New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

`BatchNorm` is not twice-differentiable #2154

Open

ToucheSir opened this issue Jan 6, 2023 · 0 comments

Labels

Member

ToucheSir commented Jan 6, 2023

Originally reported at https://discourse.julialang.org/t/compilation-error-in-zygote-flux-and-cuda-interaction/92571. The culprit is https://github.com/FluxML/Flux.jl/blob/master/src/cuda/cudnn.jl#L14-L21 not being differentiable.

ToucheSir added gradients cuda labels

alexrosen45 added a commit to alexrosen45/Flux.jl that referenced this issue


          Make BatchNorm twice-differentiable

dd4757b

Solves the BatchNorm not being twice-differentiable problem first reported at https://discourse.julialang.org/t/compilation-error-in-zygote-flux-and-cuda-interaction/92571 and made into issue FluxML#2154.

alexrosen45 mentioned this issue

Make BatchNorm twice-differentiable #2255

Open

3 tasks

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment