Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There is actually an IOCCC entry in 2019 that does almost exactly this except it’s LSTM instead of transformers: https://www.ioccc.org/2019/mills/prog.c

https://www.ioccc.org/2019/mills/hint.html



This one is actually impressive: implements Adam training of RNNs, LSTMs, GRUs.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: