AobaZero is a user-participated Shogi AI project that will test the AlphaZero Shogi experiment.

If you are interested, please join us. Anyone can contribute using Google Colab.

GitHub Source, executable files. GitHub(Japanese top page)

2021-01-11 Replay buffer is past 1000000 games from past 500000 games. 28352k games, w3077.
2020-12-28 weight is updated from each 10000 games to 34285 games. 26700k games, w3022.
2020-12-11 v1.9 Sente one ply mate returns loss(-1) not win(+1) is fixed.
2020-12-06 weight_decay is changed from 0.0002 to 0.00004. 23982k games, w2750. We reache 24000k games, but we'll continue a little more with some another hyper parameters.
update history

2021-01-18 06:33 JST(update every 30 minutes)
In past hour,number of clients are 3, 3626 games.
In past 24 hours, number of clients are 12, 130933 games.
Total 29065562 games. Latest weight= w3099. Next is in 1.2 hours. Thank you for your contribution!
In past 1000 games, Average of moves 87.6, Sente winrate 0.598, Draw rate 0.079
In past 1,000,000 games, Average of moves 89.5, Sente winrate 0.553, Draw rate 0.142

Elo progress. It is based on a self-match with the previous weight. Left vertical axis is Elo. Right is floodgate and vs Kristallweizen 1k,10k,50k,100k,200k,500k. Horizontal axis is number of trained games(1 unit is 10000 games).
As of 2021-01-17.

AobaZero 800playouts/move vs Kristallweizen 500k/move. 800 match games.

You can see the process of acquiring Shogi knowledge from the game records.
Self-play games without noise. Each game uses same weight.

You can see the transition of opening moves.

Self-play games for training.
Self-play games for every 10,000 games added. The top of the page is the latest game. It will be updated every other day.

These are self-play games for training. It often plays blunder for the first 30 moves.
And sometimes it choose a move that is not a best by adding noise on root node.

Game records
From arch000000000000.csa.xz to arch000014300000.csa.xz.
These will be updated each two weeks.
no000000000000.csa to
 are generated by not using neural network, but random function.
The first game that is generated by neural network is
no000001017999.csa. Up to here, 64x15block, window is past 100,000 games.
no000001018000.csa. From here, 256x20block, window is past 500,000 games.
From w000000000001.txt.xz to w000000001782.txt.xz.
Network size is 64 x 15 block up to w448, 256 x 20 block from w449.
w001  ...  64x15b,minibatch   64, learning rate 0.01,    wd 0.00005, momentum 0.9,   120000 games
w156  ...  64x15b,minibatch   64, learning rate 0.001,   wd 0.00005, momentum 0.9,   430000 games
w449  ... 256x20b,minibatch   64, learning rate 0.01,    wd 0.0002,  momentum 0.9,  1018000 games
w465  ... 256x20b,minibatch   64, learning rate 0.001,   wd 0.0002,  momentum 0.9,  1180000 games
w775  ... 256x20b,minibatch 4096, learning rate 0.02,    wd 0.0002,  momentum 0.9,  4220000 games
w787  ... 256x20b,minibatch  128, learning rate 0.0002,  wd 0.0002,  momentum 0.9,  4340000 games
w1450 ... 256x20b,minibatch  128, learning rate 0.00002, wd 0.0002,  momentum 0.9, 10980000 games
w2047 ... 256x20b,minibatch  128, learning rate 0.000002,wd 0.0002,  momentum 0.9, 16948000 games
w2750 ... 256x20b,minibatch  128, learning rate 0.000002,wd 0.00004, momentum 0.9, 23982000 games
w3022 ... 256x20b,minibatch  128, learning rate 0.000002,wd 0.00004, momentum 0.9, 26706447 games
w3077 ... 256x20b,minibatch  128, learning rate 0.000002,wd 0.00004, momentum 0.9, 28352543 games
Weights are updated each  2000 games ( 4000 iterations) up to w448.
Weights are updated each 10000 games (20000 iterations) from  w449.
Weights are updated each 10000 games (10000 iterations) from  w787.
Weights are updated each 34285 games (32000 iterations) from w3022.
Replay buffer is past 1000000 games(from past 500000 games) from w3077.