gemseo / mlearning

api module

Machine learning API.

The machine learning API provides methods for creating new and loading existing machine learning models. It also provides methods for listing available models and options.

gemseo.mlearning.api.create_classification_model(name, data, transformer=mappingproxy({'inputs': <gemseo.mlearning.transform.scaler.min_max_scaler.MinMaxScaler object>}), **parameters)

Create a classification model from a learning dataset.

Parameters:

• name (str) – The name of the classification algorithm.

• data (Dataset) – The learning dataset.

• transformer (MutableMapping[str, Union[str, Tuple[str, Mapping[str, Any]], Transformer]]) –

  The strategies to transform the variables. Values are instances of Transformer while keys are names of either variables or groups of variables. If IDENTITY, do not transform the variables.

  By default it is set to {‘inputs’: <gemseo.mlearning.transform.scaler.min_max_scaler.MinMaxScaler object at 0x7fcd0601a850>}.

• parameters – The parameters of the classification model.

Returns:

A classification model.

Return type:

MLClassificationAlgo

See also

get_classification_models, get_classification_options, import_classification_model

gemseo.mlearning.api.create_clustering_model(name, data, transformer=mappingproxy({}), **parameters)

Create a clustering model from a learning dataset.

Parameters:

• name (str) – The name of the clustering algorithm.

• data (Dataset) – The learning dataset.

• transformer (MutableMapping[str, Union[str, Tuple[str, Mapping[str, Any]], Transformer]]) –

  The strategies to transform the variables. Values are instances of Transformer while keys are names of either variables or groups of variables. If IDENTITY, do not transform the variables.

  By default it is set to {}.

• parameters – The parameters of the clustering model.

Returns:

A clustering model.

Return type:

MLClusteringAlgo

See also

get_clustering_models, get_clustering_options, import_clustering_model

gemseo.mlearning.api.create_mlearning_model(name, data, transformer=mappingproxy({}), **parameters)

Create a machine learning algorithm from a learning dataset.

Parameters:

• name (str) – The name of the machine learning algorithm.

• data (Dataset) – The learning dataset.

• transformer (MutableMapping[str, Union[str, Tuple[str, Mapping[str, Any]], Transformer]]) –

  The strategies to transform the variables. Values are instances of Transformer while keys are names of either variables or groups of variables. If IDENTITY, do not transform the variables.

  By default it is set to {}.

• parameters – The parameters of the machine learning algorithm.

Returns:

A machine learning model.

Return type:

MLAlgo

See also

get_mlearning_models, get_mlearning_options, import_mlearning_model

gemseo.mlearning.api.create_regression_model(name, data, transformer=mappingproxy({'inputs': <gemseo.mlearning.transform.scaler.min_max_scaler.MinMaxScaler object>, 'outputs': <gemseo.mlearning.transform.scaler.min_max_scaler.MinMaxScaler object>}), **parameters)

Create a regression model from a learning dataset.

Parameters:

• name (str) – The name of the regression algorithm.

• data (Dataset) – The learning dataset.

• transformer (MutableMapping[str, Union[str, Tuple[str, Mapping[str, Any]], Transformer]]) –

  The strategies to transform the variables. Values are instances of Transformer while keys are names of either variables or groups of variables. If IDENTITY, do not transform the variables.

  By default it is set to {‘inputs’: <gemseo.mlearning.transform.scaler.min_max_scaler.MinMaxScaler object at 0x7fcd0607f130>, ‘outputs’: <gemseo.mlearning.transform.scaler.min_max_scaler.MinMaxScaler object at 0x7fcd0607f0d0>}.

• parameters – The parameters of the regression model.

Returns:

A regression model.

Return type:

MLRegressionAlgo

See also

get_regression_models, get_regression_options, import_regression_model

gemseo.mlearning.api.get_classification_models()

Get available classification models.

Returns:

The available classification models.

Return type:

list[str]

See also

create_classification_model, get_classification_options, import_classification_model

gemseo.mlearning.api.get_classification_options(model_name, output_json=False, pretty_print=True)

Find the available options for a classification model.

Parameters:

• model_name (str) – The name of the classification model.

• output_json (bool) –

  Whether to apply JSON format for the schema.

  By default it is set to False.

• pretty_print (bool) –

  Print the schema in a pretty table.

  By default it is set to True.

Returns:

The options schema of the classification model.

Return type:

dict[str, str] | str

See also

create_classification_model, get_classification_models, import_classification_model

gemseo.mlearning.api.get_clustering_models()

Get available clustering models.

Returns:

The available clustering models.

Return type:

list[str]

See also

create_clustering_model, get_clustering_options, import_clustering_model

gemseo.mlearning.api.get_clustering_options(model_name, output_json=False, pretty_print=True)

Find the available options for clustering model.

Parameters:

• model_name (str) – The name of the clustering model.

• output_json (bool) –

  Whether to apply JSON format for the schema.

  By default it is set to False.

• pretty_print (bool) –

  Print the schema in a pretty table.

  By default it is set to True.

Returns:

The options schema of the clustering model.

Return type:

dict[str, str] | str

See also

create_clustering_model, get_clustering_models, import_clustering_model

gemseo.mlearning.api.get_mlearning_models()

Get available machine learning algorithms.

Returns:

The available machine learning algorithms.

Return type:

list[str]

See also

import_mlearning_model, create_mlearning_model, get_mlearning_options, import_mlearning_model

gemseo.mlearning.api.get_mlearning_options(model_name, output_json=False, pretty_print=True)

Find the available options for a machine learning algorithm.

Parameters:

• model_name (str) – The name of the machine learning algorithm.

• output_json (bool) –

  Whether to apply JSON format for the schema.

  By default it is set to False.

• pretty_print (bool) –

  Whether to print the schema in a pretty table.

  By default it is set to True.

Returns:

The options schema of the machine learning algorithm.

Return type:

dict[str, str] | str

See also

create_mlearning_model, get_mlearning_models, import_mlearning_model

gemseo.mlearning.api.get_regression_models()

Get available regression models.

Returns:

The available regression models.

Return type:

list[str]

See also

create_regression_model, get_regression_options, import_regression_model

gemseo.mlearning.api.get_regression_options(model_name, output_json=False, pretty_print=True)

Find the available options for a regression model.

Parameters:

• model_name (str) – The name of the regression model.

• output_json (bool) –

  Whether to apply JSON format for the schema.

  By default it is set to False.

• pretty_print (bool) –

  Print the schema in a pretty table.

  By default it is set to True.

Returns:

The options schema of the regression model.

Return type:

dict[str, str] | str

See also

create_regression_model, get_regression_models, import_regression_model

gemseo.mlearning.api.import_classification_model(directory)

Import a classification model from a directory.

Parameters:

directory (str | Path) – The path to the directory.

Returns:

A classification model.

Return type:

MLClassificationAlgo

See also

create_classification_model, get_classification_models, get_classification_options

gemseo.mlearning.api.import_clustering_model(directory)

Import a clustering model from a directory.

Parameters:

directory (str | Path) – The path to the directory.

Returns:

A clustering model.

Return type:

MLClusteringAlgo

See also

create_clustering_model, get_clustering_models, get_clustering_options

gemseo.mlearning.api.import_mlearning_model(directory)

Import a machine learning algorithm from a directory.

Parameters:

directory (str | Path) – The path to the directory.

Returns:

A machine learning model.

Return type:

MLAlgo

See also

create_mlearning_model, get_mlearning_models, get_mlearning_options

gemseo.mlearning.api.import_regression_model(directory)

Import a regression model from a directory.

Parameters:

directory (str | Path) – The path of the directory.

Returns:

A regression model.

Return type:

MLRegressionAlgo

See also

create_regression_model, get_regression_models, get_regression_options

Examples using create_classification_model

Classification API

Classification API

K nearest neighbors classification

K nearest neighbors classification

Random forest classification

Random forest classification

Examples using create_clustering_model

API

API

Gaussian Mixtures

Gaussian Mixtures

K-means

K-means

Examples using create_regression_model

Mixture of experts with PCA on Burgers dataset

Mixture of experts with PCA on Burgers dataset

MSE example - test-train split

MSE example - test-train split

Quality measure for surrogate model comparison

Quality measure for surrogate model comparison

API

API

Advanced mixture of experts

Advanced mixture of experts

GP regression

GP regression

Linear regression

Linear regression

Mixture of experts

Mixture of experts

PCE regression

PCE regression

Polynomial regression

Polynomial regression

RBF regression

RBF regression

Random forest regression

Random forest regression

Save and Load

Save and Load

Examples using get_classification_models

Classification API

Classification API

Examples using get_classification_options

Classification API

Classification API

Examples using get_clustering_models

API

API

Examples using get_clustering_options

API

API

Examples using get_regression_models

API

API

Examples using get_regression_options

API

API

Examples using import_regression_model

PCE regression

PCE regression

Save and Load

Save and Load