biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Fix per file ruff ignores related to line length (#26262)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-10-06 06:12:40 +01:00
committed by GitHub
parent 91ac7f764d
commit 6c04638214
65 changed files with 301 additions and 291 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
tests/kernels/moe/modular_kernel_tools/common.py
View File

@@ -331,7 +331,8 @@ class WeightTensors:
in_dtype=config.dtype,
quant_dtype=config.quant_dtype,
block_shape=config.quant_block_shape,
per_out_ch_quant=config.is_per_act_token_quant, # or config.is_per_out_ch_quant
# or config.is_per_out_ch_quant
per_out_ch_quant=config.is_per_act_token_quant,
)
return WeightTensors(
w1=w1, w2=w2, w1_scale=w1_scale, w2_scale=w2_scale, w1_gs=w1_gs, w2_gs=w2_gs

2
tests/kernels/moe/modular_kernel_tools/make_feature_matrix.py
View File

@@ -124,7 +124,7 @@ def make_feature_matrix(csv_file_path: str):
results_df: Optional[pd.DataFrame] = None
for m, k, n, e, topks, dtype, pf_type, experts_type, quant_config in tqdm(
combinations
): # noqa: E501
):
config = Config(
Ms=[m],
K=k,

8
tests/kernels/moe/modular_kernel_tools/mk_objects.py
View File

@@ -10,7 +10,7 @@ import vllm.model_executor.layers.fused_moe.modular_kernel as mk
from vllm.model_executor.layers.fused_moe.batched_deep_gemm_moe import (
BatchedDeepGemmExperts,
)
from vllm.model_executor.layers.fused_moe.batched_triton_or_deep_gemm_moe import ( # noqa: E501
from vllm.model_executor.layers.fused_moe.batched_triton_or_deep_gemm_moe import (
BatchedTritonOrDeepGemmExperts,
)
from vllm.model_executor.layers.fused_moe.config import (
@@ -196,10 +196,10 @@ register_experts(
# Disable on blackwell for now
if has_deep_ep() and not current_platform.has_device_capability(100):
from vllm.model_executor.layers.fused_moe.deepep_ht_prepare_finalize import ( # noqa: E501
from vllm.model_executor.layers.fused_moe.deepep_ht_prepare_finalize import (
DeepEPHTPrepareAndFinalize,
)
from vllm.model_executor.layers.fused_moe.deepep_ll_prepare_finalize import ( # noqa: E501
from vllm.model_executor.layers.fused_moe.deepep_ll_prepare_finalize import (
DeepEPLLPrepareAndFinalize,
)
@@ -233,7 +233,7 @@ if has_pplx():
)
if has_flashinfer_cutlass_fused_moe() and current_platform.has_device_capability(100):
from vllm.model_executor.layers.fused_moe.flashinfer_cutlass_moe import ( # noqa: E501
from vllm.model_executor.layers.fused_moe.flashinfer_cutlass_moe import (
FlashInferExperts,
)
from vllm.model_executor.layers.fused_moe.flashinfer_cutlass_prepare_finalize import ( # noqa: E501