D-Robotics Official Website
D-Robotics Developer Community
D-Robotics Official Website
D-Robotics Developer Community
English

AI Benchmark User Guide

Horizon's S100 AI Benchmark Sample Package contains the most frequently used performance and accuracy evaluation samples of classification, detection, segmentation, optical-flow, tracking estimation, lidar multitask, bev, depth estimation and online map construction models. The performance evaluation samples include single-frame latency evaluation and multithreading evaluation samples, which make full use of the speed of invoking the BPU for evaluation. The pre-build source code, executable programs and evaluation scripts in the AI Benchmark Sample Package allow you to experience the samples, and develop their own applications, which makes development easier.

Release Description

The AI Benchmark sample package is located in the samples/ucp_tutorial/dnn/ai_benchmark/ path of the horizon_s100_open_explorer release package and consists the following main contents:

No.NameDescription
1codeThis folder contains sample source code and compilation scripts.
2s100Dev board operating environment of the AI Benchmark Sample Package.

Sample Code Package

Note

The on-board model needs to be obtained first by executing the resolve_ai_benchmark_ptq.sh and the resolve_ai_benchmark_qat.sh in the samples/ai_toolchain/model_zoo/runtime/ai_benchmark/ directory of the OE package respectively.

Directory of the sample code package is shown as below:

├── code # sample source code │   ├── build_ptq_s100_qnx.sh │   ├── build_ptq_s100.sh │   ├── build_qat_s100_qnx.sh │   ├── build_qat_s100.sh │   ├── build.sh # common build script │   ├── CMakeLists.txt │   ├── include # source code header files │   │   ├── base │   │   ├── input │   │   ├── method │   │   ├── output │   │   ├── plugin │   │   └── utils │   ├── README.md │   ├── resolve.sh # download script for model performance evaluation dataset │   └── src # sample source code │   ├── input │   ├── method │   ├── output │   ├── plugin │   ├── simple_example.cc │   └── utils ├── s100 # Sample package runtime environment │   ├── ptq # PTQ (post-training quantization) model samples │   │   ├── data # Model performance evaluation dataset │   │   ├── mini_data # performance evaluations dataset │   │   ├── model # PTQ (post-training quantization) solution model │ │ │ ├── README.md │ │ │ └── runtime -> ../../../../../../model_zoo/runtime/ai_benchmark/ptq # Soft link to the model in the OE package, the board environment you need to specify the model path │   │   ├── README.md │   │   ├── script # execution script │ │ │ ├── aarch64 # executable files generated by the compilation and dependencies │ │ │ ├── classification # samples classification models │ │ │ ├── config # configuration files of model inference │ │ │ ├── detection # detection model samples │ │ │ ├── segmentation # segmentation model samples │ │ │ ├── env.sh # basic environment script │ │ │ └── README.md │   │   └── tools # accuracy evaluation tools │   └── qat # QAT training model samples │   ├── data # Model performance evaluation dataset │   ├── mini_data # model performance evaluations dataset │   ├── model # QAT scheme model │   │   ├── README.md │ │ └── runtime -> ../../../../../../model_zoo/runtime/ai_benchmark/qat # Soft link to the model in the OE package, the board environment you need to specify the model path │   ├── README.md │   ├── script # execution script │   │   ├── aarch64 # executable files generated by the compilation and dependencies │   │   ├── bev # bev model samples │   │   ├── classification # samples classification models │   │   ├── config # model inference profile │   │   ├── detection # detection model samples │   │   ├── disparity_pred # disparity pred model samples │   │   ├── multitask # multi task model samples │   │   ├── online_map # online map construction model samples │   │   ├── opticalflow # optical flow model samples │   │   ├── segmentation # segmentation model samples │   │   ├── tracking # tracking model samples │   │   ├── traj_pred # traj pred model samples │   │   ├── env.sh # basic environment scripts │ │ └── README.md │   └── tools # pre-processing and accuracy evaluation tools └── README.md
  • The code directory contains the source code of the evaluation program, used to evaluate model performance and accuracy.
  • The s100 directory contains various pre-compiled application programs and evaluation scripts, used to evaluate the accuracy and performance of different models in Horizon's BPU (Brain Processing Unit).
  • The build_ptq_s100.sh script is the PTQ real program one-click compilation script.
  • The build_ptq_s100_qnx.sh script is the PTQ real program one-click compilation script.
  • The build_qat_s100.sh script is the QAT real program one-click compilation script for QNX system.
  • The build_qat_s100_qnx.sh script is the QAT real program one-click compilation script for QNX system.

Sample Models

Model releases for the AI Benchmark sample package include PTQ model and QAT model releases:

  • You can get the model_zoo of the PTQ model by executing the script resolve_ai_benchmark_ptq.sh of the samples/ai_toolchain/model_zoo/runtime/ai_benchmark/ path.
  • You can get the model_zoo of the QAT model by executing the script resolve_ai_benchmark_qat.sh of the samples/ai_toolchain/model_zoo/runtime/ai_benchmark/ path.

Among them, which contain the commonly used classification, detection, segmentation and optical flow prediction models, and the naming rules of the models is {model_name}_{backbone}_{input_size}_{input_type}.

Note

Both the PTQ and QAT models in model_zoo are compiled by the original model.

For the PTQ original model details, you can refer to the section PTQ Conversion Samples Guide.

The performance data of the model in the AI Benchmark sample package can be found in the section Benchmark of Model Performance.

Public Datasets

The dataset will be used in the sample, you can download the corresponding dataset in section Dataset Download. If you have any questions during the data preparation process, please contact Horizon.

Environment Configuration

Before using the AI Benchmark sample package, you need to ensure that the development board environment and compilation environment are available:

  • Prepare the Development Board

    1. After getting the development board, upgrade the system image file to the version recommended by the sample package.

    2. Make sure the remote connection between local dev machine and the dev board.

  • Prepare the compilation environment

    Install the aarch64-none-linux-gnu-gcc and aarch64-none-linux-gnu-g++ cross-compilation tool in current environment. Then execute the build_ptq_s100.sh and build_qat_s100.sh scripts in the code directory to compile the executable program in the real machine environment with one click. If you need to compile on the QNX platform, you can run the build_ptq_s100_qnx.sh and build_qat_s100_qnx.sh scripts. The executable programs and corresponding dependencies will be copied into the aarch64 sub-folders of the s100/ptq/script and s100/qat/script folders automatically.

Note

The cross-compilation tool specified by the build.sh script is located in the /opt folder. If you want to install it into some other locations, please modify the script.

export CC=/arm-gnu-toolchain-12.2.rel1-x86_64-aarch64-none-linux-gnu/bin/aarch64-none-linux-gnu-gcc export CXX=/arm-gnu-toolchain-12.2.rel1-x86_64-aarch64-none-linux-gnu/bin/aarch64-none-linux-gnu-g++

Similarly, when compiling on the qnx platform, the location of the cross-compilation toolchain specified in the build.sh script is /opt/qnx800/host/linux/x86_64. If it is installed in another location, you can modify the value of the environment variable QNX_HOST to specify the location of the cross-compilation toolchain.

export QNX_HOST=/opt/qnx800/host/linux/x86_64

How to Use

Evaluation Scripts

Evaluation sample scripts are in the script and tools folders.

