Expand description
Quantization Types
Defines quantization formats and data structures.
@version 0.1.0 @author AutomataNexus Development Team
Structs§
- Q4Block
- A block of Q4_0 quantized data.
- Q4_
1Block - A block of Q4_1 quantized data (with min value).
- Q8Block
- A block of Q8_0 quantized data.
- Quantized
Tensor - A quantized tensor containing compressed weight data.
Enums§
- Quant
Type - Quantization type enumeration.
- Quantized
Block - Generic quantized block enum.