qbcompiler.configs.quantization_config.CalibrationConfig Class Reference

qbcompiler.configs.quantization_config.CalibrationConfig Class Reference#

Mobilint SDK qb Compiler: qbcompiler.configs.quantization_config.CalibrationConfig Class Reference

Mobilint SDK qb Compiler v1.0

MCS002-KR

Configuration for calibration during quantization. More...

Inheritance diagram for qbcompiler.configs.quantization_config.CalibrationConfig:

Public Member Functions
	__init__ (self, int quantization_method, int quantization_output, int quantization_mode, numbers.Number percentile, numbers.Number topk_ratio, int max_each, int max_total, numbers.Number hist_percentile, bool hist_use_gpu, int hist_num_bins, int hist_num_samples, int hist_buffer_size, numbers.Number hist_min_bin_width, numbers.Number hist_search_percentile_min, numbers.Number hist_search_percentile_max, int hist_num_search, int hist_search_type, numbers.Number act_scale_min, numbers.Number weight_scale_min, numbers.Number min_clip_ratio, Dict layer_overrides, bool optional=optional)
	Initialize the CalibrationConfig.
	default_config (cls)
	from_kwargs (cls, **kwargs)
	from_dict (cls, dict config_dict)
	Create CalibrationConfig from dictionary (JSON structure)
	to_dict (self)

Public Attributes
	quantization_method = quantization_method
	quantization_output = quantization_output
	quantization_mode = quantization_mode
	percentile = percentile
	topk_ratio = topk_ratio
	max_each = max_each
	max_total = max_total
	hist_percentile = hist_percentile
	hist_use_gpu = hist_use_gpu
	hist_num_bins = hist_num_bins
	hist_num_samples = hist_num_samples
	hist_buffer_size = hist_buffer_size
	hist_min_bin_width = hist_min_bin_width
	hist_search_percentile_min = hist_search_percentile_min
	hist_search_percentile_max = hist_search_percentile_max
	hist_num_search = hist_num_search
	hist_search_type = hist_search_type
	act_scale_min = act_scale_min
	weight_scale_min = weight_scale_min
	min_clip_ratio = min_clip_ratio
	layer_overrides = layer_overrides

Static Public Attributes
bool	optional = False
list	method_list = ["WChALayer", "WChAMulti", "WChALayerZeropoint", "WChAMultiZeropoint"]
list	output_list = ["Layer", "Ch", "Sigmoid"]
list	mode_list = ["max", "maxPercentile", "histogram"]
list	search_type_list = ["percentile", "mse", "kl"]
dict	DEFAULTS
dict	py_to_cpp_name
dict	min_max_check

Configuration for calibration during quantization.

Defines calibration and quantization parameterization used to derive activation/weight scales and related statistics during quantized compilation.

Definition at line 14 of file quantization_config.py.

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