Life In 19x19 http://lifein19x19.com/ |
|
KataGo v1.8.0 http://lifein19x19.com/viewtopic.php?f=18&t=18000 |
Page 1 of 1 |
Author: | lightvector [ Wed Jan 13, 2021 10:40 pm ] |
Post subject: | KataGo v1.8.0 |
KataGo v1.8.0 is released! Release notes and download here: https://github.com/lightvector/KataGo/releases Notably, it should now support the nets being trained in the new run at https://katagotraining.org/. There are also bugfixes and changes to the search that should add a minor amount of improved accuracy/strength even without changing the network used. Enjoy! |
Author: | lightvector [ Thu Jan 14, 2021 7:12 am ] |
Post subject: | Re: KataGo v1.8.0 |
Just for fun, I ran some matches against a few benchmark opponents (also cross-posted in discord chat). Here are the win/loss stats. Test 1 KataGo running on a slightly pre-v1.8.0 version using kata1 s5.51G, 1 thread with shared cache/batch up to 80 games at a time using match engine, 19x19 board with random rules, auto komi. 800 playouts per move... ... versus KataGo v1.7.0 settings with the strongest g170 40-block network that we've had for months prior to this new run, 800 playouts: 858.5 : 638.5 (57%) ... versus KataGo v1.7.0 settings with the strongest g170e "extended training" 15-block network, 16000 playouts: 117 : 20 (85%) Test 2 KataGo v1.8.0 using kata1 s5.60G, 32 threads, resign at 5%, futileVisitsThreshold=0.15, otherwise mostly default settings. 19x19 board with tromp-taylor rules. 7.5 komi. 800 playouts per move... ... versus LZ285 2000 playouts: 375:13 (96.6%) ... versus LZ-ELFv2 16000 playouts: 319:68 (82.3%) ... versus LZ157 48000 playouts: 88:4 (95.7%) All LZs used, 32 threads, batchsize 16, resign at 5%, temp 0.3 for 20 moves. Except ELF, which used resign at 2% since its value sharpness is disproportionate relative to LZ nets of the same strength. ------------- So it looks like we're doing well against many of the standard bots from years prior even with a large handicap in playouts (and what should be less total computation time, even taking into account that some of these other networks are smaller and cheaper to evaluate). Fun to get a sense of how far KataGo has come from the beginning. Hopefully, the road ahead will be long and interesting too. https://cdn.discordapp.com/attachments/ ... 0/sgfs.zip https://cdn.discordapp.com/attachments/ ... /sgfs2.zip |
Author: | Friday9i [ Thu Jan 14, 2021 5:51 pm ] |
Post subject: | Re: KataGo v1.8.0 |
Amazing job, congrats! |
Author: | goame [ Fri Jan 15, 2021 11:57 am ] |
Post subject: | Re: KataGo v1.8.0 |
lightvector wrote: KataGo v1.8.0 is released! Release notes and download here: https://github.com/lightvector/KataGo/releases Notably, it should now support the nets being trained in the new run at https://katagotraining.org/. There are also bugfixes and changes to the search that should add a minor amount of improved accuracy/strength even without changing the network used. Enjoy! Congrats Question: When I double click the katago.exe in the folder to run it, I get the answer: cudnn64_8.dll and cudnn64_11.dll are missing. Windows 10 2x RTX 2080 Ti newest update version. I downloaded the newest Lizzie. I downloaded the newest KataGo v1.8.0 into the Lizzie KataGo folder. |
Author: | lightvector [ Fri Jan 15, 2021 12:19 pm ] |
Post subject: | Re: KataGo v1.8.0 |
