Regression

Linear Models

Linear Regression

linear_model.LinearRegression([name, ...])

Creates a LinearRegression object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

LinearRegression.contour([nbins, chart])	Draws the model's contour plot.
LinearRegression.deploySQL([X])	Returns the SQL code needed to deploy the model.
LinearRegression.drop()	Drops the model from the VAST DataBase.
LinearRegression.export_models(name, path[, ...])	Exports machine learning models.
LinearRegression.features_importance([show, ...])	Computes the model's features importance.
LinearRegression.fit(input_relation, X, y[, ...])	Trains the model.
LinearRegression.get_attributes([attr_name])	Returns the model attributes.
LinearRegression.get_match_index(x, col_list)	Returns the matching index.
LinearRegression.get_params()	Returns the parameters of the model.
LinearRegression.get_plotting_lib([...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
LinearRegression.import_models(path[, ...])	Imports machine learning models.
LinearRegression.plot([max_nb_points, chart])	Draws the model.
LinearRegression.predict(vdf[, X, name, inplace])	Predicts using the input relation.
LinearRegression.regression_report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
LinearRegression.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
LinearRegression.score([metric])	Computes the model score.
LinearRegression.set_params([parameters])	Sets the parameters of the model.
LinearRegression.summarize()	Summarizes the model.
LinearRegression.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
LinearRegression.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
LinearRegression.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

LinearRegression.object_type

Ridge

linear_model.Ridge([name, overwrite_model])

Creates a Ridge object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

Ridge.contour([nbins, chart])	Draws the model's contour plot.
Ridge.deploySQL([X])	Returns the SQL code needed to deploy the model.
Ridge.drop()	Drops the model from the VAST DataBase.
Ridge.export_models(name, path[, kind])	Exports machine learning models.
Ridge.features_importance([show, chart])	Computes the model's features importance.
Ridge.fit(input_relation, X, y[, ...])	Trains the model.
Ridge.get_attributes([attr_name])	Returns the model attributes.
Ridge.get_match_index(x, col_list[, str_check])	Returns the matching index.
Ridge.get_params()	Returns the parameters of the model.
Ridge.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
Ridge.import_models(path[, schema, kind])	Imports machine learning models.
Ridge.plot([max_nb_points, chart])	Draws the model.
Ridge.predict(vdf[, X, name, inplace])	Predicts using the input relation.
Ridge.regression_report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
Ridge.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
Ridge.score([metric])	Computes the model score.
Ridge.set_params([parameters])	Sets the parameters of the model.
Ridge.summarize()	Summarizes the model.
Ridge.to_binary(path)	Exports the model to the VAST Binary format.
Ridge.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
Ridge.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
Ridge.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

Ridge.object_type

Lasso

linear_model.Lasso([name, overwrite_model])

Creates a Lasso object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

Lasso.contour([nbins, chart])	Draws the model's contour plot.
Lasso.deploySQL([X])	Returns the SQL code needed to deploy the model.
Lasso.drop()	Drops the model from the VAST DataBase.
Lasso.export_models(name, path[, kind])	Exports machine learning models.
Lasso.features_importance([show, chart])	Computes the model's features importance.
Lasso.fit(input_relation, X, y[, ...])	Trains the model.
Lasso.get_attributes([attr_name])	Returns the model attributes.
Lasso.get_match_index(x, col_list[, str_check])	Returns the matching index.
Lasso.get_params()	Returns the parameters of the model.
Lasso.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
Lasso.import_models(path[, schema, kind])	Imports machine learning models.
Lasso.plot([max_nb_points, chart])	Draws the model.
Lasso.predict(vdf[, X, name, inplace])	Predicts using the input relation.
Lasso.regression_report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
Lasso.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
Lasso.score([metric])	Computes the model score.
Lasso.set_params([parameters])	Sets the parameters of the model.
Lasso.summarize()	Summarizes the model.
Lasso.to_binary(path)	Exports the model to the VAST Binary format.
Lasso.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
Lasso.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
Lasso.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

