JupyterLab Extension for dependency management and optimization
This is a JupyterLab extension for dependency management and optimization created to guarantee reproducibility of Jupyter notebooks.
This extension provides management of dependencies for JupyterLab notebooks.
The main goals of the project are the following:
manage notebook requirements without leaving the notebook
provide a unique and optimized environment for each notebook
NOTE: The requirements are optimized using the Thoth resolution engine.
This extension requires:
JupyterLab >= 3.1.0
The extension can be installed via pip or Pipenv from PyPI:
pip install jupyterlab-requirements
You can initialize JupyterLab and start using it.
If you are seeing the frontend extension, but it is not working, check that the server extension is enabled:
jupyter server extension list
If the server extension is installed and enabled, but you are not seeing the frontend extension, check the frontend extension is installed:
jupyter labextension list
There are 3 ways to interact with this extension:
%horus magic commands directly in your notebook’s cells.
horus CLI from terminal or integrated in pipelines.
jupyterlab-requirements UI accessible through Manage Dependencies button that appears in the notebook when it is opened in JupyteLab.
There are currently two resolution engines available in the extension:
NOTE: Thoth is used by default and Pipenv can be triggered with flags or run as backup automatically.
NOTE: Currently this extension supports only Python kernels.
Using the Thoth resolution engine you can request an optimized software stack that satisfies your requirements.
Thoth: recommendation type
You can choose the type of recommendation that better fits your needs:
You can find more information and updates here.
Thoth: runtime environment
Thoth resolution engine is able to provide an optimized software stack based on the runtime environment you are using (several inputs are used, if you want to know more, have a look here here).
In general different runtime environment will provide different effect on you application (e.g. more performance), therefore we include these information in the notebook metadata so that other can find out what runtime environment has been used to run a certain notebook.
You can select the runtime environment to be used for the recommendation selecting:
Operating System Name
Operating System Version
Python Interpreter Version
NOTE: Those parameters are autodiscovered by the extension and assigned to your environment, you can customize them if you are interested.
Once lock file is created using any of available resolution engines, the dependencies will be installed in the virtualenv using micropipenv.
Virtual environment for you dependencies
The virtual environment created and assigned to the kernel to be used for your notebook according to your dependencies requirements can be checked using the following command from a terminal:
The dependencies stored in the notebook metadata can be also stored into overlays folder using the kernel name by default. If you want to know more about the use of overlays, have a look micropipenv. If you want to see a practical example on the use of overlays and how to create them from your notebook, you can check this tutorial.
If you have too many kernels, you can remove them directly from the JupyterLab menu under Kernel Section. This plugin is provided from this extension.
You can use this extension for each of your notebook to guarantee they have the correct dependencies files required for reproducibility and shareability. In this way, all the dependencies information required to repeat the environment are shipped with the notebook. In the notebook metadata you will find:
All packages required (direct dependencies).
All packages (direct and transitive dependencies) locked with all hashes (Pipfile.lock).
dependency resolution engine
Currently two resolution engine are enabled.
Only for Thoth resolution engine.
All this information can allow reproducibility of the notebook.
NOTE: You will need NodeJS to build the extension package.
The jlpm command is JupyterLab’s pinned version of yarn that is installed with JupyterLab. You may use yarn or npm in lieu of jlpm below.
Fork this repository
Clone the origin repo to your local environment
git clone email@example.com:thoth-station/jupyterlab-requirements.git
Change directory to the jupyterlab-requirements directory
Create new virtualenv
pipenv install --dev
All following commands needs to be run from the virtualenv created and accessed with command in point 5.
Install package in development mode
pip install -ve . --no-cache-dir
Link your development version of the extension with JupyterLab
jupyter labextension develop . --overwrite
jupyter serverextension enable --py jupyterlab-requirements --sys-prefix
Rebuild extension Typescript source after making changes
jlpm run build
You can watch the source directory and run JupyterLab at the same time in different terminals to watch for changes in the extension’s source and automatically rebuild the extension.
The following command watch the source directory in one terminal, automatically rebuilding when needed.
jlpm run watch
The following command run JupyterLab in another terminal.
With the watch command running, every saved change will immediately be built locally and available in your running JupyterLab. Refresh JupyterLab to load the change in your browser (you may need to wait several seconds for the extension to be rebuilt).
By default, the jlpm run build command generates the source maps for this extension to make it easier to debug using the browser dev tools. To also generate source maps for the JupyterLab core extensions, you can run the following command:
jupyter lab build --minimize=False
Run tests using the command
python3 setup.py test
When all tests passed and you are ready with a new contribution open a Pull Request!! We are very happy to receive contributions from the community!
Demo development status and new features
pip uninstall jupyterlab-requirements
Release history Release notifications | RSS feed
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Hashes for jupyterlab_requirements-0.16.2.tar.gz
Hashes for jupyterlab_requirements-0.16.2-py3-none-any.whl