impl_group() — pytorch Function Reference

Architecture documentation for the impl_group() function in generate_kernels.py from the pytorch codebase.

Function python CoreTensor Boxing calls 1 called by 1

Entity Profile

CoreTensor→ Boxing→ impl_group() — pytorch Function Reference

Dependency Diagram

graph TD
  afbd71b3_48f9_da2a_c9e9_e31dd8ce4ee4["impl_group()"]
  de6536f5_4445_ba49_9313_fdf283611fe7["impl_group()"]
  de6536f5_4445_ba49_9313_fdf283611fe7 -->|calls| afbd71b3_48f9_da2a_c9e9_e31dd8ce4ee4
  de6536f5_4445_ba49_9313_fdf283611fe7["impl_group()"]
  afbd71b3_48f9_da2a_c9e9_e31dd8ce4ee4 -->|calls| de6536f5_4445_ba49_9313_fdf283611fe7
  style afbd71b3_48f9_da2a_c9e9_e31dd8ce4ee4 fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

aten/src/ATen/native/transformers/cuda/mem_eff_attention/kernels/generate_kernels.py lines 211–214

    def impl_group(self) -> str:
        # Maps to file which will contain the implementation
        dropout_suffix = "_dropout" if self.apply_dropout else ""
        return f"{self.dtype}_{self._aligned_suffix}_k{self.max_k}{dropout_suffix}"