Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices)
请问PP-OCRv3和PP-OCRv4检测模型是用了哪些数据集训练的?hmean指标是在哪个数据集上评估的? 我在介绍文档中找不到相关的信息:https://github.com/PaddlePaddle/PaddleOCR/blob/main/doc/doc_ch/PP-OCRv4_introduction.md