Jupyter Viewer
A revisited NbViewer as a modern Web application to view Jupyter notebooks.
NbViewer is built on top of the GitHub nbviewer repository and has been useful for many data scientists since years.
Jupyter Viewer takes over the concept of easy visualisation and brings it to the Web application developers with React.js components to render Notebooks without any Kernel. It is also available as JupyterLab extension.
You can try on https://viewer.datalayer.tech.
The documentation lives on https://jupyter-viewer.datalayer.tech.
If needed, Users can connect the static view to a Kernel and make it executable.
Usage
This repository packages the Jupyter UI Viewer component as a JupyterLab extension. Install and launch with the following commands (you will need Python installed on your machine).
pip install jupyter_viewer
jupyter viewer
After launching, enjoy the 👀 views on http://localhost:8888/jupyter_viewer (served by Jupyter).
You can view a Notebook hosted on GitHub using the following pattern: http://localhost:8888/github/account/repo/branch/path
# For example...
open http://localhost:8888/jupyter_viewer/github/datalayer/examples/main/pytorch-gpu/pytorch-gpu-example.ipynb
Develop
Develop the Web application.
yarn
yarn start
# Browse the Webpack server.
open http://localhost:3063
open http://localhost:3063/github/datalayer/examples/main/pytorch-gpu/pytorch-gpu-example.ipynb
Develop the Jupyter Server
pip install -e .[test]
jupyter viewer
# Browse the Jupyter Server.
open http://localhost:8888/jupyter_viewer
open http://localhost:8888/jupyter_viewer/github/datalayer/examples/main/pytorch-gpu/pytorch-gpu-example.ipynb
Develop the JupyterLab extension.
pip install -e .[test]
jupyter labextension develop . --overwrite
jupyter labextension list
jupyter server extension list
yarn jupyterlab
# Browse JupyterLab.
open http://localhost:8888
Publish
make publish
open https://viewer.datalayer.tech
open https://viewer.datalayer.tech/github/datalayer/examples/main/pytorch-gpu/pytorch-gpu-example.ipynb
Releases
Jupyter Viewer is released as a python package in PyPI.