Mert Atagül

According to [documentation](https://keras-ocr.readthedocs.io/en/latest/api.html) top stands for _Whether to include the final classification layer in the model (set to False to use a custom alphabet)._

I used pytorch's function torch.onnx.export which you can find details [here](https://pytorch.org/tutorials/advanced/super_resolution_with_onnxruntime.html). To be more specific I have modified the demo.py to create the model and load my weights into it....

I doubt that data paralleling effects inference results. It would be great if someone who achieved training and transporting that model to openCV could help.

Unfortunately no I didn't and currently working on something else. But I'm still keen to learn if this is possible.