Reload semantics #38

Activity

1. Carreau commented on Feb 28, 2018

   

   Carreau

   on Feb 28, 2018

   Member

   More actions

   Thanks for opening the issue and the nice comments.
   We don't have a lot of bandwidth to work on this project so it has not moved forward a lot.

   One of the limitation here is the one of Python's import, and I am not familiar enough with the internal to know whether what you ask is possible.

   In JupyterLab you should be able to attach the same kernel to both notebook, which might help mitigate the issue.
2. michael-pont commented on Jan 10, 2019

   

   michael-pont

   on Jan 10, 2019

   More actions

   I got the same error when trying to change and save the file where my helper functions were defined within Jupyter Notebooks.

   Rerunning the below line in a jupyter cell would not reload the newly saved functions.

   import ipynb.fs.defs.Helper_Functions as helper
   

   @mikepurvis solution worked for me. I ran:

   import importlib
   importlib.reload(helper)
   
3. inspirepassion commented on Nov 5, 2019

   

   inspirepassion

   on Nov 5, 2019

   More actions

   eload(helper

   /opt/anaconda3/envs/ml/lib/python3.7/importlib/init.py in reload(module)
   138 """
   139 if not module or not isinstance(module, types.ModuleType):
   --> 140 raise TypeError("reload() argument must be a module")
   141 try:
   142 name = module.spec.name

   TypeError: reload() argument must be a module

   I got this error when I try to reload one function under my helper_fun.ipynb file when I ran

   from ipynb.fs.defs.helper_fun import add_num
   from importlib import reload
   reload(add_num)
   
4. tonyfast

   added a commit that references this issue on Jan 2, 2020

   Add backports for capturing output and a parameterization function. (i…

   ...

   2d1d14e