Alessio Buccino
Alessio Buccino
> I am OK with the new method but I am not agree with the signature change inunit_locaization. This make non backward compatible witg previous versions. We should have discuss...
@samuelgarcia @yger - I cleaned up the docs, looking good now :) - There was a failing test due to this: the test recording probe is too short for the...
One last fix needed to use `numpy.broadcast_to` instead of `torch.broadcast_to` if `conv_engine` is `numpy`
In last commit: propagate API changes and docs
@samuelgarcia so you suggest protection at both levels? Recording + segment?
In my opinion we need at the traces level. If you have a chain, you will redo the same check many times. If it's at the recording level, it will...
Yes I would be ok checking for None and casting to int in the segment as well, but the very important one is at the `BaseRecording.get_traces`
@samuelgarcia can you fix the test? There are some `concatenated` that trigger some errors
Closing in favor of #3765