Lasso.object_type

Elastic Net

linear_model.ElasticNet([name, overwrite_model])

Creates an ElasticNet object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

ElasticNet.contour([nbins, chart])	Draws the model's contour plot.
ElasticNet.deploySQL([X])	Returns the SQL code needed to deploy the model.
ElasticNet.drop()	Drops the model from the VAST DataBase.
ElasticNet.export_models(name, path[, kind])	Exports machine learning models.
ElasticNet.features_importance([show, chart])	Computes the model's features importance.
ElasticNet.fit(input_relation, X, y[, ...])	Trains the model.
ElasticNet.get_attributes([attr_name])	Returns the model attributes.
ElasticNet.get_match_index(x, col_list[, ...])	Returns the matching index.
ElasticNet.get_params()	Returns the parameters of the model.
ElasticNet.get_plotting_lib([class_name, ...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
ElasticNet.import_models(path[, schema, kind])	Imports machine learning models.
ElasticNet.plot([max_nb_points, chart])	Draws the model.
ElasticNet.predict(vdf[, X, name, inplace])	Predicts using the input relation.
ElasticNet.regression_report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
ElasticNet.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
ElasticNet.score([metric])	Computes the model score.
ElasticNet.set_params([parameters])	Sets the parameters of the model.
ElasticNet.summarize()	Summarizes the model.
ElasticNet.to_binary(path)	Exports the model to the VAST Binary format.
ElasticNet.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
ElasticNet.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
ElasticNet.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

ElasticNet.object_type

Linear SVR

svm.LinearSVR([name, overwrite_model])

Creates an LinearSVR object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

LinearSVR.contour([nbins, chart])	Draws the model's contour plot.
LinearSVR.deploySQL([X])	Returns the SQL code needed to deploy the model.
LinearSVR.drop()	Drops the model from the VAST DataBase.
LinearSVR.export_models(name, path[, kind])	Exports machine learning models.
LinearSVR.features_importance([show, chart])	Computes the model's features importance.
LinearSVR.fit(input_relation, X, y[, ...])	Trains the model.
LinearSVR.get_attributes([attr_name])	Returns the model attributes.
LinearSVR.get_match_index(x, col_list[, ...])	Returns the matching index.
LinearSVR.get_params()	Returns the parameters of the model.
LinearSVR.get_plotting_lib([class_name, ...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
LinearSVR.import_models(path[, schema, kind])	Imports machine learning models.
LinearSVR.plot([max_nb_points, chart])	Draws the model.
LinearSVR.predict(vdf[, X, name, inplace])	Predicts using the input relation.
LinearSVR.regression_report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
LinearSVR.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
LinearSVR.score([metric])	Computes the model score.
LinearSVR.set_params([parameters])	Sets the parameters of the model.
LinearSVR.summarize()	Summarizes the model.
LinearSVR.to_binary(path)	Exports the model to the VAST Binary format.
LinearSVR.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
LinearSVR.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
LinearSVR.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

LinearSVR.object_type

Partial Least Squares (PLS)

linear_model.PLSRegression([name, ...])

Creates an PLSRegression object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

PLSRegression.contour([nbins, chart])	Draws the model's contour plot.
PLSRegression.deploySQL([X])	Returns the SQL code needed to deploy the model.
PLSRegression.drop()	Drops the model from the VAST DataBase.
PLSRegression.export_models(name, path[, kind])	Exports machine learning models.
PLSRegression.features_importance([show, chart])	Computes the model's features importance.
PLSRegression.fit(input_relation, X, y[, ...])	Trains the model.
PLSRegression.get_attributes([attr_name])	Returns the model attributes.
PLSRegression.get_match_index(x, col_list[, ...])	Returns the matching index.
PLSRegression.get_params()	Returns the parameters of the model.
PLSRegression.get_plotting_lib([class_name, ...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
PLSRegression.import_models(path[, schema, kind])	Imports machine learning models.
PLSRegression.plot([max_nb_points, chart])	Draws the model.
PLSRegression.predict(vdf[, X, name, inplace])	Predicts using the input relation.
PLSRegression.regression_report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
PLSRegression.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
PLSRegression.score([metric])	Computes the model score.
PLSRegression.set_params([parameters])	Sets the parameters of the model.
PLSRegression.summarize()	Summarizes the model.
PLSRegression.to_binary(path)	Exports the model to the VAST Binary format.
PLSRegression.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
PLSRegression.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
PLSRegression.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

PLSRegression._model_type

Must be overridden in child class

Poisson Regression

linear_model.PoissonRegressor([name, ...])

