inference模型转nb模型失败

[renshujiajia]

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem

• 系统环境/System Environment：Linux ocrdevelop 5.11.0-27-generic 导出识别模型时都是 KeyError: 'predict'  #29~20.04.1-Ubuntu SMP Wed Aug 11 15:58:17 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux
• 版本号/Version：Paddle： PaddlePaddle-gpu 2.6.0 PaddleOCR2.7.0： 问题相关组件/Related components：ppocrv3训练及导出
• 运行指令/Command Code：paddle_lite_opt --model_file=./ch_PP-OCRv3_rec_slim_infer/inference.pdmodel --param_file=./ch_PP-OCRv3_rec_slim_infer/inference.pdiparams --optimize_out=./ch_PP-OCRv3_rec_slim_opt --valid_targets=arm --optimize_out_type=naive_buffer
• 完整报错/Complete Error Message：Loading topology data from inference.pdmodel
  Loading params data from inference.pdiparams

1. Model is successfully loaded!
   [F 1/16 10:59:55.518 ...e-Lite/lite/core/model/general/op_desc.h:118 GetAttr] Check failed: it != attrs().end(): No attributes called beta found for swish
   Aborted (core dumped)

我们提供了AceIssueSolver来帮助你解答问题，你是否想要它来解答(请填写yes/no)?/We provide AceIssueSolver to solve issues, do you want it? (Please write yes/no):yes

请尽量不要包含图片在问题中/Please try to not include the image in the issue.

[renshujiajia]

官方提供的en_PP-OCRv3_rec_train.tar和en_PP-OCRv3_rec_slim_train.tar直接转inference模型同样成功，但是转nb或者onnx存在相同问题

[mohblnk]

How was this solved? Why is this closed if not solved?