Jupyter Notebook


Jupyhai is the Valohai Jupyter notebook extension. It enables you to use Jupyter notebooks easily with the Valohai platform. It is an optional way of interacting with projects in the same capacity as our web user interface, API and CLI.

1. Requirements

To use jupyter notebook extension, you will need:

2. Installation

Pull the valohai/jupyhai docker image from DockerHub.

docker pull valohai/jupyhai

3. Notebook server

Run the notebook server in port 8888.

docker run -p 8888:8888 -v "$PWD":/home/jovyan/work valohai/jupyhai

Open in browser:

Your current working directory will be mapped into the container.

  • All the files in your current working directory will be available within the container
  • All changes within the mapped folder will persist after shutting down

4. Create notebook

Creating new notebook

Create new notebook in the /work folder. This folder is mapped to your local filesystem.

5. Login

Login to Valohai

Press the Valohai button in the toolbar and login using your Valohai credentials.

6. Settings

Valohai execution settings

Press the Valohai button in the toolbar and go to settings window.

Select the following:

  • Project: Valohai project where the executions will be version controlled
  • Environment: Environment type for the cloud executions (E.g. AWS p2.xlarge)
  • Docker Image: Docker image that provides the required libraries (E.g. Tensorflow)

These are the same settings you would choose when using Valohai website, CLI or valohai.yaml.

Once you are happy with your selections. Press save.

7. Create execution

Running Valohai execution

Press the Valohai button in the toolbar and select Create execution.

The gizmo for the new execution will appear to the right.

8. Get results

Get results back from Valohai

Each gizmo on the right side signifies a single Valohai execution. Click #1 and then click Notebook button.

This will download the finished notebook back to your local machine and open it.

9. Parameterize notebook

Adding hyperparameter

Parameterizing notebook happens using tags. Tags are Jupyter notebook feature that lets tag a cell.

Here we will mark the first cell with parameters tag, which means all variables are considered as Valohai parameters, just like in the valohai.yaml.

10. Inputs

Adding parameterized input

Here we marked the first cell with inputs tag and ran it in Valohai.

All the variables in this cell will be considered as Valohai input URIs for the execution, just like in the valohai.yaml.

11. Reusing parameterized notebook

Adding hyperparameter

Now you can run notebook based experiments without a notebook!

Because the learning_rate here is parameterized, you can set it via Valohai UI and run experiments without opening a visible notebook.


When I try to download my outputs back from finished execution, I get 404: Not Found

Always use instead of http://localhost:8888