Expand description
Matrix quantization types and implementations
This module contains the QuantizedMatrix struct and QuantizedData2D enum along with their implementations for handling quantized matrix data.
Structs§
- Quantized
Matrix - A matrix with quantized values
Enums§
- Quantized
Data2D - Storage for quantized 2D data (matrices) in different formats
Functions§
- get_
quantizedmatrix_ 2d_ i8 - Helper function to get the i8 data from a QuantizedMatrix if available