jupyter notebook and python code development workflow

[lin d]

Hi,

I have been using jupyter notebook for interactive and exploratory data analysis for a while, and I would like to consolidate the code into more structured modules as .py files. Although I can use %run -d  as debug mode in jupyter notebook and use pdb to continue the interactive work. However, the functionality is much more limited in comparison to the full jupyter notebook, such as multi-cell environment, inline plot of matplotlib, ipython magic functions, etc. 

I wonder if anyone has experience of turning the jupyer notebook into a debugging and python development workflow. If so, please do share your wisdom and give some pointers, thanks!

Frank