Use Kedro with IPython and Jupyter

This page demonstrates how to use Kedro with IPython, Jupyter Notebook and JupyterLab. We also recommend a video by Data Engineer One and a PyCon presentation by Yetunde Dada that explain the transition from the use of vanilla Jupyter Notebooks to Kedro.

Why use a Notebook?

There are reasons why you may want to use a Notebook, although in general, the principles behind Kedro would discourage their use because they have some drawbacks when used to create production or reproducible code. However, there are occasions when you’d want to put some code into a Notebook, for example:

  • To conduct exploratory data analysis

  • For experimentation as you create new Python functions (that could become Kedro nodes)

  • As a tool for reporting and presentations

Kedro IPython extension

The recommended way to interact with Kedro in IPython and Jupyter is through the Kedro IPython extension, kedro.ipython. An IPython extension is an importable Python module that has a couple of special functions to load and unload it.

The Kedro IPython extension launches a Kedro session and makes available the useful Kedro variables catalog, context, pipelines and session. It also provides the %reload_kedro line magic that reloads these variables (for example, if you need to update catalog following changes to your Data Catalog).

The simplest way to make use of the Kedro IPython extension is through the following commands:

  • kedro ipython. This launches an IPython shell with the extension already loaded and is equivalent to the command ipython --ext kedro.ipython.

  • kedro jupyter notebook. This creates a custom Jupyter kernel that automatically loads the extension and launches Jupyter Notebook with this kernel selected.

  • kedro jupyter lab. This creates a custom Jupyter kernel that automatically loads the extension and launches JupyterLab with this kernel selected.

Running any of the above from within your Kedro project will make the catalog, context, pipelines and session variables immediately accessible to you.


If these variables are not available then Kedro has not been able to load your project. This could be, for example, due to a malformed configuration file or missing dependencies. The full error message is shown on the terminal used to launch kedro ipython, kedro jupyter notebook or kedro jupyter lab. Alternatively, it can be accessed inside the IPython or Jupyter session directly with %reload_kedro.

Managed Jupyter instances

If the above commands are not available to you (e.g. you work in a managed Jupyter service such as a Databricks Notebook) then equivalent behaviour can be achieved by explicitly loading the Kedro IPython extension with the %load_ext line magic:

In [1]: %load_ext kedro.ipython

If your IPython or Jupyter instance was launched from outside your Kedro project then you will need to run a second line magic to set the project path so that Kedro can load the catalog, context, pipelines and session variables:

In [2]: %reload_kedro <project_root>

The Kedro IPython extension remembers the project path so that subsequent calls to %reload_kedro do not need to specify it:

In [1]: %load_ext kedro.ipython
In [2]: %reload_kedro <project_root>
In [3]: %reload_kedro


%reload_kedro accepts optional keyword arguments env and params. For example, to use configuration environment prod you should run %reload_kedro --env=prod. For more details, run %reload_kedro?.

Kedro variables: catalog, context, pipelines and session

The Kedro IPython extension makes the following variables available in your IPython or Jupyter session:

  • catalog (type DataCatalog): Data Catalog instance that contains all defined datasets; this is a shortcut for context.catalog

  • context (type KedroContext): Kedro project context that provides access to Kedro’s library components

  • pipelines (type Dict[str, Pipeline]): Pipelines defined in your pipeline registry

  • session (type KedroSession): Kedro session that orchestrates a pipeline run

We will now give some examples of how these variables can be used in the Iris example project. To explore the full range of attributes and methods available, you might like to consult the relevant API documentation or use the Python dir function (e.g. dir(catalog)).


catalog can be used to explore your Data Catalog, including parameters. Useful methods include catalog.list, catalog.load and For example:

