Int4WeightOnlyQATQuantizer

class torchao.quantization.qat.Int4WeightOnlyQATQuantizer(groupsize: int = 256, inner_k_tiles: Optional[int] = 8, precision: dtype = torch.bfloat16, scales_precision: dtype = torch.bfloat16)

用于对模型执行 QAT 的量化器，其中线性层具有按通道分组的 int4 伪量化权重。