search

leela-zero / leela-zero

Go engine with no human-provided knowledge, modeled after the AlphaGo Zero paper.
GNU General Public License v3.0
5.31k stars 1.01k forks source link

Leela on edge tpu? #2270

Open merlinpan opened 5 years ago

merlinpan commented 5 years ago

I suspect with one of these things https://coral.withgoogle.com/ it may actually run at a respectful speed even with a raspberry pi.

Unfortunately, linux only for now.

iopq commented 5 years ago

https://github.com/tensorflow/minigo/pull/811

l1t1 commented 5 years ago

https://github.com/tensorflow/minigo/issues/824

ayaz-amin commented 4 years ago

You could try training it on cloud TPUs (as in the exact ones that Deepmind used to train AlphaZero and its predecessors) as they are now readily available on Google Cloud (if you have some cash on you ofc) or on Google Colab for free (though there is a certain time limit before the session stops, which is quite annoying). The edge TPUs that you were pointing out to are more suited towards inference on low-power devices.

kennyfs commented 4 years ago

LZ is based on opencl, you have to rewrite a program to take advantage of tpus.

ayaz-amin commented 4 years ago

Ah, that's quite unfortunate. I mean, it wouldn't hurt to try rewriting it as a hobby project, but it would greatly increase the performance of leela-zero considering that you have both the implementation (which could be modified for performance) as well as the hardware. I hear that Google released their next-gen TPUs on the cloud, so in theory leela-zero may be able to surpass AlphaZero in performance.