The script folder contains the scripts used for evaluating frequently used classification, detection, segmentation, optical-flow, tracking estimation, lidar multitask, bev and depth estimation models in dev board. There are three scripts under each model:

ScriptDescription
fps.shThe script implements FPS statistics (multi-threading scheduling. You can freely specify number of threads as needed) .
latency.shThe script implements statistics of single-frame latency (one thread, single-frame).
accuracy.shThe script is used for evaluating model accuracy.
script: ├── aarch64 # executable files generated by the compilation and dependencies │ ├── bin │ └── lib ├── env.sh # base config ├── config │ ├── model │ │ ├── data_name_list # image name config │ │ └── input_init # model input config │ │ └── input_order # model input order │ ├── preprocess │ │ └── centerpoint_preprocess_5dim.json # preprocess config │ └── reference_points # reference points config │ │ ├── bev_gkt_mixvargenet_multitask_nuscenes │ │ └── ... ├── detection # detection models │ ├── fcos_efficientnetb0_mscoco # there are other models in this directory, just use this model directory as a reference │ │ ├── accuracy.sh │ │ ├── fps.sh │ │ ├── latency.sh │ │ ├── workflow_accuracy.json # accuracy config file │ │ ├── workflow_fps.json # fps config file │ │ └── workflow_latency.json # latency config file │ └──... ├── ... └── README.md

The (PTQ)tools folder contains the precision calculation scripts under python_tools, which used for accuracy evaluation.

python_tools └── accuracy_tools ├── cls_imagenet_eval.py ├── det_coco_eval.py ├── det_voc_eval.py ├── metrics │ ├── cityscapes_metric.py │ ├── coco_config.py │ ├── coco_metric.py │ └── voc_metric.py └── seg_cityscapes_eval.py

The (QAT)tools folder contains pre-processing scripts and precision calculation scripts, which used for QAT model accuracy evaluation.

tools/ ├── eval_preprocess │ ├── util │ ├── ...... │ └── voc.py ├── python_tools │ └── accuracy_tools │ ├── argoverse_util │ ├── nuscenes_metric_pro │ ├── whl_package │ ├── bev_eval.py │ ├── ...... │ └── yolov3_eval.py └── README.md
Attention

Run the following commands before the evaluation and copy the ptq (or the qat) directory to the dev board.

scp -r samples/ucp_tutorial/dnn/ai_benchmark/s100/ptq root@192.168.1.1:/userdata/ptq/ scp -r samples/ucp_tutorial/dnn/ai_benchmark/s100/qat root@192.168.1.1:/userdata/qat/

Parameters of json Configuration File

This section we provide you the brief introduction of the configuration for the workflow_fps.json, workflow_latency.json and workflow_accuracy.json. The configuration can be simply divided into input_config, output_config and the workflow configuration.

Note

The configuration parameters given below are the general configuration, some sample models will have additional configuration due to the model specificity, please refer to the sample model json file for details.

input_config

Parameter NameDescriptionInvolved json Files
input_typeSpecify the input data format, support image or bin file.fps.json, latency.json, accuracy.json
heightSpecify the input data height.fps.json, latency.json, accuracy.json
widthSpecify the input data width.fps.json, latency.json, accuracy.json
data_type

Specify the input data type. Supported type can refer to hbDNNDataTypesection.
The corresponding data types are sorted from top to bottom, corresponding to the numbers 0,1,2...
For example, HB_DNN_TENSOR_TYPE_S4 corresponds to the 0,
HB_DNN_TENSOR_TYPE_U4 corresponds to the 1,
HB_DNN_TENSOR_TYPE_S8 corresponds to the 2...

fps.json, latency.json, accuracy.json
image_list_fileSpecify the path of the lst file of the preprocessing dataset.fps.json, latency.json, accuracy.json
need_pre_loadSpecify whether to read the dataset using the preload method. fps.json, latency.json, accuracy.json
limitSpecify the threshold for the difference between the amount of input data being processed and has been processed, which is used to control the processing threads for the input data.fps.json, latency.json, accuracy.json
need_loopSpecify whether to use cyclic read data for evaluation.fps.json, latency.json, accuracy.json
max_cacheWhen this parameter takes effect, the image will be pre-processed and read into memory. To ensure the stable running of your application, do not set too large a value, we recommend that you set a value of no more than 30.fps.json, latency.json, accuracy.json

output_config

Parameter NameDescriptionInvolved json Files
output_typeSpecify the output data type.fps.json, latency.json and accuracy.json
in_orderSpecify whether to output in order.fps.json, latency.json and accuracy.json
enable_view_outputSpecify whether to visualize the output.fps.json and latency.json
image_list_enableWhen visualizing, set to true to save the output as the image type.fps.json and latency.json
view_output_dirSpecify the path of the visualization result output file.fps.json and latency.json
eval_enableSpecify whether to evaluate the accuracy.accuracy.json
output_fileSpecify the model output result file.accuracy.json

Workflow Configuration

Model inference configurations:

Parameter NameDescriptionInvolved json Files
method_typeSpecify the model inference method, which here needs to be configured as InferMethod.fps.json, latency.json, accuracy.json
method_configSpecify the model inference parameters.
  • core: specify the inference core id.
  • model_file: specify the model file.
fps.json, latency.json, accuracy.json

Post-processing configurations:

Parameter NameDescriptionInvolved json Files
thread_countSpecify the post-processing thread count, in range 1-8.fps.json, latency.json and accuracy.json
method_typeSpecify the post-processing method.fps.json, latency.json and accuracy.json
method_configSpecify the post-processing parameters.fps.json, latency.json and accuracy.json

Performance Evaluation

Performance evaluation is divided into latency and fps.

How to Use Performance Evaluation Scripts

latency:

In the directory of the to-be-evaluated model, run sh latency.sh to evaluate single frame latency, as shown below:

I0419 02:35:07.041095 39124 output_plugin.cc:80] Infer latency: [avg: 13.124ms, max: 13.946ms, min: 13.048ms], Post process latency: [avg: 3.584ms, max: 3.650ms, min: 3.498ms].
Note
  • infer denotes the time consumption of model inference.
  • Post process denotes the time consumption of post-processing.

fps:

This function uses multi-threaded concurrency and is designed to allow the model to reach the ultimate performance on BPU. Due to the multi-thread concurrency and data sampling, the frame rate value will be low during the start-up phase, then the frame rate will increase and gradually stabilize, with the frame rate fluctuating within 0.5%. To test the frame rate, go to the model directory and run sh fps.sh, as shown below.

I0419 02:35:00.044417 39094 output_plugin.cc:109] Throughput: 1129.39fps # 模型帧率

About Command-line Parameters

The fps.sh script is shown as below:

#!/bin/sh source ../../env.sh export SHOW_FPS_LOG=1 export STAT_CYCLE=100 # specify environment variable, fps log printing frequency ${app} \ --config_file=workflow_fps.json \ --log_level=1

The latency.sh script is shown as below:

#!/bin/sh source ../../env.sh export SHOW_LATENCY_LOG=1 # specify environment variable, print latency level log export STAT_CYCLE=50 # specify environment variable, latency log printing frequency ${app} \ --config_file=workflow_latency.json \ --log_level=1

Result Visualization

If you want to see the effect of a single inference of the model, you can modify workflow_latency.json and re-run the latency.sh script to generate the display effect in the output_dir directory.

Attention

When the display effect is generated, the script will run slowly due to the dump effect. Only the latency.sh script dump is supported.

The Visual operation steps are as follows:

  1. Modify the latency configuration file

    "output_config": { "output_type": "image", "enable_view_output": true, # turn on visualization "view_output_dir": "./output_dir", # visualization result output path "image_list_enable": true, "in_order": false }

  2. Execute the latency.sh script

    sh latency.sh
Attention

The visualization of the bev model needs to specify the scene information and the path of the homography matrix. The homography matrix is used for the conversion of the camera perspective and the bird's-eye view. Different scenes have their own homography matrices. The visualization of the online map construction model needs to specify the perception range of local map.

