Aurora11111

I meet this problem ,and I sloved it just by uninstall old gcc,and reinstall cuda/cudnn/gcc/cmake

I have tried downgrade gcc, and it can't help. you can reference Issue when install warp_ctc_pytorch :https://github.com/Aurora11111/crnn-train-pytorch

@kevgeo which means reinstall cuda,cudnn,gcc,cmake

what about both loss and accuracy are 0 ?

@gachiemchiep thank you .It did my datasets's problem!

@gachiemchiep I have successfuly trained a crnn model to recognize numbers(0-9),but when I use this method into chinese char,the issue occurs(in the upload trianing-log.txt): [training-log.txt](https://github.com/meijieru/crnn.pytorch/files/2373382/training-log.txt) there still low loss,but 0...

@gachiemchiep maybe,I'm using less datasets trianing now(10 chars every image),the loss is from 170 to decrese to 42 and still decresing now, but it's decresing speed is slow (decrese about...

@gachiemchiep I haven't fixed the unicode issue,becase I haven't find out the solution. I have print the gt ,there is no mistake,I don't know why the length of val result...

@gachiemchiep It indeed my unicode issue.the difference between python2 and python3.

I think it beacause of your trianing image size and it must match with the training model