search

summerlvsong / Aggregation-Cross-Entropy

Aggregation Cross-Entropy for Sequence Recognition. CVPR 2019.
303 stars 60 forks source link

Use ACE loss for English handwriting recognition，but not converge #14

Open liudatutu opened 5 years ago

liudatutu commented 5 years ago

I use the ACE loss function to do English handwriting recognition. When the model is trained, it does not converge, but with CTC, it gets a good convergence effect. How can this happen?

yflv-yanxia commented 4 years ago

setting a larger lr might work

bjlgcxc commented 4 years ago

I also find this problem in the training of crnn with synth90k dataset