dequantize_affine_floatx

torchao.quantization.dequantize_affine_floatx(tensor: Tensor, scale: Tensor, ebits: int, mbits: int, output_dtype: dtype = torch.float32) → Tensor