Feedback

×

• Feedback
• Was this page helpful?*
  • Yes
  • No
• Do you have any additional feedback?
  • Yes
  • No
• Additional Feedback

Decomposition & Preprocessing

---

Decomposition

PCA

decomposition.PCA([name, overwrite_model, ...])

Creates a PCA (Principal Component Analysis) object using the Vertica PCA algorithm.

Methods:

PCA.contour([nbins, chart])	Draws the model's contour plot.
PCA.deployInverseSQL([key_columns, ...])	Returns the SQL code needed to deploy the inverse model.
PCA.deploySQL([X, n_components, cutoff, ...])	Returns the SQL code needed to deploy the model.
PCA.does_model_exists(name[, raise_error, ...])	Checks whether the model is stored in the Vertica database.
PCA.drop()	Drops the model from the Vertica database.
PCA.export_models(name, path[, kind])	Exports machine learning models.
PCA.fit(input_relation[, X, return_report])	Trains the model.
PCA.get_attributes([attr_name])	Returns the model attributes.
PCA.get_match_index(x, col_list[, str_check])	Returns the matching index.
PCA.get_params()	Returns the parameters of the model.
PCA.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
PCA.get_vertica_attributes([attr_name])	Returns the model Vertica attributes.
PCA.import_models(path[, schema, kind])	Imports machine learning models.
PCA.inverse_transform(vdf[, X])	Applies the Inverse Model on a vDataFrame.
PCA.plot([dimensions, chart])	Draws a decomposition scatter plot.
PCA.plot_circle([dimensions, chart])	Draws a decomposition circle.
PCA.plot_scree([chart])	Draws a decomposition scree plot.
PCA.register(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
PCA.score([X, input_relation, metric, p])	Returns the decomposition score on a dataset for each transformed column.
PCA.set_params([parameters])	Sets the parameters of the model.
PCA.summarize()	Summarizes the model.
PCA.to_binary(path)	Exports the model to the Vertica Binary format.
PCA.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
PCA.to_pmml(path)	Exports the model to PMML.
PCA.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
PCA.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
PCA.to_tf(path)	Exports the model to the Frozen Graph format (TensorFlow).
PCA.transform([vdf, X, n_components, cutoff])	Applies the model on a vDataFrame.

Attributes:

PCA.object_type

SVD

decomposition.SVD([name, overwrite_model, ...])

Creates an SVD (Singular Value Decomposition) object using the Vertica SVD algorithm.

Methods:

SVD.contour([nbins, chart])	Draws the model's contour plot.
SVD.deployInverseSQL([key_columns, ...])	Returns the SQL code needed to deploy the inverse model.
SVD.deploySQL([X, n_components, cutoff, ...])	Returns the SQL code needed to deploy the model.
SVD.does_model_exists(name[, raise_error, ...])	Checks whether the model is stored in the Vertica database.
SVD.drop()	Drops the model from the Vertica database.
SVD.export_models(name, path[, kind])	Exports machine learning models.
SVD.fit(input_relation[, X, return_report])	Trains the model.
SVD.get_attributes([attr_name])	Returns the model attributes.
SVD.get_match_index(x, col_list[, str_check])	Returns the matching index.
SVD.get_params()	Returns the parameters of the model.
SVD.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
SVD.get_vertica_attributes([attr_name])	Returns the model Vertica attributes.
SVD.import_models(path[, schema, kind])	Imports machine learning models.
SVD.inverse_transform(vdf[, X])	Applies the Inverse Model on a vDataFrame.
SVD.plot([dimensions, chart])	Draws a decomposition scatter plot.
SVD.plot_circle([dimensions, chart])	Draws a decomposition circle.
SVD.plot_scree([chart])	Draws a decomposition scree plot.
SVD.register(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
SVD.score([X, input_relation, metric, p])	Returns the decomposition score on a dataset for each transformed column.
SVD.set_params([parameters])	Sets the parameters of the model.
SVD.summarize()	Summarizes the model.
SVD.to_binary(path)	Exports the model to the Vertica Binary format.
SVD.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
SVD.to_pmml(path)	Exports the model to PMML.
SVD.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
SVD.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
SVD.to_tf(path)	Exports the model to the Frozen Graph format (TensorFlow).
SVD.transform([vdf, X, n_components, cutoff])	Applies the model on a vDataFrame.

