Download | Download |
| 5 | predict | 1000 | 1 | 10000 | 0.000 | 0.0 | 0.981 | 0.0 | [27] | 0.001 | 0.000 | 0.113 | 0.115 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters:
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 1000 | 1000 | 10000 | 1.430 | 0.021 | 0.056 | 0.001 | True | True | deprecated | False | 1.668 | 0.023 | 0.857 | 0.857 | Download | Download |
| 3 | fit | 1000000 | 1000000 | 100 | 8.322 | 0.552 | 0.096 | 0.000 | True | True | deprecated | False | 0.210 | 0.003 | 39.654 | 39.657 | Download | Download |
predict
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | predict | 1000 | 1000 | 10000 | 0.011 | 0.0 | 7.317 | 0.0 | True | True | deprecated | False | 0.018 | 0.0 | 0.601 | 0.601 | Download | Download |
| 2 | predict | 1000 | 10 | 10000 | 0.000 | 0.0 | 5.230 | 0.0 | True | True | deprecated | False | 0.000 | 0.0 | 0.480 | 0.500 | Download | Download |
| 4 | predict | 1000000 | 1000 | 100 | 0.000 | 0.0 | 5.587 | 0.0 | True | True | deprecated | False | 0.000 | 0.0 | 0.287 | 0.319 | Download | Download |
| 5 | predict | 1000000 | 10 | 100 | 0.000 | 0.0 | 0.115 | 0.0 | True | True | deprecated | False | 0.000 | 0.0 | 0.429 | 0.448 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters: alpha=1e-06.
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 1000 | 1000 | 10000 | 0.184 | 0.005 | 0.434 | 0.0 | 0.190 | 0.003 | 0.973 | 0.973 | Download | Download |
| 3 | fit | 1000000 | 1000000 | 100 | 1.092 | 0.014 | 0.732 | 0.0 | 0.241 | 0.003 | 4.524 | 4.525 | Download | Download |
predict
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | predict | 1000 | 1000 | 10000 | 0.011 | 0.0 | 7.334 | 0.0 | 0.018 | 0.0 | 0.612 | 0.612 | Download | Download |
| 2 | predict | 1000 | 10 | 10000 | 0.000 | 0.0 | 5.347 | 0.0 | 0.000 | 0.0 | 0.417 | 0.467 | Download | Download |
| 4 | predict | 1000000 | 1000 | 100 | 0.000 | 0.0 | 5.455 | 0.0 | 0.000 | 0.0 | 0.481 | 0.506 | Download | Download |
| 5 | predict | 1000000 | 10 | 100 | 0.000 | 0.0 | 0.125 | 0.0 | 0.000 | 0.0 | 0.398 | 0.413 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters:
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 1000 | 1000 | 3 | 3.338 | 0.227 | 0.0 | 0.003 | 1000 | 0.5 | 12.0 | warn | warn | barnes_hut | euclidean | 0.0 | 2 | 300 | 30.0 | legacy | 0 | 3.588 | 0.384 | 0.93 | 0.936 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters: n_components=10.
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 100000 | 100000 | 100 | 0.481 | 0.007 | 0.166 | 0.0 | 0.031 | 0.006 | 15.261 | 15.519 | Download | Download |
| 1 | fit | 10000 | 10000 | 1000 | 0.415 | 0.003 | 0.193 | 0.0 | 0.214 | 0.004 | 1.942 | 1.943 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
scikit-learn-intelex (2021.20220207.123530) vs. scikit-learn (1.0.2)
All estimators share the following parameters: C=0.05, kernel=rbf, gamma=scale, tol=1e-16, probability=True.
fit
| 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 | sklearn_profiling | sklearnex_profiling | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | fit | 5000 | 5000 | 100 | 11.229 | 0.064 | 0.0 | 0.002 | 4.88 | 1.031 | 2.301 | 2.352 | Download | Download |
Profiling traces can be visualized using Perfetto UI.
| 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 |
| version | |
|---|---|
| pip | 22.0.4 |
| setuptools | 60.9.3 |
| 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 |
| 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 | 2 |
|---|---|
| physical_cpu_count | 2 |