PaddleServing Inference

[ShahZebYousafzai]

When running

python recognition_web_service.py -n PPTSM -c configs/PP-TSM.yaml

Nothing happens and the code just exits.

Then I changed

uci_service = VideoService(name="video")
uci_service.get_pipeline_response(read_op=args.config)
uci_service.run_service()

to

web = WebService()
web.get_pipeline_response(read_op=args.config)
web.run_service()

After some changes I get the following error

Traceback (most recent call last): File "recognition_web_service.py", line 206, in <module> web.run_service() File "C:\Users\UMAIR COMPUTER\anaconda3\envs\paddle_env\lib\site-packages\paddle_serving_server\web_service.py", line 69, in run_service self._server.run_server() File "C:\Users\UMAIR COMPUTER\anaconda3\envs\paddle_env\lib\site-packages\paddle_serving_server\pipeline\pipeline_server.py", line 292, in run_server if self._build_dag_each_worker: AttributeError: 'PipelineServer' object has no attribute '_build_dag_each_worker'

[westfish]

你好，根据你提供的错误信息，问题出在pipeline_server.py文件的第292行，这里的self._build_dag_each_worker在PipelineServer对象中并不存在。这可能是因为你的Paddle Serving版本和你的代码不兼容。

另外，如果有图像、视频理解和生成的需求，可以使用我们新的跨模态工具: https://github.com/PaddlePaddle/PaddleMIX/tree/develop