Clustering & Anomaly Detection#

Clustering#

K-Means#

cluster.KMeans([name, overwrite_model, ...])

Creates a KMeans object using the Vertica k-means algorithm.

Methods:

`KMeans.contour`([nbins, chart])	Draws the model's contour plot.
`KMeans.deploySQL`([X])	Returns the SQL code needed to deploy the model.
`KMeans.does_model_exists`(name[, ...])	Checks whether the model is stored in the Vertica database.
`KMeans.drop`()	Drops the model from the Vertica database.
`KMeans.export_models`(name, path[, kind])	Exports machine learning models.
`KMeans.fit`(input_relation[, X, return_report])	Trains the model.
`KMeans.get_attributes`([attr_name])	Returns the model attributes.
`KMeans.get_match_index`(x, col_list[, str_check])	Returns the matching index.
`KMeans.get_params`()	Returns the parameters of the model.
`KMeans.get_plotting_lib`([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
`KMeans.get_vertica_attributes`([attr_name])	Returns the model Vertica attributes.
`KMeans.import_models`(path[, schema, kind])	Imports machine learning models.
`KMeans.plot`([max_nb_points, chart])	Draws the model.
`KMeans.plot_voronoi`([max_nb_points, ...])	Draws the Voronoi Graph of the model.
`KMeans.predict`(vdf[, X, name, inplace])	Makes predictions using the input relation.
`KMeans.register`(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
`KMeans.set_params`([parameters])	Sets the parameters of the model.
`KMeans.summarize`()	Summarizes the model.
`KMeans.to_binary`(path)	Exports the model to the Vertica Binary format.
`KMeans.to_memmodel`()	Converts the model to an InMemory object that can be used for different types of predictions.
`KMeans.to_pmml`(path)	Exports the model to PMML.
`KMeans.to_python`([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
`KMeans.to_sql`([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
`KMeans.to_tf`(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

KMeans.object_type

K-Prototype#

cluster.KPrototypes([name, overwrite_model, ...])

Creates a KPrototypes object by using the Vertica k-prototypes algorithm.

Methods:

`KPrototypes.contour`([nbins, chart])	Draws the model's contour plot.
`KPrototypes.deploySQL`([X])	Returns the SQL code needed to deploy the model.
`KPrototypes.does_model_exists`(name[, ...])	Checks whether the model is stored in the Vertica database.
`KPrototypes.drop`()	Drops the model from the Vertica database.
`KPrototypes.export_models`(name, path[, kind])	Exports machine learning models.
`KPrototypes.fit`(input_relation[, X, ...])	Trains the model.
`KPrototypes.get_attributes`([attr_name])	Returns the model attributes.
`KPrototypes.get_match_index`(x, col_list[, ...])	Returns the matching index.
`KPrototypes.get_params`()	Returns the parameters of the model.
`KPrototypes.get_plotting_lib`([class_name, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
`KPrototypes.get_vertica_attributes`([attr_name])	Returns the model Vertica attributes.
`KPrototypes.import_models`(path[, schema, kind])	Imports machine learning models.
`KPrototypes.plot`([max_nb_points, chart])	Draws the model.
`KPrototypes.plot_voronoi`([max_nb_points, ...])	Draws the Voronoi Graph of the model.
`KPrototypes.predict`(vdf[, X, name, inplace])	Makes predictions using the input relation.
`KPrototypes.register`(registered_name[, ...])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
`KPrototypes.set_params`([parameters])	Sets the parameters of the model.
`KPrototypes.summarize`()	Summarizes the model.
`KPrototypes.to_binary`(path)	Exports the model to the Vertica Binary format.
`KPrototypes.to_memmodel`()	Converts the model to an InMemory object that can be used for different types of predictions.
`KPrototypes.to_pmml`(path)	Exports the model to PMML.
`KPrototypes.to_python`([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
`KPrototypes.to_sql`([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
`KPrototypes.to_tf`(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

KPrototypes.object_type

Bisecting K-Means#

cluster.BisectingKMeans([name, ...])

Creates a BisectingKMeans object using the Vertica bisecting k-means algorithm.

