Tao Luo

@lixiangpengcs have you every tried your own dataset?

@lixiangpengcs have you replace my crnn to original crnn?

@lixiangpengcs how about your learning rate set?

@lixiangpengcs if you try the public dataset such as icdar2015, you need set multi step lrs,and the initial learning rate to a little bit big,it can avoid local optimum,the value...

@lixiangpengcs you can parallel the network with pytorch multigpu tutorial. i don't adapt this.

@lixiangpengcs can u post your error log ?

@lixiangpengcs i think you can try this method: concat all the label of one image into single string,and split it during the train process.

@lixiangpengcs except your good news.

@hwlll 具体报错信息

@ivoxx 谢了~ 另外群的话，可以加qq群：785515057（已满）826714963。也可以加微信nsnovio，拉你进微信群