chenht2022 c1cc7d2cd2 1) Linear and MLP operators support qlen>1; 2) All operators now share a single memory buffer; 3) Refactor CPUInfer submit/sync logic.

2024-08-08 09:04:36 +00:00

..

bench_linear_torch.py

1) Linear and MLP operators support qlen>1; 2) All operators now share a single memory buffer; 3) Refactor CPUInfer submit/sync logic.

2024-08-08 09:04:36 +00:00

bench_linear.py

1) Linear and MLP operators support qlen>1; 2) All operators now share a single memory buffer; 3) Refactor CPUInfer submit/sync logic.

2024-08-08 09:04:36 +00:00

bench_mlp_torch.py

1) Linear and MLP operators support qlen>1; 2) All operators now share a single memory buffer; 3) Refactor CPUInfer submit/sync logic.

2024-08-08 09:04:36 +00:00

bench_mlp.py

1) Linear and MLP operators support qlen>1; 2) All operators now share a single memory buffer; 3) Refactor CPUInfer submit/sync logic.

2024-08-08 09:04:36 +00:00

bench_moe_torch.py

1) Linear and MLP operators support qlen>1; 2) All operators now share a single memory buffer; 3) Refactor CPUInfer submit/sync logic.

2024-08-08 09:04:36 +00:00

bench_moe.py

1) Linear and MLP operators support qlen>1; 2) All operators now share a single memory buffer; 3) Refactor CPUInfer submit/sync logic.

2024-08-08 09:04:36 +00:00