`torch.maximum` / `torch.clamp_min` gradients at tie point violate PyTorch's documented subgradient convention and are inconsistent with `F.relu`

May 20, 2026 · #184572
View on GitHub
Python Difficulty: Medium

Labels

module: autograd module: edge cases
Ready to work on this? Walk through the full fork-to-PR workflow so your first contribution goes smoothly.

Sign in required

Authenticate to use favourites & bookmarks

5