Parcollet Titouan

I've seen a few changes to RelPos and InputNorm going through I believe, could this be related?

Mystery / 20, I'll investigate once done with my current duties at SAIC.

Alright, I think that there are two problems here. 1. Building a wrapper that will naively replace Linear layers with LoRA / Houlsby and others is simple, I did it,...

@pplantinga I do not like the idea of adding accelerate and torch-metrics as dependencies. I do not trust the stability of both dependencies and I hate depending on other things...

Hi, the reason for this error is that you are not using the correct inference class to use speechbrain/emotion-recognition-wav2vec2-IEMOCAP. This model, as you can see from [the code snippet here](https://huggingface.co/speechbrain/emotion-recognition-wav2vec2-IEMOCAP)...

@Adel-Moumen this looks like an ontology that we do not want to apply only on this class/function but rather across the interface. We should keep it on a "todo" list...

Hi the traceback of the error is not full, could you please post the full error?

Hi @avishaiElmakies, this way of adding so many recoverables and especially in that spot of the code (the init opt function) is a bit unconventional. It's better to use the...

SIde note while waiting for the .py: the new version of SpeechBrain (v1) on the develop Branch can handle any model originating from HF nicely from the lobes.

Alright, I won't be able to have a detailed review on that code right now because it is highly customised. I will however give an advice - there is no...