int_scaled_matmul¶
- torchao.quantization.int_scaled_matmul(a: Tensor, b: Tensor, scales1: Tensor) Tensor[source]¶
Performs scaled integer matrix multiplication.
- Parameters:
a (torch.Tensor) – The first matrix to multiply.
b (torch.Tensor) – The second matrix to multiply.
scales1 (torch.Tensor) – The scaling factors for the rows of the result.
- Returns:
The result of the scaled matrix multiplication.
- Return type:
- Raises:
AssertionError – If the dimensions of the input tensors do not match the expected shapes.
