changdazhou

Results 132 comments of changdazhou

基于python部署训练环境，训练的文字识别模型在C++源码部署加入该模型文件后不适用，报异常关于Paddle_inference的问题

现在报错的是GPU版本的还是CPU版本呢，如果是CPU版本，是因为CPU版本不稳定导致的

Compling cpp libpaddle_inference.so on Ubuntu 20 failed

you can try remove Paddle/third_party/* and remove Paddle/.git/modules/third_party/*

Compling cpp libpaddle_inference.so on Ubuntu 20 failed

sorry, you need to raise this issue under the warehouse of Paddle

Compling cpp libpaddle_inference.so on Ubuntu 20 failed

ok, this issue has been recorded

Paddleocr pp_chatocrv4_doc results in error "Large model invocation failed"

This issue may be caused by an invalid or expired Access Token. We recommend you check the validity of the Access Token. If necessary, you can try regenerating or updating...

Paddleocr pp_chatocrv4_doc results in error "Large model invocation failed"

you need set LLM server url and api_key in ".\paddlex_hps_PP-ChatOCRv4-doc_sdk\server\pipeline_config.yaml"，you can refer to https://github.com/PaddlePaddle/PaddleX/blob/release/3.3/paddlex/configs/pipelines/PP-ChatOCRv4-doc.yaml#L9

PPStructureV3处理多页PDF报错

感谢您的反馈，我们将优先排查此问题。请稍候，我们会尽快提供更新。

Crashing Google Colab

try pip install paddlepaddle==3.0.0

在PP-Structure使用文本分析时会输出res.txt,但其中输出得是Unicode字符，如何修改呢

对该文件解析是使用Unicode格式加载即可，如果需要保存到其他格式，建议阅读一下相关部分源码

ai studio上paddleocrvl api并发调用 返回429，后来在调用返回403

是稳定复现的吗，目前ai studio调用量较大，如果是偶现问题，可能和服务端排队有关