Yue Gao

PyTorch and PyTorchLightning

This issue does not happen in previous version of Dash, when I creating the PyTorch docset.

Both. I did not test it on current version of Dash with a previous version of the docset.

Any news? Could this be fixed in the next version of Dash?

Thanks for your interest. The number 52 represents 26 upper and lower case letters. In Generator, the `charclass` is used for the content supervision, i.e., which character is for the...

Yes, the 52 need to be changed according to the number of glyphs in practice. We did not pass the `charclass` in generator, did you mean the discriminator? The charclass...

The demonstration app is implemented with [Flask(backend)](https://flask.palletsprojects.com/en/1.1.x/) and [Material Design Bootstrap(frontend)](https://mdbootstrap.com), the logic is very simple, you can refer to the official tutorials [quickstart](https://flask.palletsprojects.com/en/1.1.x/quickstart/).

I have the same problem with large batch_size 64, have you guys found the solution? Help, please.

Thanks @linusericsson for your great comments. Indeed there are lots of parameters, the log will help to reproduce the results provided in paper. The results of BYOL in our paper...

To finetune the models, please copy the converted models to detectron2 path, and modify the lines related pretrained or resume path of the official config files. Then follow the instructions...