In [1]: catalog.list()
In [2]: catalog.load("example_iris_data")
     sepal_length  sepal_width  petal_length  petal_width    species
0             5.1          3.5           1.4          0.2     setosa
1             4.9          3.0           1.4          0.2     setosa
2             4.7          3.2           1.3          0.2     setosa
3             4.6          3.1           1.5          0.2     setosa
4             5.0          3.6           1.4          0.2     setosa
..            ...          ...           ...          ...        ...
145           6.7          3.0           5.2          2.3  virginica
146           6.3          2.5           5.0          1.9  virginica
147           6.5          3.0           5.2          2.0  virginica
148           6.2          3.4           5.4          2.3  virginica
149           5.9          3.0           5.1          1.8  virginica
In [3]: catalog.load("parameters")
{'example_test_data_ratio': 0.2,
 'example_num_train_iter': 10000,
 'example_learning_rate': 0.01}


If you enable versioning then you can load a particular version of a dataset, e.g. catalog.load("example_train_x", version="2021-12-13T15.08.09.255Z").


context enables you to access Kedro’s library components and project metadata. For example:

In [1]: context.project_path
Out[1]: PosixPath('/Users/username/kedro_projects/iris')


pipelines is a dictionary containing your project’s registered pipelines:

In [1]: pipelines
{'de': Pipeline([
Node(split_data, ['example_iris_data', 'params:example_test_data_ratio'], {'train_x': 'example_train_x', 'train_y': 'example_train_y', 'test_x': 'example_test_x', 'test_y': 'example_test_y'}, 'split')
]), 'ds': Pipeline([
Node(train_model, ['example_train_x', 'example_train_y', 'parameters'], 'example_model', 'train'),
Node(predict, {'model': 'example_model', 'test_x': 'example_test_x'}, 'example_predictions', 'predict'),
Node(report_accuracy, ['example_predictions', 'example_test_y'], None, 'report')
]), '__default__': Pipeline([
Node(split_data, ['example_iris_data', 'params:example_test_data_ratio'], {'train_x': 'example_train_x', 'train_y': 'example_train_y', 'test_x': 'example_test_x', 'test_y': 'example_test_y'}, 'split'),
Node(train_model, ['example_train_x', 'example_train_y', 'parameters'], 'example_model', 'train'),
Node(predict, {'model': 'example_model', 'test_x': 'example_test_x'}, 'example_predictions', 'predict'),
Node(report_accuracy, ['example_predictions', 'example_test_y'], None, 'report')

This can be very useful for exploring your pipelines and the nodes they contain:

In [2]: pipelines["__default__"].all_outputs()

In [3]: [ for node in pipelines["__default__"].nodes]
Out[3]: ['split', 'train', 'predict', 'report']

session allows you to run a pipeline. With no arguments, this will run your __default__ project pipeline in a sequential manner:

In [1]:
2022-03-25 11:20:48,900 - kedro.framework.session.session - INFO - ** Kedro project iris
2022-03-25 11:20:48,911 - - INFO - Loading data from `example_iris_data` (CSVDataSet)...
2022-03-25 11:20:48,915 - - INFO - Loading data from `params:example_test_data_ratio` (MemoryDataSet)...
2022-03-25 11:20:48,915 - kedro.pipeline.node - INFO - Running node: split: split_data([example_iris_data,params:example_test_data_ratio]) -> [example_train_x,example_train_y,example_test_x,example_test_y]
2022-03-25 11:20:49,314 - kedro.runner.sequential_runner - INFO - Completed 4 out of 4 tasks
2022-03-25 11:20:49,314 - kedro.runner.sequential_runner - INFO - Pipeline execution completed successfully.
Out[1]: {}


You can only execute one successful run per session, as there’s a one-to-one mapping between a session and a run. If you wish to do multiple runs, you’ll have to run %reload_kedro to obtain a new session.

