qubee.configs.advanced_quantization_config.AdvancedQuantizationConfig Class Reference

Mobilint SDK qb Compiler v0.12.0.0 MCS002-KR

• qubee
• configs
• advanced_quantization_config
• AdvancedQuantizationConfig

Public Member Functions | Public Attributes | Static Public Attributes | List of all members

qubee.configs.advanced_quantization_config.AdvancedQuantizationConfig Class Reference

AdvancedQuantization Configuration

Configuration for advanced quantization options used during quantization. More...

Inheritance diagram for qubee.configs.advanced_quantization_config.AdvancedQuantizationConfig:

Public Member Functions
	__init__ (self, Optional[OptqConfig] optq=None, Optional[ModConfig] mod=None, Optional[EquivalentTransformationConfig] equivalent_transformation=None, Optional[SearchWeightScaleConfig] search_weight_scale=None, bool optional=optional)
	default_config (cls)
	from_dict (cls, dict config_dict)
	Create AdvancedQuantizationConfig from dictionary (JSON structure)
	to_dict (self)

Public Attributes
	optq = optq or OptqConfig.default_config()
	mod = mod or ModConfig.default_config()
tuple	equivalent_transformation
tuple	search_weight_scale

Static Public Attributes
bool	optional = False

Detailed Description

Configuration for advanced quantization options used during quantization.

Unified configuration for advanced quantization techniques, including OPTQ, MOD, Equivalent Transformation, and Weight Scale Search.

Definition at line 17 of file advanced_quantization_config.py.

---

The documentation for this class was generated from the following file:

• advanced_quantization_config.py

---

Generated by  1.14.0