The workflow_latency.json configuration file of the bev model is recommended to be modified as follows:

"output_config": { "output_type": "image", "enable_view_output": true, # turn on visualization "view_output_dir": "./output_dir", # visualization result output path "bev_ego2img_info": [ "../../config/visible/bev/scenes.json", # scene information of input file "../../config/visible/bev/boston.bin", # homography matrix of the boston scene "../../config/visible/bev/singapore.bin" # homography matrix of the singapore scene ], "image_list_enable": true, "in_order": false }

The workflow_latency.json configuration file of the online map construction model is recommended to be modified as follows:

"output_config": { "output_type": "image", "enable_view_output": true, # turn on visualization "view_output_dir": "./output_dir", # visualization result output path "bev_range": [ -30.0, -15.0, -10.0, 30.0, 15.0, 10.0 ], # perception range of local map "image_list_enable": true, "in_order": false }
Attention

QCNet integrates model visualization and accuracy evaluation.

The visualization method of qcnet is as follows:

python3 qcnet_eval.py --eval_path \ # Path to inference result eval.log file --gt_files_path \ # Mount path for dataset --msg_pack_path \ # Path for parsed eval.log saved as msgpack --vis_save_path # Output path for visualization results

The visualization results that can be achieved by different types of models are different, refer to the table below:

Model CategoryVisualization
classificationimage
detection 2dimage
detection 3dimage
segmentationimage
keypointimage
lane lineimage
optical flowimage
lidarimage
lidar multitask

image

image

bev

image

image

image

image

image

image

image

image

raj_predimage
disparity_pred

image

image

online map constructionimage
occ_predimage
Attention

If you need to visualize images other than minidata during trajectory prediction visualization, you need to configure additional road information and trajectory information files in minidata/argoverse1/visualization. You can use the densent_process.py preprocessing script to generate configuration files, and set --is-gen-visual-config parameter to true.

Model Accuracy Evaluation

Take the following 5 steps to perform the model evaluation:

  1. Data pre-process.

  2. Data mounting.

  3. The lst file generation.

  4. Model inference.

  5. Model accuracy computing.

Data Pre-processing

The following section will provide the description of the PTQ and QAT model data preprocessing methods.

PTQ Model Data Pre-processing:

To the PTQ model data pre-processing, run the hb_eval_preprocess tool in x86 to pre-process data. The so-called pre-processing refers to the special processing operations before images are fed into the model. For example: resize, crop and padding, etc. The tool is integrated into the horizon_tc_ui tool and it will be available after the tool is installed using the install script. After the raw dataset is pre-processed by the tool, the corresponding pre-processed binary file .bin file set of the model will be generated.

Tip

About the hb_eval_preprocess tool command line parameters and usage, you can type hb_eval_preprocess -h, or see the hb_eval_preprocess Tool in the PTQ tools guide.

The datasets corresponding to each model in the sample package are described in detail below, as well as the pre-processing operations for the corresponding datasets.

The datasets used for PTQ models include the following:

DatasetDescription

VOC

Dataset

For evaluation of detection models using the VOC2012 dataset.

The dataset download and reference download structure can be found in section Preparing Datasets. The sample mainly use the val.txt file under the Main folder, the source images under the JPEGImages folder and the annotations under the Annotations folder.

2. Dataset preprocessing methods example:

hb_eval_preprocess -m ssd_mobilenetv1 \ -i VOCdevkit/VOC2012/JPEGImages \ -v VOCdevkit/VOC2012/ImageSets/Main/val.txt \ -o ./pre_ssd_mobilenetv1

COCO

Dataset

For evaluation of detection models using the coco_val2017 dataset.

1. The dataset download and reference download structure can be found in section Preparing Datasets. The samples mainly use the instances_val2017.json annotation file under the annotations folder images under the images folder.

2. Dataset preprocessing methods example:

hb_eval_preprocess -m model_name \ -i coco/coco_val2017/images \ -o ./pre_model_name

ImageNet

Dataset

For evaluation of classification models using the imagenet dataset.

1. The dataset download and reference download structure can be found in section Preparing Datasets. The samples mainly use the annotation file val.txt and the source images in the val directory.

2. Dataset preprocessing methods example:

hb_eval_preprocess -m model_name \ -i imagenet/val \ -o ./pre_model_name

Cityscapes

Dataset

For evaluation of segmentation models using the cityscapes dataset.

1. The dataset download and reference download structure can be found in section Preparing Datasets. The samples mainly use the annotation files under the ./gtFine/val folder and the source images in the ./leftImg8bit/val folder.

2. Dataset preprocessing methods example:

hb_eval_preprocess -m model_name \ -i cityscapes/leftImg8bit/val \ -o ./pre_model_name

QAT Model Data Pre-processing:

The QAT model data pre-process needs to execute the preprocess scripts in ai_benchmark/s100/qat/tools/eval_preprocess in the x86 environment. The datasets corresponding to each model in the sample package are described in detail as below, as well as the pre-processing operations for the corresponding datasets.

Tip

Before use, please modify the dataset path and save path in the script to make the script run properly.

DatasetDescription

ImageNet

Dataset

The ImageNet dataset is used for evaluating the mixvargenet_imagenet, mobilenetv1_imagenet, mobilenetv2_imagenet, resnet18_imagenet, resnet50_imagenet, horizon_swin_transformer_imagenet, vit_small_imagenet and vargnetv2_imagenet QAT models.

Dataset preprocessing methods:

python3 imagenet.py --image-path=./standard_imagenet/val/ \ --save-path=./pre_model_name

COCO

Dataset

The COCO dataset is used for evaluating the fcos_efficientnetb0_mscoco, detr_resnet50_mscoco, detr_efficientnetb3_mscoco and deform_detr_resnet50_mscoco QAT models.

Dataset preprocessing methods:

  • fcos_efficientnetb0

    python3 fcos_process.py --image-path=./mscoco/images/val2017/ \ --label-path=./mscoco/images/annotations/instances_val2017.json \ --save-path=./pre_fcos_efficientnetb0

  • detr_resnet50_mscoco, detr_efficientnetb3_mscoco, deform_detr_resnet50_mscoco

    python3 detr_process.py --image-path=./mscoco/images/val2017/ \ --label-path=./mscoco/images/annotations/instances_val2017.json \ --save-path=./pre_detr

Cityscapes

Dataset

The Cityscapes dataset is used for the evaluation of the QAT segmentation model unet_mobilenetv1_cityscapes, and no preprocessing is required to use the validation set data directly.

Kitti3D

Dataset

The Kitti3D dataset is used for evaluating the pointpillars_kitti_car model.

1. You can download this dataset from the official website: Kitti3D dataset official website download address. We recommend you download the following zip files. Please contact Horizon if you encounter any problems during data preparation.

. ├── kitti3d ├── data_object_calib.zip # camera calibration matrices of object data set ├── data_object_image_2.zip # left color images of object data set ├── data_object_label_2.zip # taining labels of object data set └── data_object_veloodyne.zip # velodyne point cloud

We recommend that you unzip the downloaded dataset into the following structure.

. ├── kitti3d_origin ├── ImageSets │ ├── test.txt │ ├── train.txt │ ├── trainval.txt │ └── val.txt ├── testing │ ├── calib │ ├── image_2 │ └── velodyne └── training ├── calib ├── image_2 ├── label_2 └── velodyne

2. Dataset preprocessing methods:

python3 pointpillars_process.py --data-path=./kitti3d_origin \ --save-path=./pre_kitti3d \ --height=1 --width=150000

Culane

Dataset

The Culane dataset is used for evaluating the ganet_mixvargenet_culane model.

