WangZ
WangZ
可能文档中链接没有更新过来，我反馈一下
[2024/05/16 11:00:15] ppocr ERROR: error in loading image:abc/four-part.1.3.jpg Traceback (most recent call last): File "paddle_and_annotate.py", line 25, in result = ocr_model.ocr(img_path) File "/usr/local/lib/python3.8/site-packages/paddleocr/paddleocr.py", line 660, in ocr img = preprocess_image(img)...
This is a C++ error, seems to be a warning about file reading, more context is needed to determine the specific problem.
When you do not specify the default detection and recognition model address, paddleocr will download the official model by default. You can delete it manually, or it will not be...
目前已关注到问题，已经反馈给负责框架的同学排查
蒸馏模型目前还存在问题，已经反馈问题，等待推进
可能是数据中存在无法解析的字符，可能是\n \t等字符
你好，可以提供一个排序错误的box坐标吗
感谢讨论❤️
这个是json保存的问题吧，尝试加入ensure_ascii=False试试 json.dumps(data,ensure_ascii=False)