I think it's better merged to data_clamp for now.

Thanks @huangziwei! Unfortunately the latter 2 bullet points in #375 are still triggering the performance warning for me though
Update: just kidding, performance warning gone :)

The performance warning for adding channels is fixed though!

I guess it's device dependent (My laptop has 16Gb RAM), and there's no warning when I ran your retina notebook. It's also why I said the fix likely just delays the warning...

can you try replacing

            view["global_comp_index"] = view["comp_index"]
            view["global_branch_index"] = view["branch_index"]
            view["global_cell_index"] = view["cell_index"]

in network.py with

            global_indices = pd.DataFrame(
                {
                    "global_comp_index": view["comp_index"],
                    "global_branch_index": view["branch_index"],
                    "global_cell_index": view["cell_index"],
                }
            )
            view = pd.concat([view, global_indices], axis=1)

?

Just do this at the beginning of your script to suppress the warning:

import pandas as pd
from warnings import simplefilter
simplefilter(action="ignore", category=pd.errors.PerformanceWarning)

I think the concern is some pandas operations used up too much RAM, so we should handle them in a more efficient way.

are we really having problems with this or is it just the warning? I don't think that pandas is a limiting factor to runtime performance of Jaxley.

It is not a problem, just a low-priority thing to improve. Anyone who uses groups in Jaxley would probably get severely annoyed by the warning and have to silence it though. The warning is actually gone now for me though thanks to the changes!

I am not sure it has to do with groups, just with lots of mechanisms

Yes I think you're right

I am closing this as the major patches has been merged into the data_clamp branch and the latest commit has been addressed by another PR #384