Tom Dörr

It's currently training and get's updatet from time to time. Not sure if I can make it less resource hungry. What would you need for game design?

There were changes in the backend today, the server was scaled up and there was a huge amount of traffic. I hope it sorts itself out.

I don't plan to do that, but maybe someone else is wiling to implement this

The idea is that the model replaces words with other words that could be in the same place. You might be able to improve it to use a different model...

Don't know how to do that, but I'll leave this issue open in case someone else wants to implement this

What is the difference between AI and human text?

Could you try it again? It should run a bit faster

This is the data that is being sent (I removed the tokens): ``` data: {'auth_token': ''} data: {'roadall': [[1576429200, 0, None]], 'auth_token': ''} ``` The Unix timestamp `1576429200` is neither...

Yeah not sure what's going on there. Sometimes it helps to to python3 -m pip install ...., but I think this might not be one of those cases.

@santoryudeadman I switched to lm-format-enforcer: https://github.com/noamgat/lm-format-enforcer