brianprichardson

Update on work in progress

3

I am currently working on larger NNs (256x20). Still using supervised PGN file input. Have disabled "testeval". Trying 32 policy and value head filters per Leela, and here: https://medium.com/oracledevs/lessons-from-alpha-zero-part-6-hyperparameter-tuning-b1cfcbe4ca9a Last...

Working for me

19

Progress is slow, but it does seem to be working! Eval games take about 6 min. (set gpu mem to 25%) Opt epochs take 29 sec. (gpu mem 30%) Self...

Potential collaboration

3

See here: [https://github.com/crypt3lx2k/Zerofish](url)

Next step

11

As you now have a powerful M60 GPU (about 2-3x my GTX 1070), I am wondering what would be the most helpful next step so that I can still contribute.

Ubuntu 16.04 LTS venv eraseme tensorflow-gpu all 3 workers running run self --type mini encounters the error looks like shortly after 2017-12-09 01:26:06,376 able to restart run self and it...

Huge number of files created

3

It ran for a couple of days and found several new best models. However, it also creates numerous files (502,586 items, totalling 5.6 GB). The models directory is large and...

Fancy branch changing side to move

1

From startpos, after first "go nodes" command if a second is entered the bestmove is for black. Per UCI protocol I think side to move should stay white.