viable/strict/1792040002: Fix huber loss backward for mixed inputs (#192027)
- PyTorch: 2348 events in the last 90 days
- PyTorch: 2318th Release in the last 90 days
- Previous: earlier the same day · trunk/620cac79649a8e9aedb624229e1dfe54e180ced5: [MPS] Prescale Jacobi SVD by the largest entry (#200776)
What happened
fixes #76646 summary: backward of huber loss didnt support type promotion and raised an error when different dtypes were used in input Pull Request resolved: #192027 Approved by: https://github.com/soulitzer Co-authored-by: soulitzer via mergedog soulitzer@gmail.com
Summary assembled by rule from the sources below