Creates an PoissonRegressor object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

PoissonRegressor.contour([nbins, chart])	Draws the model's contour plot.
PoissonRegressor.deploySQL([X])	Returns the SQL code needed to deploy the model.
PoissonRegressor.drop()	Drops the model from the VAST DataBase.
PoissonRegressor.export_models(name, path[, ...])	Exports machine learning models.
PoissonRegressor.features_importance([show, ...])	Computes the model's features importance.
PoissonRegressor.fit(input_relation, X, y[, ...])	Trains the model.
PoissonRegressor.get_attributes([attr_name])	Returns the model attributes.
PoissonRegressor.get_match_index(x, col_list)	Returns the matching index.
PoissonRegressor.get_params()	Returns the parameters of the model.
PoissonRegressor.get_plotting_lib([...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
PoissonRegressor.import_models(path[, ...])	Imports machine learning models.
PoissonRegressor.plot([max_nb_points, chart])	Draws the model.
PoissonRegressor.predict(vdf[, X, name, inplace])	Predicts using the input relation.
PoissonRegressor.regression_report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
PoissonRegressor.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
PoissonRegressor.score([metric])	Computes the model score.
PoissonRegressor.set_params([parameters])	Sets the parameters of the model.
PoissonRegressor.summarize()	Summarizes the model.
PoissonRegressor.to_binary(path)	Exports the model to the VAST Binary format.
PoissonRegressor.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
PoissonRegressor.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
PoissonRegressor.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

---

Tree-based Models

Random Forest Regressor

ensemble.RandomForestRegressor([name, ...])

Creates an RandomForestRegressor object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

RandomForestRegressor.contour([nbins, chart])	Draws the model's contour plot.
RandomForestRegressor.deploySQL([X])	Returns the SQL code needed to deploy the model.
RandomForestRegressor.drop()	Drops the model from the VAST DataBase.
RandomForestRegressor.export_models(name, path)	Exports machine learning models.
RandomForestRegressor.features_importance([...])	Computes the model's features importance.
RandomForestRegressor.fit(input_relation, X, y)	Trains the model.
RandomForestRegressor.get_attributes([attr_name])	Returns the model attributes.
RandomForestRegressor.get_match_index(x, ...)	Returns the matching index.
RandomForestRegressor.get_params()	Returns the parameters of the model.
RandomForestRegressor.get_plotting_lib([...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
RandomForestRegressor.get_tree([tree_id])	Returns a table with all the input tree information.
RandomForestRegressor.import_models(path[, ...])	Imports machine learning models.
RandomForestRegressor.plot([max_nb_points, ...])	Draws the model.
RandomForestRegressor.plot_tree([tree_id, ...])	Draws the input tree.
RandomForestRegressor.predict(vdf[, X, ...])	Predicts using the input relation.
RandomForestRegressor.regression_report([...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
RandomForestRegressor.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
RandomForestRegressor.score([metric])	Computes the model score.
RandomForestRegressor.set_params([parameters])	Sets the parameters of the model.
RandomForestRegressor.summarize()	Summarizes the model.
RandomForestRegressor.to_binary(path)	Exports the model to the VAST Binary format.
RandomForestRegressor.to_graphviz([tree_id, ...])	Returns the code for a Graphviz tree.
RandomForestRegressor.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
RandomForestRegressor.to_python([...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
RandomForestRegressor.to_sql([X, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

RandomForestRegressor.object_type

GradientBoostingRegressor

ensemble.GradientBoostingRegressor([name, ...])

