关于GCU、沐曦GPGPU、MLU、0卡V100资源4月7日恢复上架的公告>>> 关于共建具身智能开源数据集的倡议>>> 关于云脑任务中统一路径访问方式的公告>>> 关于将启智集群GPU资源迁移至智算集群的公告>>>
PaddlePaddle
/
PaddleSeg
mirror of https://github.com/PaddlePaddle/PaddleSeg
Not watched
5
2
Fork 14
Code
Releases 20 Wiki Activity
Issues 0 Datasets Model Cloudbrain
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
1522 Commits
25 Branches
447 MiB
3071 Download
 
 
 
 
 
 
Tree: v2.4
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'v2.4'
${ noResults }
PaddleSeg/benchmark
wuyefeilin 1c1ec3a27f
[Doc] Add English Docs (#1702) 		2 years ago
..
configs [Benchmark] Revert useless setting changes (#1690) 2 years ago
README.md [Doc] Add English Docs (#1702) 2 years ago
README_CN.md [Doc] Add English Docs (#1702) 2 years ago
deeplabv3p.yml [Enhancement] Support dy2stat for HRNet and DeeplabV3 (#1527) 2 years ago
hrnet.yml [Enhancement] Support dy2stat for HRNet and DeeplabV3 (#1527) 2 years ago
hrnet48.yml Merge benchmark 2 years ago
run_all.sh [Benchmark] Change some settings for seg train performance (#1683) 2 years ago
run_benchmark.sh update seg benchmark case (#1506) 2 years ago
run_fp16.sh Merge release/2.2 into develop (#1326) 2 years ago
run_fp32.sh Merge release/2.2 into develop (#1326) 2 years ago

README.md

English | 简体中文

Benchmark Introduction

The content is as follow:

├── README.md  
├── configs  
│   ├── cityscapes_30imgs.yml  
│   ├── fastscnn.yml  
│   ├── ocrnet_hrnetw48.yml  
│   └── segformer_b0.yml  
├── deeplabv3p.yml  
├── hrnet.yml  
├── hrnet48.yml  
├── run_all.sh  
├── run_benchmark.sh  
├── run_fp16.sh  
└── run_fp32.sh

Environment

Use Docker to configure the environment.

  • docker image: paddlepaddle/paddle:latest-gpu-cuda10.1-cudnn7
  • CUDA 10.1 + cudnn7
  • paddle=2.1.2
  • py=37

Test

Training Test

git clone https://github.com/PaddlePaddle/PaddleSeg.git
cd PaddleSeg
bash benchmark/run_all.sh

How to Open Profiling

Add the following parameter when training.
--profiler_options="batch_range=[50, 60]; profile_path=model.profile

Delete task

Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.

cancel
confirm

飞桨高性能图像分割开发套件，端到端完成从训练到部署的全流程图像分割应用。

https://github.com/PaddlePaddle/PaddleSeg

Python Markdown Text Shell Java other

Contributors (25+)
All

Please read the following content carefully:

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

Agree and continue
Disagree, exit