Methods:

`BisectingKMeans.contour`([nbins, chart])	Draws the model's contour plot.
`BisectingKMeans.deploySQL`([X])	Returns the SQL code needed to deploy the model.
`BisectingKMeans.does_model_exists`(name[, ...])	Checks whether the model is stored in the Vertica database.
`BisectingKMeans.drop`()	Drops the model from the Vertica database.
`BisectingKMeans.export_models`(name, path[, kind])	Exports machine learning models.
`BisectingKMeans.features_importance`([...])	Computes the model's features importance.
`BisectingKMeans.fit`(input_relation[, X, ...])	Trains the model.
`BisectingKMeans.get_attributes`([attr_name])	Returns the model attributes.
`BisectingKMeans.get_match_index`(x, col_list)	Returns the matching index.
`BisectingKMeans.get_params`()	Returns the parameters of the model.
`BisectingKMeans.get_plotting_lib`([...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
`BisectingKMeans.get_score`([tree_id])	Returns the feature importance metrics for the input tree.
`BisectingKMeans.get_tree`()	Returns a table containing information about the BK-tree.
`BisectingKMeans.get_vertica_attributes`([...])	Returns the model Vertica attributes.
`BisectingKMeans.import_models`(path[, ...])	Imports machine learning models.
`BisectingKMeans.plot`([max_nb_points, chart])	Draws the model.
`BisectingKMeans.plot_tree`([pic_path])	Draws the input tree.
`BisectingKMeans.plot_voronoi`([...])	Draws the Voronoi Graph of the model.
`BisectingKMeans.predict`(vdf[, X, name, inplace])	Makes predictions using the input relation.
`BisectingKMeans.register`(registered_name[, ...])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
`BisectingKMeans.set_params`([parameters])	Sets the parameters of the model.
`BisectingKMeans.summarize`()	Summarizes the model.
`BisectingKMeans.to_binary`(path)	Exports the model to the Vertica Binary format.
`BisectingKMeans.to_graphviz`([round_score, ...])	Returns the code for a Graphviz tree.
`BisectingKMeans.to_memmodel`()	Converts the model to an InMemory object that can be used for different types of predictions.
`BisectingKMeans.to_pmml`(path)	Exports the model to PMML.
`BisectingKMeans.to_python`([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
`BisectingKMeans.to_sql`([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
`BisectingKMeans.to_tf`(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

BisectingKMeans.object_type

DBSCAN (Beta)#

cluster.DBSCAN([name, overwrite_model, eps, ...])

[Beta Version] Creates a DBSCAN object by using the DBSCAN algorithm as defined by Martin Ester, Hans-Peter Kriegel, Jörg Sander, and Xiaowei Xu.

Methods:

`DBSCAN.contour`([nbins, chart])	Draws the model's contour plot.
`DBSCAN.deploySQL`([X])	Returns the SQL code needed to deploy the model.
`DBSCAN.does_model_exists`(name[, ...])	Checks whether the model is stored in the Vertica database.
`DBSCAN.drop`()	Drops the model from the Vertica database.
`DBSCAN.export_models`(name, path[, kind])	Exports machine learning models.
`DBSCAN.fit`(input_relation[, X, key_columns, ...])	Trains the model.
`DBSCAN.get_attributes`([attr_name])	Returns the model attributes.
`DBSCAN.get_match_index`(x, col_list[, str_check])	Returns the matching index.
`DBSCAN.get_params`()	Returns the parameters of the model.
`DBSCAN.get_plotting_lib`([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
`DBSCAN.get_vertica_attributes`([attr_name])	Returns the model Vertica attributes.
`DBSCAN.import_models`(path[, schema, kind])	Imports machine learning models.
`DBSCAN.plot`([max_nb_points, chart])	Draws the model.
`DBSCAN.predict`()	Creates a `vDataFrame` of the model.
`DBSCAN.register`(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
`DBSCAN.set_params`([parameters])	Sets the parameters of the model.
`DBSCAN.summarize`()	Summarizes the model.
`DBSCAN.to_binary`(path)	Exports the model to the Vertica Binary format.
`DBSCAN.to_pmml`(path)	Exports the model to PMML.
`DBSCAN.to_python`([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
`DBSCAN.to_sql`([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
`DBSCAN.to_tf`(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

DBSCAN.object_type

Anomaly Detection#

Isolation Forest#

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

Creates an IsolationForest object using the Vertica IFOREST algorithm.