NVIDIA's licensing terms terms make it not easy to legally distribute any of those DLLs. So legally speaking, you are attempting to use the CUDA version, it is up to you to install both CUDA and CUDNN (the latter of which requires that you sign up for a free "developer account" at NVIDIA's website) and make sure that the appropriate "dll" files are within your library search path on Windows or otherwise copied to where they need to be so that they can be found. The latest windows release (unless you are compiling a custom version from source) needs CUDA 11.2 and CUDNN 8. Anyways, this is why the official recommendation (https://github.com/lightvector/KataGo#o ... a-vs-eigen) recommends the OpenCL version to everyone. CUDA is only if you are willing to do the technical work yourself, and care a lot about every bit of performance - including the significant chance that at the end, the work you do isn't worth anything because it could be that on your system the OpenCL version is faster anyways. |
Author: | lightvector [ Fri Jan 15, 2021 12:26 pm ] |
Post subject: | Re: KataGo v1.8.0 |
By the way, if you've done that work in the past before to get an older CUDA version of KataGo working, then probably the answer is that you had CUDNN 7 set up, while the newer version needs CUDNN 8 - just repeat the process and upgrade to CUDNN 8 instead. |
Author: | And [ Fri Jan 15, 2021 1:21 pm ] |
Post subject: | Re: KataGo v1.8.0 |
How big can theoretically be the difference in performance between CUDA and OpenCL versions on a powerful video card? and the CUDA version is likely to be faster? |
Author: | goame [ Fri Jan 15, 2021 2:23 pm ] |
Post subject: | Re: KataGo v1.8.0 |
lightvector wrote: By the way, if you've done that work in the past before to get an older CUDA version of KataGo working, then probably the answer is that you had CUDNN 7 set up, while the newer version needs CUDNN 8 - just repeat the process and upgrade to CUDNN 8 instead. Thx. I downloaded CUDNN 8 - this problem is now fixed. But it still shows me that I need cublas64_11.dll. Where I can find and download it? And into which folder? |
Author: | Friday9i [ Fri Jan 15, 2021 2:52 pm ] |
Post subject: | Re: KataGo v1.8.0 |
On RTX cards, OpenCL and Cuda version are almost equal in terms of speed. So unless you get access to pro cards such as V100, you'd better go for openCL anyway : as fast and simpler! |
Author: | goame [ Sat Jan 16, 2021 9:43 am ] |
Post subject: | Re: KataGo v1.8.0 |
lightvector wrote: NVIDIA's licensing terms terms make it not easy to legally distribute any of those DLLs. So legally speaking, you are attempting to use the CUDA version, it is up to you to install both CUDA and CUDNN (the latter of which requires that you sign up for a free "developer account" at NVIDIA's website) and make sure that the appropriate "dll" files are within your library search path on Windows or otherwise copied to where they need to be so that they can be found. The latest windows release (unless you are compiling a custom version from source) needs CUDA 11.2 and CUDNN 8. Anyways, this is why the official recommendation (https://github.com/lightvector/KataGo#o ... a-vs-eigen) recommends the OpenCL version to everyone. CUDA is only if you are willing to do the technical work yourself, and care a lot about every bit of performance - including the significant chance that at the end, the work you do isn't worth anything because it could be that on your system the OpenCL version is faster anyways. Thx. I have fixed my two problems. And have also looked here: https://lifein19x19.com/viewtopic.php?t=17317 Now tuning looks like this: Z:\>LG0\Lizzie\katago\katago.exe genconfig -model \LG0\Lizzie\katago\katanetwork.gz -output gtp_custom.cfg ========================================================================= RULES What rules should KataGo use by default for play and analysis? (chinese, japanese, korean, tromp-taylor, aga, chinese-ogs, new-zealand, bga, stone-scoring, aga-button): japanese ========================================================================= SEARCH LIMITS When playing games, KataGo will always obey the time controls given by the GUI/tournament/match/online server. But you can specify an additional limit to make KataGo move much faster. This does NOT affect analysis/review, only affects playing games. Add a limit? (y/n) (default n): n NOTE: No limits configured for KataGo. KataGo will obey time controls provided by the GUI or server or match script but if they don't specify any, when playing games KataGo may think forever without moving. (press enter to continue) When playing games, KataGo can optionally ponder during the opponent's turn. This gives faster/stronger play in real games but should NOT be enabled if you are running tests with fixed limits (pondering may exceed those limits), or to avoid stealing the opponent's compute time when testing two bots on the same machine. Enable pondering? (y/n, default n):y Specify max num seconds KataGo should ponder during the opponent's turn. Leave blank for no limit: ========================================================================= GPUS AND RAM Finding available GPU-like devices... Found CUDA device 0: GeForce RTX 2080 Ti Found CUDA device 1: GeForce RTX 2080 Ti Specify devices/GPUs to use (for example "0,1,2" to use devices 0, 1, and 2). Leave blank for a default SINGLE-GPU config: 0,1 By default, KataGo will cache up to about 3GB of positions in memory (RAM), in addition to whatever the current search is using. Specify a different max in GB or leave blank for default: 64 ========================================================================= PERFORMANCE TUNING Specify number of visits to use test/tune performance with, leave blank for default based on GPU speed. Use large number for more accurate results, small if your GPU is old and this is taking forever: 100000 Specify number of seconds/move to optimize performance for (default 5), leave blank for default: 1 2021-01-16 16:07:21+0100: Loading model and initializing benchmark... 2021-01-16 16:07:21+0100: nnRandSeed0 = 13173919156662199898 2021-01-16 16:07:21+0100: After dedups: nnModelFile0 = \LG0\Lizzie\katago\katanetwork.gz useFP16 auto useNHWC auto 2021-01-16 16:07:23+0100: Cuda backend thread 0: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 16:07:23+0100: Cuda backend thread 1: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 16:07:23+0100: Cuda backend thread 0: Model version 10 useFP16 = true useNHWC = true 2021-01-16 16:07:23+0100: Cuda backend thread 1: Model version 10 useFP16 = true useNHWC = true 2021-01-16 16:07:23+0100: Cuda backend thread 0: Model name: kata1-b40c256-s5675792640-d1366587029 2021-01-16 16:07:23+0100: Cuda backend thread 1: Model name: kata1-b40c256-s5675792640-d1366587029 ========================================================================= TUNING NOW Tuning using 100000 visits. Automatically trying different numbers of threads to home in on the best: 2021-01-16 16:07:27+0100: nnRandSeed0 = 2285250991643616650 2021-01-16 16:07:27+0100: After dedups: nnModelFile0 = \LG0\Lizzie\katago\katanetwork.gz useFP16 auto useNHWC auto 2021-01-16 16:07:29+0100: Cuda backend thread 0: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 16:07:29+0100: Cuda backend thread 1: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 16:07:29+0100: Cuda backend thread 0: Model version 10 useFP16 = true useNHWC = true 2021-01-16 16:07:29+0100: Cuda backend thread 1: Model version 10 useFP16 = true useNHWC = true 2021-01-16 16:07:29+0100: Cuda backend thread 0: Model name: kata1-b40c256-s5675792640-d1366587029 2021-01-16 16:07:29+0100: Cuda backend thread 1: Model name: kata1-b40c256-s5675792640-d1366587029 Possible numbers of threads to test: 2, 3, 4, 5, 6, 8, 10, 12, 16, 20, 24, 32, 40, 48, numSearchThreads = 6: 10 / 10 positions, visits/s = 1172.56 nnEvals/s = 666.52 nnBatches/s = 354.20 avgBatchSize = 1.88 (852.9 secs) numSearchThreads = 20: 10 / 10 positions, visits/s = 2167.26 nnEvals/s = 1195.58 nnBatches/s = 234.58 avgBatchSize = 5.10 (461.5 secs) numSearchThreads = 12: 10 / 10 positions, visits/s = 1797.95 nnEvals/s = 987.94 nnBatches/s = 291.59 avgBatchSize = 3.39 (556.3 secs) numSearchThreads = 32: 10 / 10 positions, visits/s = 2466.86 nnEvals/s = 1419.08 nnBatches/s = 174.40 avgBatchSize = 8.14 (405.5 secs) numSearchThreads = 40: 10 / 10 positions, visits/s = 2832.08 nnEvals/s = 1532.72 nnBatches/s = 147.34 avgBatchSize = 10.40 (353.2 secs) numSearchThreads = 48: 10 / 10 positions, visits/s = 2865.65 nnEvals/s = 1619.12 nnBatches/s = 124.66 avgBatchSize = 12.99 (349.1 secs) Optimal number of threads is fairly high, increasing the search limit and trying again. 2021-01-16 16:57:39+0100: nnRandSeed0 = 15290355468334568374 2021-01-16 16:57:39+0100: After dedups: nnModelFile0 = \LG0\Lizzie\katago\katanetwork.gz useFP16 auto useNHWC auto 2021-01-16 16:57:41+0100: Cuda backend thread 0: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 16:57:41+0100: Cuda backend thread 1: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 16:57:41+0100: Cuda backend thread 0: Model version 10 useFP16 = true useNHWC = true 2021-01-16 16:57:41+0100: Cuda backend thread 1: Model version 10 useFP16 = true useNHWC = true 2021-01-16 16:57:41+0100: Cuda backend thread 0: Model name: kata1-b40c256-s5675792640-d1366587029 2021-01-16 16:57:41+0100: Cuda backend thread 1: Model name: kata1-b40c256-s5675792640-d1366587029 Possible numbers of threads to test: 24, 32, 40, 48, 64, 80, 96, 128, numSearchThreads = 80: 10 / 10 positions, visits/s = 3019.80 nnEvals/s = 1771.93 nnBatches/s = 79.95 avgBatchSize = 22.16 (331.4 secs) numSearchThreads = 64: 10 / 10 positions, visits/s = 3021.92 nnEvals/s = 1715.16 nnBatches/s = 96.56 avgBatchSize = 17.76 (331.1 secs) Ordered summary of results: numSearchThreads = 6: 10 / 10 positions, visits/s = 1172.56 nnEvals/s = 666.52 nnBatches/s = 354.20 avgBatchSize = 1.88 (852.9 secs) (EloDiff baseline) numSearchThreads = 12: 10 / 10 positions, visits/s = 1797.95 nnEvals/s = 987.94 nnBatches/s = 291.59 avgBatchSize = 3.39 (556.3 secs) (EloDiff +134) numSearchThreads = 20: 10 / 10 positions, visits/s = 2167.26 nnEvals/s = 1195.58 nnBatches/s = 234.58 avgBatchSize = 5.10 (461.5 secs) (EloDiff +174) numSearchThreads = 32: 10 / 10 positions, visits/s = 2466.86 nnEvals/s = 1419.08 nnBatches/s = 174.40 avgBatchSize = 8.14 (405.5 secs) (EloDiff +181) numSearchThreads = 40: 10 / 10 positions, visits/s = 2832.08 nnEvals/s = 1532.72 nnBatches/s = 147.34 avgBatchSize = 10.40 (353.2 secs) (EloDiff +214) numSearchThreads = 48: 10 / 10 positions, visits/s = 2865.65 nnEvals/s = 1619.12 nnBatches/s = 124.66 avgBatchSize = 12.99 (349.1 secs) (EloDiff +191) numSearchThreads = 64: 10 / 10 positions, visits/s = 3021.92 nnEvals/s = 1715.16 nnBatches/s = 96.56 avgBatchSize = 17.76 (331.1 secs) (EloDiff +163) numSearchThreads = 80: 10 / 10 positions, visits/s = 3019.80 nnEvals/s = 1771.93 nnBatches/s = 79.95 avgBatchSize = 22.16 (331.4 secs) (EloDiff +109) Based on some test data, each speed doubling gains perhaps ~250 Elo by searching deeper. Based on some test data, each thread costs perhaps 7 Elo if using 800 visits, and 2 Elo if using 5000 visits (by making MCTS worse). So APPROXIMATELY based on this benchmark, if you intend to do a 1 second search: numSearchThreads = 6: (baseline) numSearchThreads = 12: +134 