Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
tongxingyu 9f42b02f6e | 2 years ago | |
---|---|---|
.. | ||
configs/posec3d | 2 years ago | |
data | 2 years ago | |
docs | 2 years ago | |
output/c3d | 2 years ago | |
README.md | 2 years ago | |
common_func.sh | 2 years ago | |
prepare.sh | 2 years ago | |
test_train_inference_python.sh | 2 years ago |
飞桨除了基本的模型训练和预测,还提供了支持多端多平台的高性能推理部署工具。本文档提供了PaddleOCR中所有模型的飞桨训推一体认证 (Training and Inference Pipeline Certification(TIPC)) 信息和测试工具,方便用户查阅每种模型的训练推理部署打通情况,并可以进行一键测试。
test_tipc/
├── configs/ # 配置文件目录
├── posec3d # posec3d模型的测试配置文件目录
├── train_infer_python.txt # 测试Linux上python训练预测(基础训练预测)的配置文件
├── train_infer_python.md # 测试Linux上python训练预测(基础训练预测)的使用文档
├── results/ # 预测结果
├── prepare.sh # 完成test_*.sh运行所需要的数据和模型下载
├── test_train_inference_python.sh # 测试python训练预测的主程序
└── readme.md # 使用文档
使用本工具,可以测试不同功能的支持情况,以及预测结果是否对齐,测试流程概括如下:
test_train_inference_python.sh
,产出log,由log可以看到不同配置是否运行成功;测试单项功能仅需两行命令,如需测试不同模型/功能,替换配置文件即可,命令格式如下:
# 功能:准备数据
# 格式:bash + 运行脚本 + 参数1: 配置文件选择 + 参数2: 模式选择
bash test_tipc/prepare.sh configs/[model_name]/[params_file_name] [Mode]
# 功能:运行测试
# 格式:bash + 运行脚本 + 参数1: 配置文件选择 + 参数2: 模式选择
bash test_tipc/test_train_inference_python.sh configs/[model_name]/[params_file_name] [Mode]
以下为示例:
# 功能:准备数据
# 格式:bash + 运行脚本 + 参数1: 配置文件选择 + 参数2: 模式选择
bash test_tipc/prepare.sh ./test_tipc/configs/posec3d/train_infer_python.txt 'lite_train_lite_infer'
# 功能:运行测试
# 格式:bash + 运行脚本 + 参数1: 配置文件选择 + 参数2: 模式选择
bash test_tipc/test_train_inference_python.sh ./test_tipc/configs/posec3d/train_infer_python.txt 'lite_train_lite_infer'
Revisiting Skeleton-based Action Recognition(PoseC3D 基于Paddle复现)
Python Shell Markdown Text
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》