1. You can download this dataset from the official website: Culane dataset official website download address. We recommend you download the following zip files. Please contact Horizon if you encounter any problems during data preparation.

. ├── culane ├── annotations_new.tar.gz ├── driver_23_30frame.tar.gz ├── driver_37_30frame.tar.gz ├── driver_100_30frame.tar.gz ├── driver_161_90frame.tar.gz ├── driver_182_30frame.tar.gz ├── driver_193_90frame.tar.gz ├── laneseg_label_w16.tar.gz └── list.tar.gz

In which, the annotations_new.tar.gz needs to be unpacked last in order to make corrections to the original annotations file. We recommend that you unzip the downloaded dataset into the following structure.

. ├── culane # root directory ├── driver_23_30frame # datasets and annotations │ ├── 05151640_0419.MP4 # a section of the dataset, containing each frame of the picture │ │ ├──00000.jpg # source images │ │ ├──00000.lines.txt # annotation file where each line gives the x,y coordinates of the lane marking keypoints │ ...... ├── driver_37_30frame ├── driver_100_30frame ├── driver_161_90frame ├── driver_182_30frame ├── driver_193_90frame ├── laneseg_label_w16 # lane segment labels └── list # train, validate, test list

2. Dataset preprocessing methods:

python3 ganet_process.py --image-path=./culane \ --save-path=./pre_culane

Nuscenes

Dataset

The Nuscenes dataset is used for evaluating the fcos3d_efficientnetb0_nuscenes, centerpoint_pointpillar_nuscenes, centerpoint_mixvargnet_multitask_nuscenes, bev_gkt_mixvargenet_multitask_nuscenes, bev_lss_efficientnetb0_multitask_nuscenes, bev_ipm_efficientnetb0_multitask_nuscenes, bev_ipm_4d_efficientnetb0_multitask_nuscenes, bevformer_tiny_resnet50_detection_nuscenes, maptroe_henet_tinym_bevformer_nuscenes, detr3d_efficientnetb3_nuscenes, petr_efficientnetb3_nuscenes and flashocc_henet_lss_occ3d_nuscenes models.

1. You can download this dataset from the official website: Nuscenes dataset official website download address. We recommend you download the following zip files. Please contact Horizon if you encounter any problems during data preparation.

. ├── Nuscenes ├── nuScenes-map-expansion-v1.3.zip ├── nuScenes-map-expansion-v1.2.zip ├── nuScenes-map-expansion-v1.1.zip ├── nuScenes-map-expansion-v1.0.zip ├── v1.0-mini.tar ├── v1.0-test_blobs.tar ├── v1.0-test_meta.tar ├── v1.0-trainval01_blobs.tar ├── v1.0-trainval02_blobs.tar ├── v1.0-trainval03_blobs.tar ├── v1.0-trainval04_blobs.tar ├── v1.0-trainval05_blobs.tar ├── v1.0-trainval06_blobs.tar ├── v1.0-trainval07_blobs.tar ├── v1.0-trainval08_blobs.tar ├── v1.0-trainval09_blobs.tar ├── v1.0-trainval10_blobs.tar └── v1.0-trainval_meta.tar

For the lidar multitask model, you have to download the lidar segmentation tag lidarseg from the official website as well, and follow the tutorial on the official nuscenes website to update the v1.0-trainval. We recommend that you unzip the downloaded dataset into the following structure.

. ├── Nuscenes ├── can_bus ├── lidarseg ├── maps ├── nuscenes │ └── meta │ ├── maps │ ├── v1.0-mini │ └── v1.0-trainval ├── samples ├── sweeps ├── v1.0-mini └── v1.0-trainval

For the online map construction model, you also need to download .shp format map files for the following regions from OpenStreetMap websize: Boston Seaport, Singapore One-North, Singapore Holland Village, and Singapore Queenstown.

For the Occupancy model, you have to download the "gts" from CVPR2023-3D-Occupancy-Prediction, We recommend that you unzip the downloaded dataset into the following structure.

. ├── Nuscenes ├── maps ├── samples ├── sweeps ├── v1.0-mini ├── v1.0-trainval └── gts (new)

2. Dataset preprocessing methods:

Attention:

  • In addition to generating pre-processed images, the fcos3d_process.py will also process the internal parameters of the camera to be used, and generate the corresponding camera internal reference configuration files.
  • In addition to generating preprocessed data, centerpoint_preprocess.py, the bev_preprocess.py and lidar_preprocess.py will also generate a val_gt_infos.pkl file under the preprocessed data path for precision calculation.
  • In addition to generating preprocessed data, maptr_preprocess.py also generate a nuscenes_map_anns_val.json file under the preprocessed data path for precision calculation. Also, it needs to specify the name of the model through --model, the option is maptroe_henet_tinym_bevformer_nuscenes.
  • The bev_preprocess.py needs to specify the name of the model through --model, the options are bev_gkt_mixvargenet_multitask_nuscenes, bev_lss_efficientnetb0_multitask_nuscenes, bev_ipm_efficientnetb0_multitask_nuscenes, bev_ipm_4d_efficientnetb0_multitask_nuscenes, bevformer_tiny_resnet50_detection_nuscenes, detr3d_efficientnetb3_nuscenes, petr_efficientnetb3_nuscenes, flashocc_henet_lss_occ3d_nuscenes.

The reference commands are as follows:

  • fcos3d_efficientnetb0_nuscenes

    python3 fcos3d_process.py --src-data-dir=./Nuscenes \ --file-path=../../script/config/model/data_name_list/nuscenes_names.txt \ --save-path=./processed_fcos3d_images

  • centerpoint_pointpillar_nuscenes

    python3 centerpoint_preprocess.py --data-path=./Nuscenes \ --save-path=./nuscenes_lidar_val

  • bev model

    python3 bev_preprocess.py --model=model_name \ --data-path=./Nuscenes \ --meta-path=./Nuscenes/meta \ --reference-path=../../script/config/reference_points \ --save-path=./nuscenes_bev_val

  • centerpoint_mixvargnet_multitask_nuscenes

    python3 lidar_preprocess.py --data-path=./Nuscenes \ --save-path=./nuscenes_lidar_val

  • flashocc_henet_lss_occ3d_nuscenes

    python3 bev_preprocess.py --model=model_name \ --data-path=./Nuscenes \ --meta-path=./Nuscenes/meta \ --reference-path=../../script/config/reference_points \ --save-path=./nuscenes_bev_val --with-occ-gt

  • maptroe_henet_tinym_bevformer_nuscenes

    python3 maptr_preprocess.py --model=model_name \ --data-path=./Nuscenes \ --meta-path=./Nuscenes/meta \ --sdmap-path=./data_osm \ --reference-path=../../script/config/reference_points \ --save-path=./nuscenes_map_val

Mot17

Dataset

The Mot17 dataset is used for evaluating the motr_efficientnetb3_mot17 model.

1. You can download this dataset from the official website: Mot17 dataset official website download address. We recommend that you unzip the downloaded dataset into the following structure. Please contact Horizon if you encounter any problems during data preparation.

. ├── valdata # root directory ├── gt_val │ ├── MOT17-02-SDP │ ├── MOT17-04-SDP │ ├── MOT17-05-SDP │ ├── MOT17-09-SDP │ ├── MOT17-10-SDP │ ├── MOT17-11-SDP │ ├── MOT17-13-SDP ├── images │ └── train │ ├── MOT17-04-SDP │ ├── MOT17-05-SDP │ ├── MOT17-09-SDP │ ├── MOT17-10-SDP │ ├── MOT17-11-SDP │ ├── MOT17-13-SDP └── mot17.val

2. Dataset preprocessing methods:

python3 motr_process.py --image-path=./valdata/images/train \ --save-path=./processed_motr

Carfusion

Dataset

The Carfusion dataset is used for evaluating the keypoint_efficientnetb0_carfusion model.

