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

Convert formatting to use ruff instead of yapf + isort (#26247)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-10-05 15:06:22 +01:00
committed by GitHub
parent 17edd8a807
commit d6953beb91
1508 changed files with 115244 additions and 94146 deletions
Download Patch File Download Diff File Expand all files Collapse all files

50
tests/kernels/quantization/test_awq.py
View File

@@ -8,40 +8,42 @@ from tests.kernels.utils import opcheck
from vllm import _custom_ops as ops # noqa: F401
@pytest.mark.skipif(not hasattr(torch.ops._C, "awq_dequantize"),
reason="AWQ is not supported on this GPU type.")
@pytest.mark.skipif(
not hasattr(torch.ops._C, "awq_dequantize"),
reason="AWQ is not supported on this GPU type.",
)
def test_awq_dequantize_opcheck(monkeypatch: pytest.MonkeyPatch):
with monkeypatch.context() as m:
m.setenv("VLLM_USE_TRITON_AWQ", "0")
qweight = torch.randint(-2000000000,
2000000000, (8192, 256),
device='cuda',
dtype=torch.int32)
scales = torch.rand((64, 2048), device='cuda', dtype=torch.float16)
zeros = torch.empty((64, 256), device='cuda', dtype=torch.int32)
qweight = torch.randint(
-2000000000, 2000000000, (8192, 256), device="cuda", dtype=torch.int32
)
scales = torch.rand((64, 2048), device="cuda", dtype=torch.float16)
zeros = torch.empty((64, 256), device="cuda", dtype=torch.int32)
split_k_iters = 0
thx = 0
thy = 0
opcheck(torch.ops._C.awq_dequantize,
(qweight, scales, zeros, split_k_iters, thx, thy))
opcheck(
torch.ops._C.awq_dequantize,
(qweight, scales, zeros, split_k_iters, thx, thy),
)
@pytest.mark.skip(reason="Not working; needs investigation.")
@pytest.mark.skipif(not hasattr(torch.ops._C, "awq_gemm"),
reason="AWQ is not supported on this GPU type.")
@pytest.mark.skipif(
not hasattr(torch.ops._C, "awq_gemm"),
reason="AWQ is not supported on this GPU type.",
)
def test_awq_gemm_opcheck(monkeypatch: pytest.MonkeyPatch):
with monkeypatch.context() as m:
m.setenv("VLLM_USE_TRITON_AWQ", "0")
input = torch.rand((2, 8192), device='cuda', dtype=torch.float16)
qweight = torch.randint(-2000000000,
2000000000, (8192, 256),
device='cuda',
dtype=torch.int32)
scales = torch.randint(-2000000000,
2000000000, (64, 256),
device='cuda',
dtype=torch.int32)
qzeros = torch.empty((64, 2048), device='cuda', dtype=torch.float16)
input = torch.rand((2, 8192), device="cuda", dtype=torch.float16)
qweight = torch.randint(
-2000000000, 2000000000, (8192, 256), device="cuda", dtype=torch.int32
)
scales = torch.randint(
-2000000000, 2000000000, (64, 256), device="cuda", dtype=torch.int32
)
qzeros = torch.empty((64, 2048), device="cuda", dtype=torch.float16)
split_k_iters = 8
opcheck(torch.ops._C.awq_gemm,
(input, qweight, qzeros, scales, split_k_iters))
opcheck(torch.ops._C.awq_gemm, (input, qweight, qzeros, scales, split_k_iters))