For the latest version of JMP Help, visit JMP.com/help.


Predictive and Specialized Modeling > Formula Depot > Formula Depot Platform Options
Publication date: 11/10/2021

Image shown hereFormula Depot Platform Options

Add Formula From Column

Enables you to add existing prediction formulas from data table columns to the current formula depot. If your table has only one formula, that formula is added to the depot. If you have multiple column formulas, a dialog box enables you to select the formulas of interest. If your formula columns are selected in your data table, they will be pre-selected in the dialog box.

Show Scripts

Opens a new Formula Window (or appends to an open Formula Window) that contains scripts for all of the formulas that are in the current formula depot.

Copy Scripts

Copies all scripts from the current formula depot to the clipboard.

Copy Formulas as Functions

Enables you to select models from the current formula depot to be copied as functions for use in JSL. Selected models are copied onto the clipboard. The functions act on scalar variables rather than columns. One Namespace is generated for each selected model. Each Namespace has one function, called predict, that takes the model inputs as parameters and returns the model predictor.

Copy Formulas as Transforms

Enables you to select models from the current formula depot to be copied. Selected models are copied onto the clipboard within a Transform Columns() statement.

Run Scripts

Enables you to save models from the current formula depot to new columns in your JMP data table. Formulas with intermediate column formulas will add the intermediate columns to your JMP data table.

Note: If you have more than one data table open, a Choose Table window appears that enables you to specify which data table to use for Run Scripts.

Generate C Code, Generate Python Code, Generate JavaScript Code, Generate SAS Code, and Generate SQL Code

Enables you to select models from the current formula depot for code generation. A new script window appears that contains scoring code for the selected models in C, Python, JavaScript, SAS DS2, or SQL, respectively. You can use this code to facilitate the deployment of the model in the environment or framework of your choice. See Generating Scoring Code from the Formula Depot Platform.

Model Comparison

Enables you to select models from the current formula depot to be compared using the model comparison utility. If models from multiple tables are stored in the depot, first select the table of interest and then the models of interest. Hold the shift key when selecting the Model Comparison option to launch the Model Comparison in a new window. See Model Comparison.

Note: If you have more than one data table open, a Choose Table window appears that enables you to specify which data table to use for the Model Comparison.

Remove Model Comparison

Removes all Model Comparison reports from the current formula depot.

Profiler

Enables you to select models from the current formula depot to be profiled using the profiler. If models from multiple tables are stored in the depot you first select the table of interest and then the models of interest. Hold the shift key when selecting the profiler option to launch the profiler in a new window. See Profiler in Profilers.

Note: If you have more than one data table open, a Choose Table window appears that enables you to specify which data table to use for the Profiler.

Remove Profiler

Removes all Profilers from the current formula depot.

See Redo Menus in Reports and Save Script Menus in Reports in Using JMP for more information about the following options:

Redo

Contains options that enable you to repeat or relaunch the analysis. In platforms that support the feature, the Automatic Recalc option immediately reflects the changes that you make to the data table in the corresponding report window.

Save Script

Contains options that enable you to save a script that reproduces the report to several destinations.

Want more information? Have questions? Get answers in the JMP User Community (community.jmp.com).