Horizon Robotics Official Website
Horizon Robotics Developer Community

English

中文
English

3.8.1

3.9.0
3.8.1
3.7.1
3.7.0
3.5.0
3.2.0
3.0.31

Release Notes

OE Document Introduction

OpenExplorer Overview

Product Introduction

OpenExplorer Introduction

Toolchain Overview

Key Concepts

Environment Deployment

Pre-installation Preparation

Software Installation

Quick Start

Model Quantization and Compilation

Post-training Quantization (PTQ)

PTQ Conversion Principle and Process

Basic Process

Prepare Floating-point Model

Model Checking

Data Preparation

Model Quantization and Compilation

Model Performance Evaluation

Model Accuracy Evaluation

PTQ Deployment Consistency Analysis

PTQ Conversion Samples

Release Introduction

Common Algorithm Model Samples

FAQ and Common Failure Resolutions

Post-training Quantization (PTQ) FAQ

Common Failure Resolutions

Appendix

Data Normalization Parameter Configuration

Image Processing Transformer

Quantized Awareness Training (QAT)

QAT Conversion Introduction

Terminology

Basic Process

Build Floating-point Model

Build QConfig

Prepare Description

Model Calibration

Quantized Awareness Training

Convert to Fixed-Point Model

Model Compilation and Performance Evaluation

QAT Deployment Consistency

Operator Fusion

Eager Mode

FAQ and Common Failure Resolutions

Model Deployment

Board Resources Evaluation

Board Deployment

Model Inference Application Development Guide

Alignment Rule

Dynamic Input Introduction

Basic Sample User Guide

AI Benchmark User Guide

Model Tuning Guide

Model Performance Tuning Guide

Model Performance Optimization

Efficient Model Design Advice

Model Accuracy Tuning Guide

PTQ

Model Accuracy Tuning

Accuracy Tuning Practice

QAT

Quantization Accuracy Tuning Guide

Quantization Accuracy Tuning Practice

Reference Algorithm

Overview

Framework

Module Framework

Execution Engine

Model List and Performance Benchmarks

Tools Guide

Tools Instruction

Model Compilation and Performance Analysis Tool

Picture Inference Tool

Dataset Processing Tool

Model Export Tool

Dataset Assistance Tool

Quantization Analysis Tool

Config File Introduction

Keywords Introduction

Overriding Config Parameters Using Commands

Registration Mechanism

Config Construction Sample

Turn on AMP

Quantized Training

Quantization Training Process

Data Calibration

QAT Mode

Qconfig Configuration

Examples

Docker Image

Multi-machine Instructions

Resnet50 Classification Model Training

CenterPoint Detection Model Training

Deformable DETR Detection Model Training

FCOS Detection Model Training

FCOS3D Detection Model Training

GaNet Lane Line Detection Model Training

PointPillars Detection Model Training

UNet Segmentation Model Training

QCNet Trajectory Prediction Model Training

Occupancy Prediction Model

MapTROE Training

Motr Multiple Object Track Model Training

Bev Multi-task Model Training

Bev Sparse Multitask Model Training

Lidar Fusion MultiTask Model Training

Lidar Fusion Sparse-bev Perception Model

FAQ

Unify Compute Platform (UCP)

