Dear NanoHub, 

I am reaching out to kindly request your help in uploading and deploying Jupyter Notebook Tools. My tools are based on some Python modules or versions of modules which don't install globally like mp-api, torch, dash. However, I have found that there are two different sets of instructions which are causing confusion.

One set of instructions can be found at https://nanohub.org/tools/jupyterexamples -> manual -> Publishing Notebooks and Installing Local Python Modules. It advises me to install packages in the lib folder, add paths in invoke, insert paths inside notebooks, and write a Makefile to build binaries and executables in "bin".

The other set of instructions is available at https://nanohub.org/whypublish/whypublishdeployworkflow, which simply states that I just need to "Get started working on your Jupyter notebook at https://nanohub.org/tools/jupyter" and then go to the tool page and click “My code is committed, working, and ready to be installed”.

I have already set up the required Python environment and packages in my storage. I am wondering how I can upload and deploy my tool correctly. Should I follow the instructions to use Git/SVN to upload my code and set up libraries myself? Or should I just set up the environment in my storage?

I also have another question regarding using Dash (https://dash.plotly.com/workspaces/using-dash-in-jupyter-and-workspaces?de-version=5.2). NanoHub is a hosted Jupyter notebook server, so how can I use Dash here? I found a tool that uses Dash on NanoHub at https://nanohub.org/resources/snap. Should I follow this example?

I would greatly appreciate it if you could provide me with the answers and solutions to my problems.

Thank you all for considering my inquiries. 

Best regards, pigz2538