ipython6 and jupyter: Slow tab-completion on large data object in notebook


The problem

Using ipython6 with Jupyter notebook, you see much longer delays for tab-completion on large variables (arrays ~1GB in size in memory) which don't occur with previus version.


In the meantime, looks like there is a way one can disable jedi in ipython with Completer.use_jedi: https://ipython.readthedocs.io/en/stable/config/options/terminal.html

Specifically you can use ipython locate profile to find your current profile directory, and edit the ipython_config.py to add c.IPCompleter.use_jedi = False