Nikita Vedeneev dd2c22f4bb bsr_dense_bmm(): enable more precise float32 support with float64 accumulators (#100882) ...

Float64 is there in Triton! This PR increases precision for float32 inputs with float64 accumulation dtype.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/100882
Approved by: https://github.com/cpuhrsch

2023-05-11 11:22:55 +00:00

..

__init__.py

Fix a bug in nesting check_sparse_tensor_invariants context managers (#95372)

2023-02-23 18:22:13 +00:00

_triton_ops.py

bsr_dense_bmm(): enable more precise float32 support with float64 accumulators (#100882)

2023-05-11 11:22:55 +00:00