Life In 19x19

Thanks so much for the replies, will look into it!

I know this is a longshot and offtopic, but akigo, since you are such a tensorflow wizard and a charitable person I thought I would try :bow: . Would you know if there is any way force GPU support of the sorting of abstracts in this hobby project I am using to learn python:

https://github.com ...

if you want the strongest net, the 60b net in this link should work fine for you
https://media.katagotraining.org/upload ... 017.bin.gz

I was getting this same feeling using the snapdragon optimized network against golaxy. it definitely felt at least as strong as the s509 network. What an achievement by all of you to create such a strong, small and fast network. You have made katago play at a professional level even on cheap mobile ...

Thanks And for the attempt, I really appreciate the effort! Of course the optimized version may be a little weaker than the same net, but i am very curious how it would compare to the older non-distributed training katago. You gave me an idea, maybe i will try two games at 1 playout. My guess is the ...

I know its a lot to ask, but would it be possible to run 5-10 matches of the latest snapdragon optimized network vs the last non distributed training 40b release at 300 playouts each? It is so impressive what you have done. Probably the slight difference is just from the parrallel processing?

I published a new accelerated network for Snapdragon 888 devices on android 12. For details and download go here: https://github.com/aki65/aki65.github.io/releases/tag/v1.13.2
I have a snapdragon 636 e ink device (onyx boox note air 1) and the speed of this new optimized net is incredible! Is it ...

Gansheng Shi had a movie made about him, was one of the first American pros, and now is officially a medical doctor! is there anything he can't do? What an great inspiration to the go playing youth.
https://ibb.co/dWSwvVq
he added the subscript, "coming home, baby!“

My mistake, I miswrote, I meant how fast he released the net. In terms of playouts, its probably faster than the non-optimized 20b net. On Snapdragon 625, 4gb ram, I get maybe 2 playouts/second, somewhere between 1-2. Which is really nice for these devices

Wow i cant believe how fast you make the optimized network. Thatis really amazing.

If you don't want to train katago, but just play rating games, would this work?
https://github.com/portkata/KataGo/blob ... ames.ipynb
probably have to replace cuda with auto.

Is it possible to do an optimized version of s680? Thanks either way!

Yes, sorry about that. In the picture below, paste the following code where the bottom most > sign is. Ill delete the previous redundant post.
https://i.ibb.co/WWwqG0H/Screenshot-20210227-034036.png
geojson lat long
function ClickConnect(){
console.log("Clicked on connect button");
document ...

edit: Is it possible to make a script like this to train the 15b net on the new s663 40b data? thanks!

Wow, thanks for this And. Does it use GPU by any chance. Either way, really neat program.