Skip to content

Mixed-precision training workflow for DrivAerML? #6

Description

@vpuri3

The public training scripts (train_surface.py / main_drivaerml_surface.py) appear to run in FP32 (model.float(), no autocast / GradScaler). The paper mentions training in fp16 or bf16.

I tried running Transolver-3 on DrivAerML in fp16 and training blows up (NaNs). Could you share the mixed-precision recipe you used (fp16 vs bf16, AMP settings, any extra stability tricks)?

Activity

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

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions