俞航
俞航
@Zeta36 I've created about 3.5GB game play data this Thanksgiving break. But I don't know where to find evaluation log? I will upload the best model to google drive so...
@Zeta36 I just noticed that the self-play pipeline is manual. 😂 I am trying to train it with ```opt``` mode but 1) the total loss blows up to NaN but...
@Zeta36 I have to apologize that I didn't notice it earlier. Since my computer (gaming laptop) is not a dedicated server, I can't promise your that these programmes will run...
@wonderkid27 It happens because the last commit has uncommented this [line](https://github.com/Zeta36/chess-alpha-zero/blob/5c794a7f2ff419e45e6570f0d66cdd4722e43708/src/chess_zero/agent/player_chess.py#L71) for the purpose of training. Does it resolve your problem?
@yuandong-tian @zhanghongjie101 I modified the commented flag in ```eval_minirts.sh``` (```#--omit_keys Wt,Wt2,Wt3```) to be yours, it fails to read it. ``` Load from ./rts/game_MC/model/model-winrate-80.0-357800.bin Omit_keys = ['Wt', 'Wt1', 'Wt2', 'Wt3'] /home/ubuntu/miniconda3/envs/elf/lib/python3.6/site-packages/torch/serialization.py:286:...
@derceg I have the exact same problem as @cristianCanek. Does your commit solve this problem? How can enbale this feature? I am using v1.3.5 (stable).
@hxhb is it solve? Can we use Hotpatcher for UE5 release version already?
@kstenschke thanks, this is helpful
@cleeus thanks, this is helpful
win10 x64 user, nvidia 430.16 works