deep-text-recognition-benchmark
deep-text-recognition-benchmark copied to clipboard
clovaai
Text recognition (optical character recognition) with deep learning methods, ICCV 2019
> 0, 'datasets should not be an empty iterable' # type: ignore[arg-type] AssertionError: datasets should not be an empty iterable My config.py : """ Default CONFIGURATIONS """ exp_name = 'logs'...
Can you please tell me , how did you overcome overfitting _Originally posted by @Amithvassit in https://github.com/clovaai/deep-text-recognition-benchmark/issues/154#issuecomment-644687741_ e..., I've tried using weight_decay, and I' ve found that it's good for...
Hello, I've started training the model on **custom dataset** and found that after 10K iterations, **Training loss** has **decreased to 0.00002** but **validation loss** is still on **higher side** and...
Hello, I am trying to add additional characters (--sensitive option) by fine-tuning existing non-sensitive model I am using None-VGG-BiLSTM-CTC (a.k.a CRNN) model, which had accuracy of `62.795%` on IC15_2077 test...
Shape mismatch is happening if I try to add extra 2 characters. Any solution for the same? If i want to do fine tuning on TPS-ResNet-BiLSTM-Attn.pth `RuntimeError: Error(s) in loading...
Paper said "This result showed that the diversity of training data can be more important than the number of training examples, and that the effects of using different training datasets...
The EasyOCR ReadME states that the following three files are required to use a custom model: custom_model.pth, custom_model.yaml, and custom_model.py. How are we supposed to create the custom_model.yaml and custom_model.py...
Below is the configuration file and training log. Looking forward to your reply.   @ku21fan
I've train with long words. But cannot predict long numeric words or email address like below:    Any help will appreciate me.
Hello, We got very low accuracy on long string recognition. The strings are 20-character long, and consist of mixed capital and numeric numbers. The recognition error include incorrect # of...
