ppv3_rec模型转换不成功

[AstonyJ]

1、转换的pp模型链接： https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_train.tar 2、运行脚本： python ./converter/ch_ppocr_v3_rec_converter.py 3、报错：

Traceback (most recent call last): File "./converter/ch_ppocr_v3_rec_converter.py", line 60, in load_paddle_weights self.net.state_dict()[k].copy_(torch.Tensor(para_state_dict[ppname])) KeyError: 'Student.backbone.conv1._conv.weight'

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "./converter/ch_ppocr_v3_rec_converter.py", line 94, in converter = PPOCRv3RecConverter(cfg, paddle_pretrained_model_path) File "./converter/ch_ppocr_v3_rec_converter.py", line 22, in init self.load_paddle_weights([para_state_dict, opti_state_dict]) File "./converter/ch_ppocr_v3_rec_converter.py", line 64, in load_paddle_weights print('paddle: {}, {}'.format(ppname, para_state_dict[ppname].shape)) KeyError: 'Student.backbone.conv1._conv.weight'

[frotms]

python ./converter/ch_ppocr_v3_rec_converter.py --src_model_path paddle_ch_PP-OCRv3_rec_train_dir

[AstonyJ]

python ./converter/ch_ppocr_v3_rec_converter.py --src_model_path paddle_ch_PP-OCRv3_rec_train_dir

我写到py文件里了：

[frotms]

python ./converter/multilingual_ppocr_v3_rec_converter.py --src_model_path paddle_multilingual_PP-OCRv3_rec_train_dir

[AstonyJ]

转换成功了 感谢

[AstonyJ]

https://github.com/PaddlePaddle/PaddleOCR/blob/release%2F2.5/doc/doc_ch/algorithm_rec_svtr.md 请问该网页下的模型可以转换吗？

[frotms]

暂时还没整理好，这周末看下

[AstonyJ]

暂时还没整理好，这周末看下 嗯嗯 感谢呀 我也学习一下 看看能不能转成功

[frotms]

@AstonyJ svtr已更新。但请注意paddle的nn.Dropout(p=last_drop, mode="downscale_in_infer")的"downscale_in_infer"参数

[AstonyJ]

收到 已成功转换 感谢！

[BLMFY]

所以楼主这个问题是怎么解决的呢，为什么会出现这个keyerror呢？

[yangy996]

转换成功了 感谢

楼主这个问题是怎么解决的呢，我现在转换也报这个错