Attributes:

SVD.object_type

MCA (Beta)

decomposition.MCA([name, overwrite_model])

Creates a MCA (multiple correspondence analysis) object using the Vertica PCA algorithm.

Methods:

MCA.contour([nbins, chart])	Draws the model's contour plot.
MCA.deployInverseSQL([key_columns, ...])	Returns the SQL code needed to deploy the inverse model.
MCA.deploySQL([X, n_components, cutoff, ...])	Returns the SQL code needed to deploy the model.
MCA.does_model_exists(name[, raise_error, ...])	Checks whether the model is stored in the Vertica database.
MCA.drop()	Drops the model from the Vertica database.
MCA.export_models(name, path[, kind])	Exports machine learning models.
MCA.fit(input_relation[, X, return_report])	Trains the model.
MCA.get_attributes([attr_name])	Returns the model attributes.
MCA.get_match_index(x, col_list[, str_check])	Returns the matching index.
MCA.get_params()	Returns the parameters of the model.
MCA.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
MCA.get_vertica_attributes([attr_name])	Returns the model Vertica attributes.
MCA.import_models(path[, schema, kind])	Imports machine learning models.
MCA.inverse_transform(vdf[, X])	Applies the Inverse Model on a vDataFrame.
MCA.plot([dimensions, chart])	Draws a decomposition scatter plot.
MCA.plot_circle([dimensions, chart])	Draws a decomposition circle.
MCA.plot_contrib([dimension, chart])	Draws a decomposition contribution plot of the input dimension.
MCA.plot_cos2([dimensions, chart])	Draws a MCA (multiple correspondence analysis) cos2 plot of the two input dimensions.
MCA.plot_scree([chart])	Draws a decomposition scree plot.
MCA.plot_var([dimensions, method, chart])	Draws the MCA (multiple correspondence analysis) graph.
MCA.register(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
MCA.score([X, input_relation, metric, p])	Returns the decomposition score on a dataset for each transformed column.
MCA.set_params([parameters])	Sets the parameters of the model.
MCA.summarize()	Summarizes the model.
MCA.to_binary(path)	Exports the model to the Vertica Binary format.
MCA.to_memmodel()	Converts the model to an InMemory object that can be used for different types of predictions.
MCA.to_pmml(path)	Exports the model to PMML.
MCA.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
MCA.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
MCA.to_tf(path)	Exports the model to the Frozen Graph format (TensorFlow).
MCA.transform([vdf, X, n_components, cutoff])	Applies the model on a vDataFrame.

Attributes:

MCA.object_type

---

Preprocessing

One-Hot Encoder

preprocessing.OneHotEncoder([name, ...])

Creates a Vertica OneHotEncoder object.

Methods:

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

Attributes:

OneHotEncoder.object_type

---

Scaler

preprocessing.Scaler([name, ...])

Creates a Vertica Scaler object.

Methods:

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

Attributes:

Scaler.object_type

Standard Scaler

preprocessing.StandardScaler([name, ...])

i.e. Scaler with param method = 'zscore'.

Methods:

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

Attributes:

StandardScaler.object_type

Min Max Scaler

preprocessing.MinMaxScaler([name, ...])

i.e. Scaler with param method = 'minmax'.

Methods:

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

Attributes:

MinMaxScaler.object_type

Robust Scaler

preprocessing.RobustScaler([name, ...])

i.e. Scaler with param method = 'robust_zscore'.

Methods:

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

Attributes:

RobustScaler.object_type

---

Balance

preprocessing.balance(name, input_relation, y)

Creates a view with an equal distribution of the input data based on the response_column.

---

Density Estimation

Kernel Density (Beta)

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

[Beta Version] Creates a KernelDensity object.

Methods:

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

Attributes:

KernelDensity.object_type

---