OOM eventually when using create_graph=True with BatchL2Grad about backpack HOT 2 OPEN

Apparently if I tell ESGD-M to do a Hessian-vector product every step instead of every ten for compute efficiency, I don't OOM anymore. Normally the graphs made with create_graph=True are freed on their own if ESGD-M doesn't do an HVP that step but Backpack is hanging onto them somewhere?

from backpack.

Hi,

thanks for your report. From your description I think BackPACK's memory cleanup should be triggered during the backward pass.

Maybe you can try to explicitly disable BackPACK's hooks during the HVP using

from backpack import disable

with disable():
   # HVP

Otherwise, it would be great to reproduce this issue in an MWE and track down the memory leak.

Best,
Felix

from backpack.