pub fn simd_quantized_matvec(
qmatrix: &QuantizedMatrix,
qparams: &QuantizationParams,
vector: &ArrayView1<'_, f32>,
) -> LinalgResult<Array1<f32>>
Expand description
SIMD-accelerated quantized matrix-vector multiplication
Performs matrix-vector multiplication where the matrix is in quantized form and the vector is in f32 format. The result is returned as f32.
§Arguments
qmatrix
- Quantized matrixqparams
- Quantization parameters for the matrixvector
- Vector to multiply with
§Returns
- Result vector of the multiplication