fengxinjie

There are label nosie in IC15 test set, and I have relabeled.

> > Hi @zobeirraisi > > I am also interested in this work. It'd be greatly appreciated if you can post the results on datasets that you have tried. >...

> > Hi @zobeirraisi > > I am also interested in this work. It'd be greatly appreciated if you can post the results on datasets that you have tried. >...

> SVTP has 645 images, if you carefully read each one, you will find there are around 20 images humans cannot recognize. How can this algorithm get 98.6%?? > >...

> There are several papers that adopt transformer on OCR. > > 1. NRTR: A no-recurrence sequence-to-sequence model for scene text recognition > 2. MASTER: Multi-Aspect Non-local Network for Scene...

我没遇到过这个问题，我用的是cuda9

> @fengxinjie 谢谢回复。转成cuda9是ok的。原因是我的数据中含有< >"这样的字典，在转化过程标签对应的索引有问题，请问如何把这些符号也加入到vocab中啊。 我是为了方便打标签才用做边界符的，你可以直接指定边界符的id就可以了，不用符号。

We use ResNet101 pretrained model of pyTorch model zoo, and transformer is trained from scratch.