Skip to main content

Jupyter Notebook

The Jupyter Notebook is an open-source web application that allows you to create and share documents that contain live code, equations, visualizations, and narrative text. Uses include data cleaning and transformation, numerical simulation, statistical modeling, data visualization, machine learning, and much more.

Datagrok lets you create, edit, import, link, and apply Notebooks into tables.

Jupyter Notebook

Create a notebook​

There are several ways to create a new notebook:

  1. From the Top Menu, select ML > Notebooks > New Notebook..
  2. Go to Browse > Platform > Notebooks and click NEW NOTEBOOK from the notebook toolbar.
  3. Run CmdNewNotebook() from the Console.
  4. Select a table and click Actions > Open in Notebooks in the Context Panel.
  5. Right-click the table's header and select Table > Open in Notebook.

If you want to link a notebook to a table, use one of the last two methods.

Notebooks browser​

The Notebooks Browser lists all available notebooks. If a notebook applies to any open table, its context menu shows an Apply to option.

Apply existing notebooks into tables​

To use one of the notebooks available in the platform, open a table and search for applicable notebooks in the Notebooks Browser. Select a notebook and use its context menu to apply it to your data.

Return a table into the platform​

To return a table to the platform, call the grok function:

    grok(< table >)

where table in a Pandas dataframe.

Supported language​

  • Python (3.8)

Environments​

You can run each script in an isolated environment with a predefined package configuration, just like Script Environments. The environment can be specified in the notebook properties.

Importing notebooks​

To import a notebook, just drag-and-drop the corresponding .ipynb file to the platform or use import.

Videos​

Notebooks

See also: