[MoE] Move GPT OSS Triton kernel experts into fused_moe/experts/ (#39007)
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com> Signed-off-by: Jackmin801 <ongjackm@gmail.com> Co-authored-by: Robert Shaw <robertgshaw2@gmail.com>
This commit is contained in:
co-authored by
Robert Shaw
parent
ecf5ff7ce3
commit
1a9353bb02
@@ -29,18 +29,22 @@ class TestTritonMoeForwardExpertMap:
|
||||
torch.tensor([0, -1, 1, -1], device=device) if expert_map_present else None
|
||||
)
|
||||
|
||||
from vllm.utils.import_utils import import_triton_kernels
|
||||
|
||||
import_triton_kernels()
|
||||
|
||||
with (
|
||||
patch("triton_kernels.topk.topk") as mock_topk,
|
||||
patch(
|
||||
"vllm.model_executor.layers.fused_moe."
|
||||
"vllm.model_executor.layers.fused_moe.experts."
|
||||
"gpt_oss_triton_kernels_moe.make_routing_data"
|
||||
) as mock_make_routing,
|
||||
patch(
|
||||
"vllm.model_executor.layers.fused_moe."
|
||||
"vllm.model_executor.layers.fused_moe.experts."
|
||||
"gpt_oss_triton_kernels_moe.triton_kernel_fused_experts"
|
||||
) as mock_fused_experts,
|
||||
):
|
||||
from vllm.model_executor.layers.fused_moe.gpt_oss_triton_kernels_moe import ( # noqa: E501
|
||||
from vllm.model_executor.layers.fused_moe.experts.gpt_oss_triton_kernels_moe import ( # noqa: E501
|
||||
triton_kernel_moe_forward,
|
||||
)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user