Feedback

×

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

ML Ops

---

Model Versioning

model_versioning.RegisteredModel(registered_name)

Creates a RegisteredModel object to load and use a registered model.

Methods:

RegisteredModel.change_status(version, ...)	Changes the status of a model.
RegisteredModel.list_models()	Returns the list of the registered models with the same name that the user has USAGE privilege on.
RegisteredModel.list_status_history([version])	Returns the model status-change history.
RegisteredModel.predict(vdf[, X, name, ...])	Makes predictions on the input relation.
RegisteredModel.predict_proba(vdf[, X, ...])	Returns the model's probabilities using the input relation.

---

Model Tracking

model_tracking.vExperiment(experiment_name, ...)

Creates a vExperiment object that can be used for tracking native vertica models trained as part of an experiment.

Methods:

vExperiment.add_model(model[, metrics])	Adds a model to the experiment.
vExperiment.drop([keeping_models])	Drops all models of the experiment except those in the keeping_models list.
vExperiment.get_plotting_lib([class_name, ...])	Returns the first available library (Plotly, Matplotlib, or Highcharts) to draw a specific graphic.
vExperiment.list_models()
vExperiment.load_best_model(metric)
vExperiment.plot(parameter, metric[, chart])	Draws the scatter plot of a metric vs a parameter