search

qjadud1994 / CRNN-Keras

CRNN (CNN+RNN) for OCR using Keras / License Plate Recognition
MIT License
527 stars 191 forks source link

Size and nature of the dataset. #21

Open Sreerag-ibtl opened 5 years ago

Sreerag-ibtl commented 5 years ago

I have a 25451 images in the train and 2000 images in the test set. The problem I am facing is the train and validation loss is high in every epochs. In my dataset a few images are of plates with two lines. So I would like to get a suggestion about the dataset size and it's nature. TIA

ghost commented 5 years ago

Hi @Sreerag-ibtl, In my opinion, the number of dataset is small. because i tested this problem, i've solved this problem with number of large dataset. around 500k images.

Sreerag-ibtl commented 5 years ago

Thanks for the reply. I have a fair amount of synthetic images. I tried a different implementation of the CRNN a while ago, with those. But the prediction was not better.