Overview

Function Support Comparison List

Vision Process

VP Overview

Quickstart

Module Architecture

Performance Indicator

Vision Algorithm

Sample

Basic Image Processing Sample

Image Transformation Sample

Feature Extraction Sample

Optical Flow Sample

Surround View Image Stitching Sample

Video Encoding and Decoding Sample

High Performance Library

HPL Introduction

Quickstart

Module Architecture

Performance Indicator

High Performance Algorithm

Overview

Fast Fourier Transform

Inverse Fast Fourier Transform

Fast Fourier Transform 2D

Inverse Fast Fourier Transform 2D

Sample

UCP Custom Operator

Introduction

DSP Development

DSP Development Documentation

DSP Development Process

DSP Sample

Instruction

Point Cloud Pre-processing Introduction

Quantization and Dequantization Introduction

SLAM Sample Introduction

Demosaicing Sample Introduction

GPU Development Documentation

Object Detection Full Process Sample

FAQ and Error Code

FAQ

Error Code

API

HMCT API Reference

HBDK Tool API Reference

Horizon Plugin Pytorch API Reference

March

qconfig

Fake Quantize

observers

quantization.observer_v2.MinMaxObserver

quantization.observer_v2.HistogramObserver

QAT

quantization.fuse_modules

quantization.fuse_known_modules

fx.fx_helper.wrap

quantization.prepare

qconfig_setter

quantization.qconfig_setter

quantization.qconfig_setter.templates

Model Export

quantization.hbdk4.export

quantization.hbdk4.get_export_hbir_plugin_version

ONNX

Horizon Operator

nn.functional.filter

nn.functional.point_pillars_preprocess

nn.BgrToYuv444

nn.Correlation

nn.PointPillarsScatter

nn.MultiScaleDeformableAttention

UCP API Reference

UCP Common API Introduction

Overview

Data Structure

Version Information Type

HB_UCP_INITIALIZE_SCHED_PARAM

Functional Interface

VP API Reference

Data Structure

hbVPInterpolationType

Function Interface

Model Inference API Reference

Model Inference API Overview

Data Type and Data Structure

HB_DNN_TENSOR_MAX_DIMENSIONS

hbDNNTensorProperties

hbDNNTensor

Function Interface

Version Info

Model Loading & Releasing

HPL API Reference

Data Structure

Functional Interface

DSP API Reference

Data Structure

Functional Interface

hb_dsp_init_global_tm

Tools Guide

PTQ Conversion Tools

PTQ Conversion Tool Overview

The hb_compile Tool

Model Checking

Model Quantized Compilation

Model Modification and HBIR Model Compilation

The quant_config Introduction

The hb_model_info Tool

The hb_verifier Tool

The hb_eval_preprocess Tool

The hb_config_generator Tool

The HBRuntime Inference Library

Accuracy Debug Tool

QAT Accuracy Tuning Tool

Performance Analysis Tool

Model Inference Tools

Common Operations

The hrt_model_exec Tool

View the Model Information

Model Inference

Model Performance Evaluation

Transmission Optimization

Common issues

UCP Performance Analysis Tools

UCP Trace Instructions

The hrt_ucp_monitor Tool Introduction

Samples

Benchmark of Model Performance

J6P/H Benchmark of Model Performance

J6E/M Benchmark of Model Performance

J6B Benchmark of Model Performance

Model Deployment Practice Guidance Examples

The Practical Guide to Deploying the ResNet18 Model with RGB Input

The Practical Guide to Deploying the ResNet18 Model with Pyramid Input

The Practical Guide to Deploying Multi-Batch ResNet18 Model with Pyramid Input

The Practical Guide to Deploying the ResNet18 Model with Resizer Input

The Practical Guide to Deploying ResNet18 Model with Dequantize Nodes Removed

Appendix

Toolchain Operator Support Constraint List

Operator Support List

J6P/H ONNX Operator Support List

J6P/H Torch Operator Support List

J6E/M ONNX Operator Support List

J6E/M Torch Operator Support List

J6B ONNX Operator Support List

J6B Torch Operator Support List

Operator BPU Constraint List

J6P/H ONNX Operator BPU Constraint List

J6P/H Torch Operator BPU Constraint List

J6E/M ONNX Operator BPU Constraint List

J6E/M Torch Operator BPU Constraint List

J6B ONNX Operator BPU Constraint List

J6B Torch Operator BPU Constraint List

Dataset Download

License Agreement and Third-party Software Vulnerability Description

J6 Algorithm Toolchain License Agreement

Third-party Software and License Statement

Third-party Software Vulnerability Description

>Model Quantization and Compilation>Quantized Awareness Training (QAT)>Basic Process

Basic Process

The basic workflow of the quantized-aware training is as follows:

Previous pageTerminology Next pageBuild Floating-point Model

#Basic Process

Basic Process