You can also specify the following optional arguments for

Argument name

Accepted types




Construct the pipeline using only nodes which have this tag attached. A node is included in the resulting pipeline if it contains any of those tags



An instance of Kedro AbstractRunner. Can be an instance of a ParallelRunner



Run only nodes with specified names



A list of node names which should be used as a starting point



A list of node names which should be used as an end point



A list of dataset names which should be used as a starting point



A list of dataset names which should be used as an end point


Dict[str, str]

A mapping of a dataset name to a specific dataset version (timestamp) for loading. Applies to versioned datasets only



Name of the modular pipeline to run. Must be one of those returned by the register_pipelines function in src/<package_name>/

Kedro and Jupyter

We recommend that you store your Jupyter Notebooks in the notebooks folder of your Kedro project. If you are using kedro jupyter notebook or kedro jupyter lab then you should use the default kernel selected for you, which is listed as Kedro (<package_name>). This will run the Kedro IPython extension automatically when the kernel is started, so that the catalog, context, pipelines and session variables are available immediately to you.


Restarting the kernel will reload the Kedro IPython extension and hence refresh the catalog, context, pipelines and session variables.

For example, to create a new Notebook in the Iris example project use the following button:

Make a new Notebook using the Kedro (iris) kernel

From inside a Notebook, the kernel selected appears as follows:

Select the Kedro (iris) kernel inside a Notebook

If you are not able to execute kedro jupyter notebook or kedro jupyter lab then follow the instructions on managed Jupyter instances to manually load the Kedro IPython extension.

Manage Jupyter kernels

Behind the scenes, the kedro jupyter notebook and kedro jupyter lab commands create a Jupyter kernel named kedro_<package_name>. This kernel is identical to the default IPython kernel but with a slightly customised kernel specification that automatically loads kedro.ipython when the kernel is started. The kernel specification is installed at a user level rather than system-wide.


If a Jupyter kernel with the name kedro_<package_name> already exists then it is replaced. This ensures that the kernel always points to the correct Python executable. For example, if you change conda environment in a Kedro project then you should re-run kedro jupyter notebook/lab to replace the kernel specification with one that points to the new environment.

As each Kedro project has its own Jupyter kernel, you can switch between multiple Kedro projects from a single Jupyter instance simply by selecting the appropriate kernel.

You can use the jupyter kernelspec set of commands to manage your Jupyter kernels. For example, to remove the kernel created as part of the Iris example project, you would run jupyter kernelspec remove kedro_iris.

Use an alternative Jupyter client

In addition to Jupyter Notebook and JupyterLab, you can use any other Jupyter client to connect to a Kedro project kernel. For example, the Qt Console can be launched using the kedro_iris kernel as follows:

jupyter qtconsole --kernel=kedro_iris

This will automatically load the Kedro IPython in a console that supports graphical features such as embedded figures: Plot of example iris data in a Qt Console

Convert functions from Jupyter Notebooks into Kedro nodes

Built into the Kedro Jupyter workflow is the ability to convert multiple functions defined in the Jupyter Notebook(s) into Kedro nodes. You need a single CLI command.

Here is how it works:

  • Start a Jupyter Notebook session: kedro jupyter notebook

  • Create a new Notebook and paste the following code into the first cell:

def some_action():
    print("This function came from `notebooks/my_notebook.ipynb`")
  • Enable tags toolbar: View menu -> Cell Toolbar -> Tags Enable the tags toolbar graphic

  • Add the node tag to the cell containing your function Add the node tag graphic


The Notebook can contain multiple functions tagged as node, each of them will be exported into the resulting Python file

  • Save your Jupyter Notebook to notebooks/my_notebook.ipynb

  • Run kedro jupyter convert notebooks/my_notebook.ipynb from the terminal to create a Python file src/<package_name>/nodes/ containing some_action function definition


You can also convert all your Notebooks at once by calling kedro jupyter convert --all.

  • The some_action function can now be used in your Kedro pipelines

Kedro-Viz line magic

If you have Kedro-Viz installed then you can display an interactive visualisation of your pipeline directly in your Notebook using the line magic %run_viz. You should see a visualisation like the following:

 line magic in a Notebook