Add GPTQ support (#916)

This commit is contained in:
CHU Tianxiang
2023-12-15 03:04:22 -08:00
committed by GitHub
parent c06170cc8e
commit 0fbfc4b81b
35 changed files with 1782 additions and 82 deletions
@@ -67,17 +67,19 @@ class SqueezeLLMLinearMethod(LinearMethodBase):
def __init__(self, quant_config: SqueezeLLMConfig):
self.quant_config = quant_config
def create_weights(self, input_size: int, output_size: int,
params_dtype: torch.dtype) -> Dict[str, torch.Tensor]:
if input_size % self.quant_config.pack_factor != 0:
def create_weights(self, input_size_per_partition: int,
output_size_per_partition: int, input_size: int,
output_size: int,
params_dtype: torch.dtype) -> Dict[str, Any]:
if input_size_per_partition % self.quant_config.pack_factor != 0:
raise ValueError(
"The input size is not aligned with the quantized "
"weight shape. This can be caused by too large "
"tensor parallel size.")
qweight = Parameter(
torch.empty(
input_size // self.quant_config.pack_factor,
output_size,
input_size_per_partition // self.quant_config.pack_factor,
output_size_per_partition,
device="cuda",
dtype=torch.int32,
),
@@ -108,7 +110,7 @@ class SqueezeLLMLinearMethod(LinearMethodBase):
}
def apply_weights(self,
weights: Dict[str, torch.Tensor],
weights: Dict[str, Any],
x: torch.Tensor,
bias: Optional[torch.Tensor] = None) -> torch.Tensor:
qweight = weights["qweight"]