# Supported Algorithms¶

Applying Intel® Extension for Scikit-learn* will impact the following scikit-learn algorithms:

## on CPU¶

Task |
Functionality |
Parameters support |
Data support |
---|---|---|---|

Classification |
SVC |
All parameters are supported |
No limitations. |

Classification |
NuSVC |
All parameters are supported |
No limitations. |

Classification |
RandomForestClassifier |
All parameters except |
Multi-output and sparse data are not supported. |

Classification |
KNeighborsClassifier |
All parameters except |
Multi-output and sparse data is not supported. |

Classification |
LogisticRegression |
All parameters except |
Only dense data is supported. |

Regression |
SVR |
All parameters are supported |
No limitations. |

Regression |
NuSVR |
All parameters are supported |
No limitations. |

Regression |
RandomForestRegressor |
All parameters except |
Multi-output and sparse data are not supported. |

Regression |
KNeighborsRegressor |
All parameters except |
Multi-output and sparse data is not supported. |

Regression |
LinearRegression |
All parameters except |
Only dense data is supported, #observations should be >= #features. |

Regression |
Ridge |
All parameters except |
Only dense data is supported, #observations should be >= #features. |

Regression |
ElasticNet |
All parameters except |
Multi-output and sparse data is not supported, #observations should be >= #features. |

Regression |
Lasso |
All parameters except |
Multi-output and sparse data is not supported, #observations should be >= #features. |

Clustering |
KMeans |
All parameters except |
No limitations. |

Clustering |
DBSCAN |
All parameters except |
Only dense data is supported. |

Dimensionality reduction |
PCA |
All parameters except |
Sparse data is not supported. |

Dimensionality reduction |
TSNE |
All parameters except |
Sparse data is not supported. |

Unsupervised |
NearestNeighbors |
All parameters except |
Sparse data is not supported. |

Other |
train_test_split |
All parameters are supported. |
Only dense data is supported. |

Other |
assert_all_finite |
All parameters are supported. |
Only dense data is supported. |

Other |
pairwise_distance |
With |
Only dense data is supported. |

Other |
roc_auc_score |
Parameters |
No limitations. |

## on GPU¶

Task |
Functionality |
Parameters support |
Data support |
---|---|---|---|

Classification |
SVC |
All parameters except |
Only binary dense data is supported. |

Classification |
RandomForestClassifier |
All parameters except |
Multi-output, sparse data, out-of-bag score and sample_weight are not supported. |

Classification |
KNeighborsClassifier |
All parameters except |
Only dense data is supported. |

Classification |
LogisticRegression |
All parameters except |
Only dense data is supported. |

Regression |
RandomForestRegressor |
All parameters except |
Multi-output, sparse data, out-of-bag score and sample_weight are not supported. |

Regression |
KNeighborsRegressor |
All parameters except |
Only dense data is supported. |

Regression |
LinearRegression |
All parameters except |
Only dense data is supported, #observations should be >= #features. |

Clustering |
KMeans |
All parameters except |
Sparse data is not supported. |

Clustering |
DBSCAN |
All parameters except |
Only dense data is supported. |

Dimensionality reduction |
PCA |
All parameters except |
Sparse data is not supported. |

## Scikit-learn tests¶

Monkey-patched scikit-learn classes and functions passes scikit-learn’s own test suite, with few exceptions, specified in deselected_tests.yaml.

The results of the entire latest scikit-learn test suite with Intel® Extension for Scikit-learn*: CircleCI.