X-UMX: lr_scheduler_step() missing 1 required positional argument: 'metric' about asteroid HOT 4 CLOSED

Hi again,

The problem disappeared after downgrading Lightning to 1.9.4. I guess there are some changes in the 2.0 version that breaks the X-UMX training routine of Asteroid.

Best,
David

from asteroid.

I'm not sure this code has an easy backward compatible fix. So we should

• First change requirements to limit lightning to 1.x versions, and release asteroid 0.6.1,
• Then fix the issue and upgrade to pytorch-lightning>=2.0.0 with asteroid 0.7.0.

from asteroid.

Hello 👋

I had the same problem, it seems that the function definition is def lr_scheduler_step(self, scheduler, metric) in version 2.0 instead of def lr_scheduler_step(self, scheduler, optimizer_idx, metric) in previous versions (see the custom learning rate schedulers documentation here).

@mpariente Should we change this and update the requirements to pytorch-lightning>=2.0.0?

from asteroid.

I'm closing this issue since it has been fixed with #682 and the release 0.7.0.

from asteroid.