Feedback

×

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

Time Series

Multi-Timeseries Model (Beta)

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

This model is built based on multiple base models.

Methods:

TimeSeriesByCategory.contour([nbins, chart])	Draws the model's contour plot.
TimeSeriesByCategory.deploySQL([vdf, ts, y, ...])	Returns the SQL code needed to deploy the model.
TimeSeriesByCategory.does_model_exists(name)	Checks whether the model is stored in the Vertica database.
TimeSeriesByCategory.drop()	Drops the model from the Vertica database.
TimeSeriesByCategory.export_models(name, path)	Exports machine learning models.
TimeSeriesByCategory.features_importance([...])	Computes the input submodel's features importance.
TimeSeriesByCategory.fit(input_relation, ts, ...)	Trains the model.
TimeSeriesByCategory.get_attributes([attr_name])	Returns the model attributes.
TimeSeriesByCategory.get_match_index(x, col_list)	Returns the matching index.
TimeSeriesByCategory.get_params()	Returns the parameters of the model.
TimeSeriesByCategory.get_plotting_lib([...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
TimeSeriesByCategory.get_vertica_attributes([...])	Returns the model Vertica attributes.
TimeSeriesByCategory.import_models(path[, ...])	Imports machine learning models.
TimeSeriesByCategory.plot([idx, vdf, ts, y, ...])	Draws the input submodel.
TimeSeriesByCategory.predict([vdf, ts, y, ...])	Predicts using the input relation.
TimeSeriesByCategory.register(registered_name)	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
TimeSeriesByCategory.regression_report([...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
TimeSeriesByCategory.report([metrics, ...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
TimeSeriesByCategory.score([metric, start, ...])	Computes the model score.
TimeSeriesByCategory.set_params([parameters])	Sets the parameters of the model.
TimeSeriesByCategory.summarize()	Summarizes the model.
TimeSeriesByCategory.to_binary(path)	Exports the model to the Vertica Binary format.
TimeSeriesByCategory.to_pmml(path)	Exports the model to PMML.
TimeSeriesByCategory.to_python([...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
TimeSeriesByCategory.to_sql([X, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
TimeSeriesByCategory.to_tf(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

TimeSeriesByCategory.object_type

---

ARIMA

tsa.ARIMA([name, overwrite_model, order, ...])

Creates a inDB ARIMA model.

Methods:

ARIMA.contour([nbins, chart])	Draws the model's contour plot.
ARIMA.deploySQL([ts, y, start, ...])	Returns the SQL code needed to deploy the model.
ARIMA.does_model_exists(name[, raise_error, ...])	Checks whether the model is stored in the Vertica database.
ARIMA.drop()	Drops the model from the Vertica database.
ARIMA.export_models(name, path[, kind])	Exports machine learning models.
ARIMA.features_importance([show, chart])	Computes the model's features importance.
ARIMA.fit(input_relation, ts, y[, ...])	Trains the model.
ARIMA.get_attributes([attr_name])	Returns the model attributes.
ARIMA.get_match_index(x, col_list[, str_check])	Returns the matching index.
ARIMA.get_params()	Returns the parameters of the model.
ARIMA.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
ARIMA.get_vertica_attributes([attr_name])	Returns the model Vertica attributes.
ARIMA.import_models(path[, schema, kind])	Imports machine learning models.
ARIMA.plot([vdf, ts, y, start, ...])	Draws the model.
ARIMA.predict([vdf, ts, y, start, ...])	Predicts using the input relation.
ARIMA.register(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
ARIMA.regression_report([metrics, start, ...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
ARIMA.report([metrics, start, npredictions, ...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
ARIMA.score([metric, start, npredictions, ...])	Computes the model score.
ARIMA.set_params([parameters])	Sets the parameters of the model.
ARIMA.summarize()	Summarizes the model.
ARIMA.to_binary(path)	Exports the model to the Vertica Binary format.
ARIMA.to_pmml(path)	Exports the model to PMML.
ARIMA.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
ARIMA.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
ARIMA.to_tf(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

ARIMA.object_type

---

ARMA

tsa.ARMA([name, overwrite_model, order, ...])

Creates a inDB ARMA model.

