ONE - On-device Neural Engine
Loading...
Searching...
No Matches
mpqsolver::core::Quantizer::Context Struct Reference

#include <Quantizer.h>

Data Fields

std::string output_model_dtype = "uint8"
 
std::string granularity = "channel"
 
std::string input_type = "uint8"
 
std::string output_type = "uint8"
 
bool TF_style_maxpool = false
 
bool save_min_max = false
 

Detailed Description

Definition at line 47 of file Quantizer.h.

Field Documentation

◆ granularity

std::string mpqsolver::core::Quantizer::Context::granularity = "channel"

◆ input_type

std::string mpqsolver::core::Quantizer::Context::input_type = "uint8"

Definition at line 51 of file Quantizer.h.

Referenced by entry(), and mpqsolver::core::Quantizer::quantize().

◆ output_model_dtype

std::string mpqsolver::core::Quantizer::Context::output_model_dtype = "uint8"

Definition at line 49 of file Quantizer.h.

Referenced by entry(), and mpqsolver::core::Quantizer::quantize().

◆ output_type

std::string mpqsolver::core::Quantizer::Context::output_type = "uint8"

Definition at line 52 of file Quantizer.h.

Referenced by entry(), and mpqsolver::core::Quantizer::quantize().

◆ save_min_max

bool mpqsolver::core::Quantizer::Context::save_min_max = false

Definition at line 54 of file Quantizer.h.

Referenced by entry(), and mpqsolver::core::Quantizer::quantize().

◆ TF_style_maxpool

bool mpqsolver::core::Quantizer::Context::TF_style_maxpool = false

Definition at line 53 of file Quantizer.h.

Referenced by entry(), and mpqsolver::core::Quantizer::quantize().


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