bilibiliGO

How ot imporve throughput on tritonserver

3

Description: This is the result of perf_analyzer when I start one tritonserver：  After that, I turned on the SO_REUSEPORT switch of grpc and started 8 tritonservers. This is the...

[Tosa backend] Support MatmulStaticBroadcast_basic for tosa

4

This PR is to solve this issue: #2581 # Test case ``` class Matmul(nn.Module): def __init__(self): super(Matmul, self).__init__() def forward(self, x, weight): r = torch.matmul(x, weight) return r def gen_mt_mlir():...

[TORCH-LINALG] Add recomposition for select+copy_

2

I am relatively new to the world of developing open-source projects and currently working on Torch-MLIR. It's possible that I might have overlooked certain processes or encountered errors in my...