PaddleOCR/latest/version3.x/module_usage/formula_recognition

[giscus[bot]]

PaddleOCR/latest/version3.x/module_usage/formula_recognition

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)

https://paddlepaddle.github.io/PaddleOCR/latest/version3.x/module_usage/formula_recognition.html

[lifloveyou]

python3 tools/train.py -c configs/rec/PP-FormuaNet/PP-FormulaNet_plus-M.yaml -o Global.pretrained_model=./rec_ppformulanet_plus_m_train/best_accuracy.pdparams
执行公式识别训练命令时报错：
File "/paddleocr/PaddleOCR-main/ppocr/modeling/heads/rec_unimernet_head.py", line 619, in forward
attn_weights.reshape([bsz, self.num_heads, tgt_len, src_len])
TypeError: unsupported operand type(s) for +: 'Tensor' and 'Tensor'

[liuhongen1234567]

您好，可以提供下paddle的版本等复现环境吗？理论上paddle3.0以上都是没问题的。