1. You can download this dataset from the official website: Carfusion dataset official website download address. We recommend that you unzip the downloaded dataset into the following structure. Please contact Horizon if you encounter any problems during data preparation.

. ├── carfusion # root directory ├── train └── test

2. Dataset preprocessing methods:

# First generate the data required for evaluation(If you are using this dataset for the first time, # you must use the following script to generate the data required for the evaluation) python3 gen_carfusion_data.py --src-data-path=carfusion \ --out-dir=cropped_data \ --num-workers 2

The directory after executing the first script is as follows:

. ├── cropped_data # root directory ├── test └── simple_anno

Ensure that the data root address is at the same level as the cropped_data, and then run the following command:

python3 keypoints_preprocess.py --data-root=./ \ --label-path=cropped_data/simple_anno/keypoints_test.json \ --save-path=./processed_carfusion

Argoverse1

Dataset

The Argoverse1 dataset is used for evaluating the densetnt_argoverse1 model.

1. You can download this dataset from the official website: Argoverse1 dataset official website download address. We recommend that you unzip the downloaded dataset into the following structure. Please contact Horizon if you encounter any problems during data preparation.

. ├── arogverse-1 # root directory ├── map_files └── val

2. Dataset preprocessing methods:

The densetnt_process.py generates the appropriate review meta file under src-path in addition to the preprocessed input. The evaluation only needs to use the two parameters --src-path and --dst-path, and there is no need to pay attention to other parameters.

python3 densetnt_process.py --src-path=arogverse-1 \ --dst-path=processed_arogverse1

Argoverse2

Dataset

The Argoverse2 dataset is used for evaluating the qcnet_oe_argoverse2 model.

1. You can download this dataset from the official website: Argoverse2 dataset official website download address. Please download val.tar. We recommend that you unzip the downloaded dataset into the following structure. Please contact Horizon if you encounter any problems during data preparation.

. |-- argoverse-2 # root directory |-- val

2. Dataset preprocessing methods:

The required data for evaluation can be generated using qcnet_argoverse2_preprocess.py and qcnet_argoverse2_dataset.py. You only need to use the --src-data-dir and --target-data-dir parameters, without the need to pay attention to other parameters.

python3 qcnet_argoverse2_preprocess.py --src-data-dir arogverse-2 \ --target-data-dir processed_arogverse2 \ --pack-type lmdb --num-samples 24988

SceneFlow

Dataset

The SceneFlow dataset is used for evaluating the stereonetplus_mixvargenet_sceneflow model.

1. You can download this dataset from the official website: SceneFlow dataset official website download address. We recommend that you unzip the downloaded dataset into the following structure. Please contact Horizon if you encounter any problems during data preparation.

. ├── SceneFlow # root directory ├── FlyingThings3D │ ├── disparity │ ├── frames_finalpass └── SceneFlow_finalpass_test.txt

2. Dataset preprocessing methods:

In addition to generating preprocessed data, the stereonet_preprocess.py, bev_preprocess.py and lidar_preprocess.py will also generate a val_gt_infos.pkl file under the preprocessed data path for precision calculation.

python3 stereonet_preprocess.py --data-path=SceneFlow/ \ --data-list=SceneFlow/SceneFlow_finalpass_test.txt \ --save-path=sceneflow_val

Model Mounting

Because datasets are huge, it is recommended to mount them for dev board to load, rather than to copy them into the dev board, you need to do the following on the server PC terminal and the board terminal:

Server PC terminal:

Attention

Note that the root permission is required on the server pc terminal to perform the following actions.

  1. Edit one line into /etc/exports: /nfs *(insecure,rw,sync,all_squash,anonuid=1000,anongid=1000,no_subtree_check). Wherein, /nfs denotes mounting path of local machine, it can be replaced by the directory you specify.

  2. Run exportfs -a -r to bring /etc/exports into effect.

Board terminal:

  1. Create the directory to be mounted: mkdir -p /mnt.

  2. Mount: mount -t nfs {PC terminal IP}:/nfs /mnt -o nolock.

Mount the /nfs folder at PC terminal to the /mnt folder in dev board. In this way, mount the folder in which contains preprocessed folder to dev board and create a soft link of /data folder in the /ptq or /qat folder (at the same directory level as /script) in dev board.

Generate lst Files

The running process of precision calculation script in the sample is:

  1. According to the value of image_list_file in workflow_accuracy.json, find the lst file of the corresponding dataset.

  2. Load each preprocessing file according to the path information of preprocessing file stored in lst file, and then perform the inference.

Therefore, after generating the preprocessing file, you need to generate the corresponding LST file, and write the path of each preprocessing file into the lst file, which is related to the storage location of the dataset at the board end. Here, we recommend that its storage location shall be the same level as the ./data/dataset_name/pre_model_name folder.

The structure of the PTQ pre-processed dataset is as follows:

|── ptq | |── data | | |── cityscapes | | | |── pre_deeplabv3plus_efficientnetb0 | | | | |── xxxx.bin # pre-processed binary file | | | | |── .... | | | |── pre_deeplabv3plus_efficientnetb0.lst # lst file: record the path of each preprocessing file | | | |── .... | | |── .... | |── model | | |── ... | |── script | | |── ...

The structure of the QAT pre-processed data set is as follows:

|── qat | |── data | | |── carfusion | | | |── pre_keypoints | | | | |── xxxx # pre-processed data | | | | |── .... | | | |── pre_carfusion.lst # lst file: record the path of each preprocessing file | | |── .... | |── model | | |── ... | |── script | | |── ...

The corresponding lst file is generated by reference as follows:

Except for the densetnt_argoverse1, qcnet_oe_argoverse2, bev, motr_efficientnetb3_mot17, stereonetplus_mixvargenet_sceneflow and maptroe_henet_tinym_bevformer_nuscenes models, the reference generation method of the lst files for the other models:

find ../../../data/coco/pre_centernet_resnet101 -name "*bin*" > ../../../data/coco/pre_centernet_resnet101.lst
Note

The parameters after -name need to be adjusted according to the format of the preprocessed dataset, such as bin, png.

The path stored in the generated lst file is a relative path: ../../../data/coco/pre_centernet_resnet101/ , which can match the workflow_accuracy.json default configuration path.

If you need to change the storage location of the preprocessing dataset, you need to ensure that the corresponding lst file can be used by workflow_accuracy.json. Secondly, it is necessary to ensure that the program can read the corresponding preprocessing file according to the path information in lst file.

For the densetnt_argoverse1, qcnet_oe_argoverse2, bev, motr_efficientnetb3_mot17, stereonetplus_mixvargenet_sceneflow and maptroe_henet_tinym_bevformer_nuscenes models, the reference generation method of the lst files:

Model NameReference Generation Method and Description of lst File
densetnt_argoverse1

Reference generation method:

sh generate_acc_lst.sh

The path stored in the generated lst file is a relative path: ../../../data/argoverse1/densetnt/, which can match the workflow_accuracy.json default configuration path.

qcnet_oe_argoverse2

We provide a dedicated script, generate_acc_lst.sh, for generating lst files. The usage is as follows:

sh generate_acc_lst.sh ../../../data/argoverse2_qcnet/fp32_sample/val \ ../../../data/argoverse2_qcnet

This script requires two input parameters: preprocessed data mount path and output path for the generated list. The resulting lst file will store paths as relative paths: ../../../data/argoverse2_qcnet/, which can align with the default configuration path of workflow_accuracy.json.