Methods:

`IsolationForest.contour`([nbins, chart])	Draws the model's contour plot.
`IsolationForest.decision_function`(vdf[, X, ...])	Returns the anomaly score using the input relation.
`IsolationForest.deploySQL`([X, cutoff, ...])	Returns the SQL code needed to deploy the model.
`IsolationForest.does_model_exists`(name[, ...])	Checks whether the model is stored in the Vertica database.
`IsolationForest.drop`()	Drops the model from the Vertica database.
`IsolationForest.export_models`(name, path[, kind])	Exports machine learning models.
`IsolationForest.features_importance`([...])	Computes the model's features importance.
`IsolationForest.fit`(input_relation[, X, ...])	Trains the model.
`IsolationForest.get_attributes`([attr_name])	Returns the model attributes.
`IsolationForest.get_match_index`(x, col_list)	Returns the matching index.
`IsolationForest.get_params`()	Returns the parameters of the model.
`IsolationForest.get_plotting_lib`([...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
`IsolationForest.get_score`([tree_id])	Returns the feature importance metrics for the input tree.
`IsolationForest.get_tree`([tree_id])	Returns a table with all the input tree information.
`IsolationForest.get_vertica_attributes`([...])	Returns the model Vertica attributes.
`IsolationForest.import_models`(path[, ...])	Imports machine learning models.
`IsolationForest.plot`([max_nb_points, chart])	Draws the model.
`IsolationForest.plot_tree`([tree_id, pic_path])	Draws the input tree.
`IsolationForest.predict`(vdf[, X, name, ...])	Predicts using the input relation.
`IsolationForest.register`(registered_name[, ...])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
`IsolationForest.set_params`([parameters])	Sets the parameters of the model.
`IsolationForest.summarize`()	Summarizes the model.
`IsolationForest.to_binary`(path)	Exports the model to the Vertica Binary format.
`IsolationForest.to_graphviz`([tree_id, ...])	Returns the code for a Graphviz tree.
`IsolationForest.to_memmodel`()	Converts the model to an InMemory object that can be used for different types of predictions.
`IsolationForest.to_pmml`(path)	Exports the model to PMML.
`IsolationForest.to_python`([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
`IsolationForest.to_sql`([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
`IsolationForest.to_tf`(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

IsolationForest.object_type

Local Outlier Factor (Beta)#

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

[Beta Version] Creates a LocalOutlierFactor object by using the Local Outlier Factor algorithm as defined by Markus M.

Methods:

`LocalOutlierFactor.contour`([nbins, chart])	Draws the model's contour plot.
`LocalOutlierFactor.deploySQL`([X])	Returns the SQL code needed to deploy the model.
`LocalOutlierFactor.does_model_exists`(name[, ...])	Checks whether the model is stored in the Vertica database.
`LocalOutlierFactor.drop`()	Drops the model from the Vertica database.
`LocalOutlierFactor.export_models`(name, path)	Exports machine learning models.
`LocalOutlierFactor.fit`(input_relation[, X, ...])	Trains the model.
`LocalOutlierFactor.get_attributes`([attr_name])	Returns the model attributes.
`LocalOutlierFactor.get_match_index`(x, col_list)	Returns the matching index.
`LocalOutlierFactor.get_params`()	Returns the parameters of the model.
`LocalOutlierFactor.get_plotting_lib`([...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
`LocalOutlierFactor.get_vertica_attributes`([...])	Returns the model Vertica attributes.
`LocalOutlierFactor.import_models`(path[, ...])	Imports machine learning models.
`LocalOutlierFactor.predict`()	Creates a `vDataFrame` of the model.
`LocalOutlierFactor.register`(registered_name)	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
`LocalOutlierFactor.set_params`([parameters])	Sets the parameters of the model.
`LocalOutlierFactor.summarize`()	Summarizes the model.
`LocalOutlierFactor.to_binary`(path)	Exports the model to the Vertica Binary format.
`LocalOutlierFactor.to_pmml`(path)	Exports the model to PMML.
`LocalOutlierFactor.to_python`([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
`LocalOutlierFactor.to_sql`([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
`LocalOutlierFactor.to_tf`(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

LocalOutlierFactor.object_type