English

The hb_eval_preprocess Tool

The hb_eval_preprocess tool is used for preprocessing image data on x86 system.

Preprocessing refers to certain image data operations before being fed into the model, e.g., image resizing, image cropping, image padding, etc.

The basic flow of the hb_eval_preprocess tool to preprocess an image and generate a binary file is shown below:

Support Range

The board-side model is only supported by the AIbenchmark PTQ sample model, whose model is included:

mobilenetv1, mobilenetv2, resnet50, googlenet, efficientnet_lite0, efficientnet_lite1, efficientnet_lite2, efficientnet_lite3, efficientnet_lite4, vargconvnet, efficientnasnet_m, efficientnasnet_s, resnet18, yolov2_darknet19, yolov3_darknet53, yolov5x、ssd_mobilenetv1, centernet_resnet101, yolov3_vargdarknet, deeplabv3plus_efficientnetb0, fastscnn_efficientnetb0, deeplabv3plus_efficientnetm1, deeplabv3plus_efficientnetm2.

Usage

Usage: hb_eval_preprocess [OPTIONS]

Example: hb_eval_preprocess -m mobilenetv1 -i ./files

Options:
  -h, --help                      Show this message and exit.
  --version                       Show the version and exit.
  -m, --model_name                Input model name.  [required]                                
                                  [mobilenetv1|mobilenetv2|resnet50|googlenet|efficientnet_lite0|efficientnet_lite1|
                                  efficientnet_lite2|efficientnet_lite3|efficientnet_lite4|vargconvnet|efficientnasnet_m|
                                  efficientnasnet_s|resnet18|yolov2_darknet19|yolov3_darknet53|yolov5x|ssd_mobilenetv1|
                                  centernet_resnet101|yolov3_vargdarknet|deeplabv3plus_efficientnetb0|fastscnn_efficientnetb0|
                                  deeplabv3plus_efficientnetm1|deeplabv3plus_efficientnetm2]                                
  -i, --image_dir TEXT            Input image dir.  [required]
  -o, --output_dir TEXT           Output dir.
  -v, --val_txt TEXT

Parameters Introduction

Parameter Name	Parameter Description
`-h, --help`	Display help information and exit.
`--version`	Display version information and exit.
`-m, --model_name`	Specify the model name. You can use the `hb_eval_preprocess --help` command to see supported models.
`-i, --image_dir`	Specify the path of input images.
`-o, --output_dir`	Specify the output path.
`-v, --val_txt`	Specify the name of the image file required by the evaluation. The preprocessing will use these names to generate images.

Usage Example

Taking the resnet50 model as an example, you can use the following command to preprocess the image:

hb_eval_preprocess -m resnet50 -i ./files -o ./output

Output Contents

The hb_eval_preprocess tool will generate binary image files in the directory specified by the --output_dir parameter.

Hint

More application samples of the hb_eval_preprocess tool in the accuracy evaluations of on-board models can be found in Data Pre-process.

On This Page

PTQ Conversion Tools

The hb_compile Tool

PTQ Conversion Steps

PTQ Conversion Samples

FAQ and Common Failure Resolutions

Appendix

Tutorial

Advanced Tutorial

API Reference

QAT

qconfig_setter

Model Export

Horizon Operator

FAQ and Common Failure Resolutions

Model Inference Development

Model Inference API Introduction

Data Type and Data Structure

Function Interface

Model Inference Tool Introduction

The hrt_model_exec Tool Introduction

The hbm_infer Tool Introduction

UCP Common API Introduction

Data Structure

Functional Interface

UCP Performance Analysis Tools

FAQ and Error code

Model Deployment Principle and Process

Model Deployment Practice Guidance Examples

HMCT API Reference

Toolchain Operator Support Constraint List

Operator Support List

Operator BPU Constraint List

Community Quality Articles

The hb_eval_preprocess Tool

Support Range

Usage

Parameters Introduction

Usage Example

Output Contents

The hb_compile Tool

QAT

qconfig_setter

Model Export

Horizon Operator

Model Inference API Introduction

Data Type and Data Structure

Function Interface

Model Inference Tool Introduction

The hrt_model_exec Tool Introduction

The hbm_infer Tool Introduction

Data Structure

Functional Interface

Operator Support List

Operator BPU Constraint List

#The hb_eval_preprocess Tool

#Support Range

#Usage

#Parameters Introduction

#Usage Example

#Output Contents

The hb_eval_preprocess Tool

Support Range

Usage

Parameters Introduction

Usage Example

Output Contents