Bev

maptroe_henet_tinym_bevformer_nuscenes

Take the bev_ipm_efficientnetb0_multitask_nuscenes as an example. This model has two types of input: images and reference points. The input image and reference point of the same frame have the same name. In order to ensure that the input corresponds, you need to add sort to sort by name when executing the find command, same with maptroe_henet_tinym_bevformer_nuscenes. Reference generation method:

find ../../../data/nuscenes_bev/images -name "*bin*" | \ sort > ../../../data/nuscenes_bev/images.lst find ../../../data/nuscenes_bev/reference_points0 -name "*bin*" | \ sort > ../../../data/nuscenes_bev/reference_points0.lst

In addition, bev_ipm_4d_efficientnetb0_multitask_nuscenes is a timing model, which requires input order. Therefore, we provide a script gen_file_list.sh for generating lst files, the usage is as follows:

sh gen_file_list.sh

The path stored in the generated lst file is a relative path: ../../../data/nuscenes_bev/ , which can match the workflow_accuracy.json default configuration path.

If you need to change the storage location of the preprocessing dataset, you need to ensure that the corresponding lst file can be used by workflow_accuracy.json. Secondly, it is necessary to ensure that the program can read the corresponding preprocessing file according to the path information in lst file.

motr_efficientnetb3_mot17

Reference generation method:

sh generate_acc_lst.sh

The path stored in the generated lst file is a relative path: ../../../data/mot17/motr/ , which can match the workflow_accuracy.json default configuration path.

If you need to change the storage location of the preprocessing dataset, you need to ensure that the corresponding lst file can be used by workflow_accuracy.json.

Secondly, it is necessary to ensure that the program can read the corresponding preprocessing file according to the path information in lst file.

stereonetplus_mixvargenet_sceneflow

Take the stereonetplus_mixvargenet_sceneflow as an example. The input left image and right the same frame have the same name. In order to ensure that the input corresponds, you need to add sort to sort by name when executing the find command.

find ../../../data/sceneflow/left -name "*png*" | \ sort > ../../../data/sceneflow/left.lst find ../../../data/sceneflow/right -name "*png*" | \ sort > ../../../data/sceneflow/right.lst

Model Inference

The accuracy.sh script is shown as below:

#!/bin/sh source ../../base_config.sh # load basic configurations export SHOW_FPS_LOG=1 # specify environment variable, print fps level log ${app} \ # executable program defined in the accuracy.sh script --config_file=workflow_accuracy.json \ # load workflow configuration file of accuracy evaluation --log_level=2 # specify log level

After the data has been mounted, log in dev board and run the accuracy.sh script in the centernet_resnet101 directory, as shown below:

root@dvb-hynix8G:/userdata/ptq/script/detection/centernet_resnet101# sh accuracy.sh ../../aarch64/bin/example --config_file=workflow_accuracy.json --log_level=2 ... I0419 03:14:51.158655 39555 infer_method.cc:107] Predict DoProcess finished. I0419 03:14:51.187361 39556 ptq_centernet_post_process_method.cc:558] PTQCenternetPostProcessMethod DoProcess finished, predict result: [{"bbox":[-1.518860,71.691170,574.934631,638.294922],"prob":0.750647,"label":21,"class_name":" I0118 14:02:43.636204 24782 ptq_centernet_post_process_method.cc:558] PTQCenternetPostProcessMethod DoProcess finished, predict result: [{"bbox":[3.432283,164.936249,157.480042,264.276825],"prob":0.544454,"label":62,"class_name":" ...

Inference results will be saved into the eval.log file dumped by dev board program.

Model Accuracy Computing

Attention

Please perform the accuracy calculation in docker environment or Linux environment.

Accuracy computing is presented in two cases: PTQ model accuracy computing and QAT model accuracy computing.

PTQ Model Accuracy Computing:

For the PTQ model, the model accuracy computing scripts are placed under the ptq/tools/python_tools/accuracy_tools folder, in which:

ScriptDescription
cls_imagenet_eval.pyThe script is used for computing accuracy of classification models evaluated using the ImageNet dataset.
det_coco_eval.pyThe script is used for computing the accuracy of models evaluated using the COCO dataset.
seg_cityscapes_eval.pyThe script is used for computing the accuracy of segmentation models evaluated using the Cityscapes dataset.
det_voc_eval.pyThe script is used for computing the accuracy of detection models using the VOC dataset.

Below we provide you with the description of the different types of PTQ model accuracy computing:

Model TypeDescription of the Accuracy Computing
Classification Model

Method to compute the accuracy of those models using the ImageNet datasets is shown as below:

python3 cls_imagenet_eval.py --eval-log-path=eval.log \ --gt-file=val.txt

In which:

  • eval-log-path refers to inference result file of classification models.
  • gt-file refers to the annotation files of ImageNet datasets.
Detection Model

1.Method to compute the accuracy of those models using the COCO dataset is shown as below:

python3 det_coco_eval.py --eval-log-path=eval.log \ --annotation-path=instances_val2017.json

In which:

  • eval-log-pathrefers to inference result file of detection models.
  • annotation-pathrefers to the annotation file of of the COCO dataset.

2.Method to compute the accuracy of those detection models using the VOC dataset is shown as below:

python3 det_voc_eval.py --eval-log-path=eval.log \ --annotation-path=../Annotations \ --val-txt-path=../val.txt

In which:

  • eval-log-path refers to the inference result file of detection models.
  • annotation-path refers to the annotation file of the VOC dataset.
  • val-txt-path refers to the val.txt file in the .../ImageSets/Main folder in the VOC dataset.
Segmentation Model

Method to compute the accuracy of those segmentation models using the Cityscapes dataset is shown as below:

python3 seg_cityscapes_eval.py --eval-log-path=eval.log \ --gt-path=cityscapes/gtFine/val

In which:

  • eval-log-path refers to the inference result file of segmentation models.
  • gt-path refers to the annotation file of the Cityscapes dataset.

QAT Model Accuracy Computing:

For the QAT model, the model accuracy computing scripts are placed under the qat/tools/python_tools/accuracy_tools folder, in which:

ScriptDescription
bev_eval.pyThe script is used to calculate the accuracy of bev model.
centerpoint_eval.pyThe script is used to calculate the accuracy of centerpoint_pointpillar_nuscenes lidar 3D model.
cls_eval.pyThe script is used to calculate the accuracy of the classification model.
densetnt_eval.pyThe script is used to calculate the accuracy of densetnt_argoverse1 model.
qcnet_eval.pyThe script is used to calculate the accuracy of qcnet_oe_argoverse2 model.
detr_eval.pyThe script is used to calculate the accuracy of detr and deform_detr_resnet50_mscoco model.
fcos3d_eval.pyThe script is used to calculate the accuracy of fcos3d_efficientnetb0_nuscenes model.
fcos_eval.pyThe script is used to calculate the accuracy of fcos model.
ganet_eval.pyThe script is used to calculate the accuracy of ganet_mixvargenet_culane model.
keypoints_eval.pyThe script is used to calculate the accuracy of keypoints model.
lidar_multitask_eval.pyThe script is used to calculate the accuracy of lidar centerpoint_mixvargnet_multitask_nuscenes model.
motr_eval.pyThe script is used to calculate the accuracy of motr_efficientnetb3_mot17 model.
parsing_eval.pyThe script is used to calculate the accuracy of segmentation model.
pointpillars_eval.pyThe script is used to calculate the accuracy of pointpillars model.
stereonet_eval.pyThe script is used to calculate the accuracy of stereonet_plus model.
maptroe_eval.pyThe script is used to calculate the accuracy of maptr model.
flashocc_eval.pyThe script is used to calculate the accuracy of flashocc model.

