David J Wu

Ah, the answer is probably that CMake did not get cleaned when you run `make clean`. This is because `make` is on a lower-level layer than CMake - CMake creates...

It depends entirely on what you are doing. If you just want to entirely write your own Go bot, then you don't need this repo, you want to just implement...

No, it's not worth doing this. You should implement your own bot. Take a look at the jtauber repo there - it already provides all the starting point you need...

As far as I know, the options aren't great for that because the existing options are *just* passable enough that nobody has bothered making better ones, and the KataGo repo...

I haven't implemented something like that myself because as I said earlier, the other options do work well enough that it's not worth me spending all the time to implement...

Apologies, I haven't had the bandwidth to install CUDA 12.3 or 12.4 myself and test things out, so I put off responding to this. Do you still think this is...

Are you familiar with AlphaZero, or Expert Iteration, or similar methods? The original papers are pretty good background reading if you're not: https://arxiv.org/pdf/1705.08439.pdf https://discovery.ucl.ac.uk/id/eprint/10045895/1/agz_unformatted_nature.pdf The main thing about these methods...

That sounds strange. Do you have a case that reproduces this with reasonable reliability so that I can use it to debug, or is it nondeterministic?

Thanks, this is fixed in master now. It should be `!=` instead of `==`.

That's pretty weird, because I'm pretty sure KataGo doesn't use double-precision floats in the neural net. I looked over that kernel and didn't see anywhere where it's used, only single...