Working with models

Written by

Updated at March 6, 2025

Models are designed for storing and exporting computation results in DataSphere projects.

Creating a model

Select the relevant project in your community or on the DataSphere homepage in the Recent projects tab.
Open the project.
Select or create a notebook with the code that will create a model variable in RAM.
Run the code in the notebook using the selected configuration.
In the right-hand panel, select . In the window that opens, click Create model.
Select the name of the variable to create your model from.
Give your model a name and add a description if needed.
Click Create.

The new model will appear in the list of available project models. Creating a model may take some time.

You can load the saved model to the notebook's interpreter state and use it in your computations:

Select the relevant project in your community or on the DataSphere homepage in the Recent projects tab.
Open the project.
In the right-hand panel, select . In the window that opens, select the model to load and click .
Select Load.
Specify the name of the variable to load the model to and click Load.

Once loaded, the model will be available in the code in the variable you specified.

You can delete a model in the JupyterLab interface or on the project page under Resources.

JupyterLab

Project resources

Open the JupyterLab interface and select in the right-hand panel.
In the list of models, choose the one you want to delete, click , and select Delete.
Click Confirm.

Select the relevant project in your community or on the DataSphere homepage in the Recent projects tab.
Under Project resources, click Model.
In the list of models, select the one you want to delete.
On the view model page, click More and select Delete.
Click Confirm.

Warning

The actual deletion of resources can take up to 72 hours.