Below we provide you with the description of the different types of QAT model accuracy computing:

Model TypeDescription of the Accuracy Computing
Bev Model

1.Method to compute the accuracy of those models using the Nuscenes dataset is shown as below:

python3 bev_eval.py --det_eval_path=bev_det_eval.log \ --seg_eval_path=bev_seg_eval.log \ --gt_files_path=./nuscenes_bev_val/val_gt_infos.pkl \ --meta_dir=./Nuscenes/meta/

In which:

  • det_eval_path refers to detection inference result file of bev models.
  • seg_eval_path refers to segment inference result file of bev models.
  • gt_files_path refers to gt file generated by preprocessing the nuscenes dataset.
  • meta_dir refers to Nuscenes dataset meta data.

notice：detr3d_efficientnetb3_nuscenes, petr_efficientnetb3_nuscenes and bevformer_tiny_resnet50_detection_nuscenes does not require seg_eval_path parameter.

2.Method to compute the accuracy of flashocc using the Nuscenes dataset is shown as below:

python3 flashocc_eval.py --seg_eval_path=bev_seg_eval.log \ --gt_files_path=./nuscenes_bev_val/val_gt_infos.pkl

In which:

  • seg_eval_path refers to inference result file of flashocc models.
  • gt_files_path refers to gt file generated by preprocessing the nuscenes dataset with occ gts.
Classification Model

Method to compute the accuracy of those models using the CIFAR-10 and ImageNet datasets is shown as below:

python3 cls_eval.py --log_file=eval.log \ --gt_file=val.txt

In which:

  • log_file refers to inference result file of classification models.
  • gt_file refers to the annotation files of CIFAR-10 and ImageNet datasets.
Detection Model

1. Method to compute the accuracy of those models using the COCO dataset is shown as below:

python3 fcos_eval.py --eval_result_path=eval.log \ --annotation_path=instances_val2017.json \ --image_path=./mscoco/images/val2017/ # The qat fcos model needs to add --is_qat=True python3 fcos_eval.py --eval_result_path=eval.log \ --annotation_path=instances_val2017.json \ --image_path=./mscoco/images/val2017/ \ --is_qat=True

In which:

  • eval_result_path refers to inference result file of detection model fcos.
  • annotation_path refers to the annotation file of of the COCO dataset.
  • image_path refers to COCO dataset source data.
  • is_qat refers to whether result evaluation of qat fcos model.
python3 retinanet_eval.py --eval_result_path=eval.log \ --annotation_path=instances_val2017.json \ --image_path=./mscoco/images/val2017/

In which:

  • eval_result_path refers to inference result file of detection model retinanet.
  • annotation_path refers to the annotation file of of the COCO dataset.
  • image_path refers to COCO dataset source data.
python3 detr_eval.py --eval_result_path=eval.log \ --annotation_path=instances_val2017.json \ --image_path=./mscoco/images/val2017/

In which:

  • eval_result_path refers to inference result file of detection model retinanet.
  • annotation_path refers to the annotation file of of the COCO dataset.
  • image_path refers to COCO dataset source data.

2. Method to compute the accuracy of those detection models using the KITTI dataset is shown as below:

python3 pointpillars_eval.py --eval_result_path=eval.log \ --annotation_path=./val_gt_infos.pkl

In which:

  • eval_result_path refers to the inference result file of detection models.
  • annotation_path refers to the annotation file val_gt_infos.pkl of the KITTI dataset.

3. Method to compute the accuracy of those detection models using the Culane dataset is shown as below:

python3 ganet_eval.py --eval_path=eval.log \ --image_path=./culane

In which:

  • eval_result_path refers to the inference result file of detection models.
  • image_path refers to the Culane dataset.

4. Method to compute the accuracy of those models using the Nuscenes dataset is shown as below:

python3 fcos3d_eval.py --eval_result_path=eval.log \ --image_path=./Nuscenes

In which:

  • eval_result_pathrefers to inference result file of detection models.
  • image_path refers to Nuscenes dataset source data.
python3 centerpoint_eval.py --predict_result_path=eval.log \ --gt_files_path=./nuscenes_lidar_val/val_gt_infos.pkl \ --meta_dir=./Nuscenes/meta/

In which:

  • predict_result_path refers to inference result file of detection models.
  • gt_files_path refers to gt file generated by preprocessing the nuscenes dataset.
  • meta_dir refers to Nuscenes dataset meta data.

5. Method to compute the accuracy of those detection models using the Carfusion dataset is shown as below:

python3 keypoints_eval.py --anno_path=./processed_carfusion/processed_anno.json \ --eval_result_path=eval.log

In which:

  • anno_path refers to the processed_anno.json file generated by preprocessing.
  • eval_result_path refers to inference result file of detection models.
Segmentation Model

Method to compute the accuracy of those segmentation models using the Cityscapes dataset is shown as below:

python3 parsing_eval.py --log_file=eval.log \ --gt_path=cityscapes/gtFine/val

In which:

  • log_file refers to the inference result file of segmentation models.
  • gt_path refers to the annotation file of the Cityscapes dataset.
Tracking Model

Method to compute the accuracy of those models using the mot17 datasets is shown as below:

python3 motr_eval.py --eval_result_path=eval_log \ --gt_val_path=valdata/gt_val

In which:

  • eval_result_path refers to inference result file of tracking models.
  • gt_val_path refers to the annotation files of motr17 datasets.
Multitask Model

Method to compute the accuracy of those models using the Nuscenes dataset is shown as below:

python3 lidar_multitask_eval.py --det_eval_path=det_eval.log \ --seg_eval_path=seg_eval.log \ --gt_files_path=./nuscenes_lidar_val/val_gt_infos.pkl \ --data_dir=./Nuscenes

In which:

  • det_eval_path refers to detection inference result file of models.
  • seg_eval_path refers to segment inference result file of models.
  • gt_files_path refers to gt file generated by preprocessing the nuscenes dataset.
  • data_dir refers to Nuscenes dataset.
Traj Pred Model

1. Method to compute the accuracy of those models using the argoverse1t datasets is shown as below:

python3 densetnt_eval.py --eval_result_path=eval.log \ --meta_path=argoverse1/meta

In which:

  • eval_result_path refers to inference result file of traj pred models.
  • meta_pathrefers to the annotation file generated by the argoverse1 dataset, after using the preprocessing script, it is generated in the original dataset directory meta.

2. Method to compute the accuracy of those models using the argoverse2 datasets is shown as below:

python3 qcnet_eval.py --eval_path \ --gt_files_path \ --msg_pack_path

In which:

  • eval_path: refers to detection inference result file of models.
  • gt_files_path: Refers to gt file generated by preprocessing the argoverse2 dataset.The qcnet_eval.py evaluation script will automatically retrieve the corresponding labels for the samples from the data folder.
  • msg_pack_path: The path where the trajectory prediction results from the parsed eval.log file are stored as msgpack. After the script parses eval.log, it will save the data as the file <msg_pack_path>qcnet_predict_data.msgpack. The script will subsequently read the msgpack file for evaluation.
Disparity Pred Model

Method to compute the accuracy of those models using the Sceneflow datasets is shown as below:

python3 stereonet_eval.py --log_file=eval.log \ --gt_file=val_gt_infos.pkl

In which:

  • log_filerefers to inference result file of disparity pred models.
  • gt_filerefers to gt file generated by preprocessing the Sceneflow dataset.
Maptr Model

Method to compute the accuracy of those models using the Nuscenes datasets is shown as below:

python3 maptroe_eval.py --predict_result_path=eval.log \ --gt_files_path=nuscenes_map_anns_val.json

In which:

  • predict_result_pathrefers to inference result file of maptr model.
  • gt_files_pathrefers to gt file generated by preprocessing the Nuscenes dataset.