Methods:

ARMA.contour([nbins, chart])	Draws the model's contour plot.
ARMA.deploySQL([ts, y, start, npredictions, ...])	Returns the SQL code needed to deploy the model.
ARMA.does_model_exists(name[, raise_error, ...])	Checks whether the model is stored in the Vertica database.
ARMA.drop()	Drops the model from the Vertica database.
ARMA.export_models(name, path[, kind])	Exports machine learning models.
ARMA.features_importance([show, chart])	Computes the model's features importance.
ARMA.fit(input_relation, ts, y[, ...])	Trains the model.
ARMA.get_attributes([attr_name])	Returns the model attributes.
ARMA.get_match_index(x, col_list[, str_check])	Returns the matching index.
ARMA.get_params()	Returns the parameters of the model.
ARMA.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
ARMA.get_vertica_attributes([attr_name])	Returns the model Vertica attributes.
ARMA.import_models(path[, schema, kind])	Imports machine learning models.
ARMA.plot([vdf, ts, y, start, npredictions, ...])	Draws the model.
ARMA.predict([vdf, ts, y, start, ...])	Predicts using the input relation.
ARMA.register(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
ARMA.regression_report([metrics, start, ...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
ARMA.report([metrics, start, npredictions, ...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
ARMA.score([metric, start, npredictions, method])	Computes the model score.
ARMA.set_params([parameters])	Sets the parameters of the model.
ARMA.summarize()	Summarizes the model.
ARMA.to_binary(path)	Exports the model to the Vertica Binary format.
ARMA.to_pmml(path)	Exports the model to PMML.
ARMA.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
ARMA.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
ARMA.to_tf(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

ARMA.object_type

---

AR

tsa.AR([name, overwrite_model, p, method, ...])

Creates a inDB Autoregressor model.

Methods:

AR.contour([nbins, chart])	Draws the model's contour plot.
AR.deploySQL([ts, y, start, npredictions, ...])	Returns the SQL code needed to deploy the model.
AR.does_model_exists(name[, raise_error, ...])	Checks whether the model is stored in the Vertica database.
AR.drop()	Drops the model from the Vertica database.
AR.export_models(name, path[, kind])	Exports machine learning models.
AR.features_importance([show, chart])	Computes the model's features importance.
AR.fit(input_relation, ts, y[, ...])	Trains the model.
AR.get_attributes([attr_name])	Returns the model attributes.
AR.get_match_index(x, col_list[, str_check])	Returns the matching index.
AR.get_params()	Returns the parameters of the model.
AR.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
AR.get_vertica_attributes([attr_name])	Returns the model Vertica attributes.
AR.import_models(path[, schema, kind])	Imports machine learning models.
AR.plot([vdf, ts, y, start, npredictions, ...])	Draws the model.
AR.predict([vdf, ts, y, start, ...])	Predicts using the input relation.
AR.register(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
AR.regression_report([metrics, start, ...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
AR.report([metrics, start, npredictions, method])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
AR.score([metric, start, npredictions, method])	Computes the model score.
AR.set_params([parameters])	Sets the parameters of the model.
AR.summarize()	Summarizes the model.
AR.to_binary(path)	Exports the model to the Vertica Binary format.
AR.to_pmml(path)	Exports the model to PMML.
AR.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
AR.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
AR.to_tf(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

AR.object_type

---

MA

tsa.MA([name, overwrite_model, q, penalty, ...])

Creates a inDB Moving Average model.

Methods:

MA.contour([nbins, chart])	Draws the model's contour plot.
MA.deploySQL([ts, y, start, npredictions, ...])	Returns the SQL code needed to deploy the model.
MA.does_model_exists(name[, raise_error, ...])	Checks whether the model is stored in the Vertica database.
MA.drop()	Drops the model from the Vertica database.
MA.export_models(name, path[, kind])	Exports machine learning models.
MA.features_importance([show, chart])	Computes the model's features importance.
MA.fit(input_relation, ts, y[, ...])	Trains the model.
MA.get_attributes([attr_name])	Returns the model attributes.
MA.get_match_index(x, col_list[, str_check])	Returns the matching index.
MA.get_params()	Returns the parameters of the model.
MA.get_plotting_lib([class_name, chart, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
MA.get_vertica_attributes([attr_name])	Returns the model Vertica attributes.
MA.import_models(path[, schema, kind])	Imports machine learning models.
MA.plot([vdf, ts, y, start, npredictions, ...])	Draws the model.
MA.predict([vdf, ts, y, start, ...])	Predicts using the input relation.
MA.register(registered_name[, raise_error])	Registers the model and adds it to in-DB Model versioning environment with a status of 'under_review'.
MA.regression_report([metrics, start, ...])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
MA.report([metrics, start, npredictions, method])	Computes a regression report using multiple metrics to evaluate the model (r2, mse, max error...).
MA.score([metric, start, npredictions, method])	Computes the model score.
MA.set_params([parameters])	Sets the parameters of the model.
MA.summarize()	Summarizes the model.
MA.to_binary(path)	Exports the model to the Vertica Binary format.
MA.to_pmml(path)	Exports the model to PMML.
MA.to_python([return_proba, ...])	Returns the Python function needed for in-memory scoring without using built-in Vertica functions.
MA.to_sql([X, return_proba, ...])	Returns the SQL code needed to deploy the model without using built-in Vertica functions.
MA.to_tf(path)	Exports the model to the Frozen Graph format (TensorFlow).

Attributes:

MA.object_type