Serving icon indicating copy to clipboard operation
Serving copied to clipboard

A flexible, high-performance carrier for machine learning models(『飞桨』服务化部署框架)

Results 218 Serving issues
Sort by recently updated
recently updated
newest added

从0.6.2升级到0.7.0,原来的正常的服务报这样的错,是什么原因呢?

question

Great to see you here

用paddle inference 不会报错, 用pipline serving会报错 CRITICAL 2022-02-09 10:46:11,025 [operator.py:1315] [ner|1] failed to init op: (External) CUDA error(3), initialization error. [Hint: 'cudaErrorInitializationError'. The API call failed because the CUDA driver and...

question

win10环境 docker 安装 paddle serving,是否能使用T4 Cuda11 GPU 驱动?

multi-OS platforms

[问题说明.docx](https://github.com/PaddlePaddle/Serving/files/8460597/default.docx)

硬件适配

python37 -m paddle_serving_server.serve --model uci_housing_model --thread 10 --port 9292 --name uci /usr/local/python3/lib/python3.7/runpy.py:125: RuntimeWarning: 'paddle_serving_server.serve' found in sys.modules after import of package 'paddle_serving_server', but p rior to execution of 'paddle_serving_server.serve'; this...

help wanted

/site-packages/paddle_serving_server/serve.py 中的binurl使用的是https://paddle-serving.bj.bcebos.com/bin/serving-cpu-avx-mkl-0.8.3.tar.gz,这个地址已经无法访问,现在只能使用https://paddle-serving.bj.bcebos.com/test-dev/bin/serving-cpu-avx-mkl-0.8.3.tar.gz 请及时更新paddle_serving_server,不然第一次启动paddleserving的时候会出现无法下载bin的问题

enhancement

使用以下命令部署模型是,输入的thread参数不起作用,cpu使用率不超过400%: > python -m paddle_serving_server.serve --model my_model --thread 10 --port 9292 --name my_model 原因是文件: https://github.com/PaddlePaddle/Serving/blob/develop/python/paddle_serving_server/serve.py 中第433行中processes=4,并没有更新为用户的输入参数;

good first issue

ImportError: /usr/lib/x86_64-linux-gnu/libcrypto.so.1.1: version `OPENSSL_1.0.1_EC' not found (required by /usr/local/lib/python3.7/dist-packages/paddle_serving_client/serving_client.so) 机器上已经安装openssl 1.1.1, ubuntu 18.04

install

使用paddleServing 0.8.2 C++ 部署了一个自训练的PPyolo_mb_Small模型。 使用/Serving/examples/C++/PaddleDetection/ppyolo_r50vd_dcn_1x_coco/test_client.py 修改resize尺寸后,可以正常使用 ![image](https://user-images.githubusercontent.com/15975613/159264817-4c1ea9cf-5270-4c80-82ce-7a7dfb808907.png) ![image](https://user-images.githubusercontent.com/15975613/159265091-4e7eac9d-aec4-42cd-8370-abe41b2970e1.png) 但是使用java client sdk时,遇到一些问题。由于javaSDK只提供了yolov4的模型样例,我尝试修改了feedData。 代码如下 ![image](https://user-images.githubusercontent.com/15975613/159265568-ab14774f-e727-4960-9c5f-0f476aad96d0.png) serving_client_conf.prototxt内容 ![image](https://user-images.githubusercontent.com/15975613/159265647-39769abb-7ed0-45b9-a6f7-8568202f44f0.png)

多语言client