I have successfully rendered folium maps on my jupyter notebook by increasing the data limit when launching jupyter from anaconda prompt like this:
"jupyter notebook --NotebookApp.iopub_data_rate_limit=1.0e10"
However, when I have uploaded the notebook to Github, rather than the map I had processed, I got a blank window.
How can I render these maps? Is it possible to increase the data_rate_limit on the notebook when it is being launched through github?
If there's any Javascript in Folium maps, it will not render on GitHub. For security reasons, GitHub does not execute any javascript while rendering notebooks.
From GitHub docs,
The interactive features of the notebook, such as custom JavaScript
plots, will not work in your repository on GitHub
Related
I have quite a large jupyter notebook I'm working on in VScode, while using the high performance computing facility (connected through SSH).
I am trying to create some internal hyperlinks so I can jump from my table of contents to a specific section in the notebook. However, this does not work.
Code I'm using to create the link:
[Summary](#Summary)
some other cells
## Summary
It does create a link, but clicking on it does not work in VScode. This one keeps me right at the top of the notebook, other links to sections further down bring me to seemingly random cells down my file. Also, when opening the notebook in Github, clicking the link brings me back to the folder where my notebook is located.
The same construction does work both in a local jupyter instance, and in vscode, when not connected to the HPC.
One difference I can think of is that I use miniconda on the HPC, and the full anaconda installation locally.
Are the links not supported in miniconda? Or is it something else?
I have multiple people working on the same AWS EMR cluster to run some Spark jobs. This is being done through Jupyter Notebooks which are created/modified using the Jupyter extension installed on a SSH Target through vscode. The modules are installed on the base conda environment that is included with the /emr/notebook-env/. Some people can see the correct kernel that is associated with the base conda environment in their vscode window when working on notebooks. However, some don't see this kernel as an option. How do I make sure that everyone's vscode lists the appropriate kernel when they are creating new notebooks or modifying existing notebooks?
Another potential reason this could happen is that the Jupyter exentsion of VSCode is not installed.
To add the Jupyter extension for VSCode, search for it by clicking extensions icon in lefthand toolbar, then searching for Jupyter and installing.
The user having the issue had to update their vscode and that fixed the issue
I uploaded my first Jupyter Notebook to Github here:
https://github.com/Athelian/OkCupid/blob/master/OkC%20Analysis.ipynb
But I cannot figure out how to get syntax highlighting to display in the preview after clicking the link. I've noticed others don't have this issue when they upload .ipynb files.
Is there any way to get it display like a real notebook?
What is the difference between JupyterLab and JupyterHub? How should I decide which one to deploy for users?
JupyterHub is encapsulated environments for multiple users.
JupyterLab is a new version (for the lack of better words) of Jupyter Notebook. It is notebook, text editor and python console together with a file explorer.
Jupyter Notebook plugins may not work with JupyterLab (which is currently beta now).
If you want to use Jupyter Notebook, just use Jupyter Notebook.
If you want multiple users to juse their own Jupyter Notebooks encapsulated in the same machine, Jupyter Hub.
If you want to play with beta software, Jupyter Lab.
Is it possible to automatically refresh the notebook opened in browser, after the notebook is modified from a different tool?