-
你好，用作者给的train data 训练ocr，还是用自己做的数据集训练ocr, loss 一直在降，但是predict的结果一直为空，准确率为0，请问是为什么呢？感谢！
-
训练速度要低很多
-
Error while make in cmake fashion. Complains
> cuda_compile_generated_ctc_loss_layer.cu.o' failed
cmake Log
```
-- The C compiler identification is GNU 5.4.0
-- The CXX compiler identificati…
-
I have some variable saved from crnn-pytorch, which runs on gpu and uses warp-ctc. Then I use those variables in my code, I can run it on cpu and the results are correct. But when I run it on gpu, I g…
-
I am following the pytorch implementation. I couldnt find any support for beam search decoding in either of the implementations. Does the decoding by default use beam search?
-
我生成了5w的样本，4w做训练，1w做测试，accuracy为0，可能的原因是什么呢，还有就是你的accuracy和ctc loss分别是什么呢
-
哈喽，向你请教一个问题。如果我想在返回生成图片的同时保存每个字符在当前图片中的位置坐标，请问应该修改哪部分的代码呢？
-
Hello, I'm using a aws instance with 4 gpus and when activated (in the params.py file - True multigpu and 4 for the number) I get the following error: (P.S: For 4, 3, 2 and even 1 which is incomprehen…
-
### Bug description
when I run the following code, the training time of the epoch will increase epoch by epoch. For example, the first epoch takes 3:39 min, and the second on takes 4:21min, and the t…
-
Hi @yalecyu ,
Thanks for sharing your code! I have trained the number of 0~9, and the test result is very effective.
But I can not train the 36 Alpha by "0~9 + A~Z"!
The maximum length of my charac…