Model Integration

Pre-processing

You can add model pre-processing as needed and deploy it to CPU or DSP, taking centerpoint_pointpillar_nuscenes as an example:

  1. Add the preprocessing file qat_centerpoint_preprocess_method.cc and the header file qat_centerpoint_preprocess_method.h.

  2. Add model preprocessing configuration file.

Add Pre-processing File and Header File

The pre-processing qat_centerpoint_preprocess_method.cc files are placed under the ai_benchmark/code/src/method/ folder. While the header file qat_centerpoint_preprocess_method.h files are placed under the ai_benchmark/code/include/method/ floder.

|── ai_benchmark | |── code # source code of samples | | |── include | | | |── method # add your header files into this folder | | | | |── qat_centerpoint_preprocess_method.h | | | | |── ... | | |── src | | | |── method # add your .cc preprocess files into this folder | | | | |── qat_centerpoint_preprocess_method.cc | | | | |── ...

Add Preprocessing Configuration File

|── ai_benchmark | |── s100/qat/script # sample script folder | | |── config | | | |── preprocess | | | | |── centerpoint_preprocess_5dim.json # pre-processing configuration file

Preprocess of centerpoint_pointpillar_nuscenes can deploy to CPU or DSP depends on whether you config the run_on_dsp parameter in the centerpoint_pointpillar_5dim.json. If run_on_dsp in config file is set to true then preprocess will be running on DSP otherwise it running on CPU.

To Evalute the Latency of Preprocessing

Run sh latency.sh to evaluate single frame latency of preprocess, as shown below:

I0807 15:21:15.665417 21457 output_plugin.cc:95] Pre process latency: [avg: 11.338ms, max: 14.744ms, min: 8.200ms], Infer latency: [avg: 49.327ms, max: 51.065ms, min: 48.409ms], Post process latency: [avg: 9.681ms, max: 9.886ms, min: 9.544ms].

In which:

  • Pre process denotes the time consumption of pre-processing.
  • Infer denotes the time consumption of model inference.
  • Post process denotes the time consumption of post-processing.

Postprocessing

Post-processing consists of 2 steps. Let's take integration of CenterNet model as an example:

  1. Add the post-processing file ptq_centernet_post_process_method.cc and the header file ptq_centernet_post_process_method.h.

  2. Add a model execution script and a configuration file.

Add Post-processing File

Post-processing code file can reuse any post-processing files in the src/method directory. You only need to modify the InitFromJsonString function and the PostProcess function.

The InitFromJsonString function is used for loading the post-processing related parameters in the workflow.json. You can customize the corresponding input parameters. The PostProcess function is used for implementing post-processing logic.

The post-processing ptq_centernet_post_process_method.cc files are placed under the ai_benchmark/code/src/method/ folder. While the header files ptq_centernet_post_process_method.h are placed under the ai_benchmark/code/include/method/ folder.

|── ai_benchmark | |── code # source code of samples | | |── include | | | |── method # add your header files into this folder | | | | |── ptq_centernet_post_process_method.h | | | | |── ... | | |── src | | | |── method # add your .cc postprocess files into this folder | | | | |── ptq_centernet_post_process_method.cc | | | | |── ...

Add Model Execution and Configuration Files

Directory structure of scripts is shown as below:

  • The centerpoint_pointpillar_nuscenes model:

    |── ai_benchmark | |── s100/qat/script # sample script folder | | |── detection | | | |── centerpoint_pointpillar_nuscenes | | | | |── accuracy.sh # accuracy evaluation script | | | | |── fps.sh # performance evaluation script | | | | |── latency.sh # single-frame latency sample script | | | | |── README.md # dsp deployment introduction document | | | | |── workflow_accuracy # accuracy configuration folder | | | | |── workflow_fps.json # performance configuration file | | | | |── workflow_latency.json # single-frame latency configuration file

    To process on DSP, you need to execute dsp_deploy.sh to deploy the DSP environment. For a detailed introduction to dsp deployment, please refer to README.md.

  • The motr_efficientnetb3_mot17 model:

    |── ai_benchmark | |── s100/qat/script # sample script folder | | |── tracking | | | |── motr_efficientnetb3_mot17 | | | | |── accuracy.sh # accuracy evaluation script | | | | |── fps.sh # performance evaluation script | | | | |── generate_acc_lst.sh # generate accuracy lst script | | | | |── latency.sh # single-frame latency sample script | | | | |── workflow_accuracy # accuracy configuration folder | | | | |── workflow_fps.json # performance configuration file | | | | |── workflow_latency.json # single-frame latency configuration file

  • The models except for the centerpoint_pointpillar_nuscenes and motr_efficientnetb3_mot17:

    |── ai_benchmark | |── s100/ptq/script # sample script folder | | |── detection | | | |── centernet_resnet101 | | | | |── accuracy.sh # accuracy evaluation script | | | | |── fps.sh # performance evaluation script | | | | |── latency.sh # single-frame latency sample script | | | | |── workflow_accuracy.json # accuracy configuration file | | | | |── workflow_fps.json # performance configuration file | | | | |── workflow_latency.json # single-frame latency configuration file

Helper Tools

Log

There are 2 types of logs: sample Log and DNN Log. Wherein, sample log refers to the log in the AI Benchmark Sample Package deliverables, while DNN log refers to the log in the embedded runtime library. You can specify logs as needed.

Sample Log

  1. Log level.

Both glog and vlog are used in sample log and there are 4 customized log levels:

  • 0: SYSTEM level, this log level is used for generating error information in sample code.
  • 1: REPORT level, this log level is used for generating performance data in sample code.
  • 2: DETAIL level, this log level is used for generating current system status in sample code.
  • 3: DEBUG level, this log level is used for generating debugging information in sample code.
  1. Set log levels.

Rules to set log levels: The default ranks of log level: DEBUG>DETAIL>REPORT>SYSTEM, the higher the level, the more logs will be output. That is, if you set a high level, the logs corresponding to your own level and the level below it will be output.

When running samples, specify the log_level parameter to set log levels. For example, if log_level=0, then SYSTEM log should be dumped; else if log_level=3, then DEBUG, DETAIL, REPORT and SYSTEM logs should be dumped.

dnn Log

For the configuration of dnn logs, please read the Configuration Info section in the Model Inference API Instruction.

OP Time Consumption

Overview

Use the HB_DNN_PROFILER_LOG_PATH environment variable to specify statistics of OP performance. Types and values of this environment variable are described as below:

HB_DNN_PROFILER_LOG_PATH=${path}: denotes the output path of OP node. After the program is executed, a profiler.log file should be generated.

Sample

Taking the mobilenetv1 as an example, as shown in the following code block: Start 1 threads to run the model at the same time, set export HB_DNN_PROFILER_LOG_PATH=. /, then the profiler.log file will output the performance data of the OPs.

The output information contains model_latency and task_latency. Wherein, model_latency contains the time consumption required to run each operator of the model; while task_latency contains the time consumption of each task of the model.

Dump Tool

Enable the HB_DNN_DUMP_PATH environment variable to dump the input and output of each node in inference process. The dump tool can check if there are consistency problems between simulator and real machine, i.e. Whether the output of the real machine and the simulator are exactly the same, given the same model and the same inputs.