Creates an GradientBoostingRegressor object using scikit-learn for training and the scalability of VAST DataBase for the inferences.

Methods:

GradientBoostingRegressor.contour([nbins, chart])	Draws the model's contour plot.
GradientBoostingRegressor.deploySQL([X])	Returns the SQL code needed to deploy the model.
GradientBoostingRegressor.drop()	Drops the model from the VAST DataBase.
GradientBoostingRegressor.export_models(...)	Exports machine learning models.
GradientBoostingRegressor.features_importance([...])	Computes the model's features importance.
GradientBoostingRegressor.fit(...[, ...])	Trains the model.
GradientBoostingRegressor.get_attributes([...])	Returns the model attributes.
GradientBoostingRegressor.get_match_index(x, ...)	Returns the matching index.
GradientBoostingRegressor.get_params()	Returns the parameters of the model.
GradientBoostingRegressor.get_plotting_lib([...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
GradientBoostingRegressor.get_tree([tree_id])	Returns a table with all the input tree information.
GradientBoostingRegressor.import_models(path)	Imports machine learning models.
GradientBoostingRegressor.plot([...])	Draws the model.
GradientBoostingRegressor.plot_tree([...])	Draws the input tree.
GradientBoostingRegressor.predict(vdf[, X, ...])	Predicts using the input relation.
GradientBoostingRegressor.regression_report([...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
GradientBoostingRegressor.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
GradientBoostingRegressor.score([metric])	Computes the model score.
GradientBoostingRegressor.set_params([...])	Sets the parameters of the model.
GradientBoostingRegressor.summarize()	Summarizes the model.
GradientBoostingRegressor.to_binary(path)	Exports the model to the VAST Binary format.
GradientBoostingRegressor.to_graphviz([...])	Returns the code for a Graphviz tree.
GradientBoostingRegressor.to_json([path])	Creates a Python GradientBoosting JSON file that can be imported into the Python GradientBoosting API.
GradientBoostingRegressor.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
GradientBoostingRegressor.to_python([...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
GradientBoostingRegressor.to_sql([X, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

GradientBoostingRegressor.object_type

---

Neighbors

K-Nearest Neighbors Regressor (Beta)

neighbors.KNeighborsRegressor([name, ...])

[Beta Version] Creates a KNeighborsRegressor object using the k-nearest neighbors algorithm.

Methods:

KNeighborsRegressor.contour([nbins, chart])	Draws the model's contour plot.
KNeighborsRegressor.deploySQL([X, ...])	Returns the SQL code needed to deploy the model.
KNeighborsRegressor.drop()	KNeighborsRegressor models are not stored in the VAST DataBase.
KNeighborsRegressor.export_models(name, path)	Exports machine learning models.
KNeighborsRegressor.fit(input_relation, X, y)	Trains the model.
KNeighborsRegressor.get_attributes([attr_name])	Returns the model attributes.
KNeighborsRegressor.get_match_index(x, col_list)	Returns the matching index.
KNeighborsRegressor.get_params()	Returns the parameters of the model.
KNeighborsRegressor.get_plotting_lib([...])	Returns the first available library (Plotly, Matplotlib) to draw a specific graphic.
KNeighborsRegressor.import_models(path[, ...])	Imports machine learning models.
KNeighborsRegressor.predict(vdf[, X, name, ...])	Predicts using the input relation.
KNeighborsRegressor.regression_report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
KNeighborsRegressor.report([metrics])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
KNeighborsRegressor.score([metric])	Computes the model score.
KNeighborsRegressor.set_params([parameters])	Sets the parameters of the model.
KNeighborsRegressor.summarize()	Summarizes the model.
KNeighborsRegressor.to_binary(path)	Exports the model to the VAST Binary format.
KNeighborsRegressor.to_python([...])	Returns the Python function needed for in-memory scoring without using built-in VAST functions.
KNeighborsRegressor.to_sql([X, ...])	Returns the SQL code needed to deploy the model without using built-in VAST functions.

Attributes:

KNeighborsRegressor.object_type