summerlvsong / Aggregation-Cross-Entropy

Aggregation Cross-Entropy for Sequence Recognition. CVPR 2019.

summerlvsong/Aggregation-Cross-Entropy Issues

loss decline but accuracy near to 0
Updated 2 years ago5
Need Help! Loss nan
Closed 5 years ago3
是否可以用在中文OCR上面呢
Updated 3 years ago2
Can’t reproduce your results in your cvpr paper
Updated 3 years ago3
It doesn't work on my data, why don't you provide a pretrained model?
Updated 3 years ago5
Prediction script
Closed 4 years ago2
HWDB results
Updated 4 years ago
Why use ResLSTM for Offline Handwritten Chinese Text Recognition ?
Closed 5 years ago1
Use ACE loss for English handwriting recognition，but not converge
Updated 4 years ago2
combined with CTC?
Updated 4 years ago1
关于论文公式（2）的理解问题
Updated 4 years ago
def vis(self,iteration) function, possible error
Updated 4 years ago
What is the accuracy criterion in 2D prediction?
Updated 4 years ago
Would MSE loss work if i remove the softmax?
Updated 4 years ago
I would like to ask you how to accurately predict the character order of a word.
Updated 5 years ago6
Compared with CTC, the effect is very poor in Chinese long text
Updated 5 years ago
Could you share your trained models?
Updated 5 years ago
KL Divergence
Updated 5 years ago
Do ACE work? same point about HCTR Result
Updated 5 years ago
how to get val label
Updated 5 years ago1
Would it be helpful in Speech Recognition especially in acoustic model to replace CTC with ACE?
Updated 5 years ago1
can you share 288 cropped natural images for testing in CUTE80
Closed 5 years ago1
code about HCTR will be publicly or not?
Closed 5 years ago1
CUDA_VISIBLE_DEVICES=0 python -u main.py \ 2>&1 | tee $filename
Closed 5 years ago2
could you upload your trained model for scene text recognition?
Updated 5 years ago1
Can you provide the data generator tool ?
Closed 5 years ago4
When will the code be publicly available?
Closed 5 years ago