Elo numSearchThreads = 20: +174 Elo numSearchThreads = 32: +181 Elo numSearchThreads = 40: +214 Elo (recommended) numSearchThreads = 48: +191 Elo numSearchThreads = 64: +163 Elo numSearchThreads = 80: +109 Elo Using 40 numSearchThreads! ========================================================================= DONE Writing new config file to gtp_custom.cfg You should be now able to run KataGo with this config via something like: LG0\Lizzie\katago\katago.exe gtp -model '\LG0\Lizzie\katago\katanetwork.gz' -config 'gtp_custom.cfg' Feel free to look at and edit the above config file further by hand in a txt editor. For more detailed notes about performance and what options in the config do, see: https://github.com/lightvector/KataGo/b ... xample.cfg Can someone explain the performance tuning part in more detail? 1. Is it better to use 10.000 or 100.000 or 1.000.000 visits for the tuning? Or is it better to use default, which means it depends on gpu speed-but how does it work compared to the other option? 2. How tuning "seconds per move" works in detail? Is it better to use 1 second or 10 or 60 seconds? If I tune for 1 second, would it be better at 1 second per move compared to a tuning at 60 seconds per move? And after tuned for 1 second how good is the elo increase when I use for long analysis? I mean would a 60 seconds per move tuning scale better when going from 1 to 2 to 3 to 4 to...60 seconds compared to the 1 second per move tuning? And how does it look when doing 5 minutes in some special positions? 3.Please compare also the new tuning (elo) with the old one (elo) from link above -> 50000 visits So APPROXIMATELY based on this benchmark, if you intend to do a 5 second search: numSearchThreads = 5: (baseline) numSearchThreads = 6: +57 Elo numSearchThreads = 10: +208 Elo numSearchThreads = 12: +264 Elo numSearchThreads = 16: +334 Elo numSearchThreads = 20: +362 Elo numSearchThreads = 24: +381 Elo numSearchThreads = 32: +408 Elo numSearchThreads = 40: +436 Elo numSearchThreads = 48: +471 Elo (recommended) numSearchThreads = 64: +467 Elo numSearchThreads = 80: +451 Elo Using 48 numSearchThreads! |
Author: | goame [ Sat Jan 16, 2021 11:02 am ] |
Post subject: | Re: KataGo v1.8.0 |
It works with Lizzie. But looking at the GPU-Z tool, when I run KataGo, only 1 GPU is running. The other GPU has only 1% GPU load. Maybe I should try to change 0,1 to something like 1,2 to use both GPUs (maybe 0 is for using the CPU)? 2021-01-16 14:18:50+0100: GTP Engine starting... 2021-01-16 14:18:50+0100: KataGo v1.8.0 2021-01-16 14:18:50+0100: Using TrompTaylor rules initially, unless GTP/GUI overrides this 2021-01-16 14:18:50+0100: Using 10 CPU thread(s) for search 2021-01-16 14:18:50+0100: nnRandSeed0 = 718745476986450218 2021-01-16 14:18:50+0100: After dedups: nnModelFile0 = katanetwork.gz useFP16 auto useNHWC auto 2021-01-16 14:18:52+0100: Cuda backend thread 0: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 14:18:52+0100: Cuda backend thread 0: Model version 10 useFP16 = true useNHWC = true 2021-01-16 14:18:52+0100: Cuda backend thread 0: Model name: kata1-b40c256-s5675792640-d1366587029 2021-01-16 15:08:22+0100: GTP Engine starting... 2021-01-16 15:08:22+0100: KataGo v1.8.0 2021-01-16 15:08:22+0100: Using TrompTaylor rules initially, unless GTP/GUI overrides this 2021-01-16 15:08:22+0100: Using 10 CPU thread(s) for search 2021-01-16 15:08:22+0100: nnRandSeed0 = 7753452469933976295 2021-01-16 15:08:22+0100: After dedups: nnModelFile0 = katanetwork.gz useFP16 auto useNHWC auto 2021-01-16 15:08:24+0100: Cuda backend thread 0: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 15:08:24+0100: Cuda backend thread 0: Model version 10 useFP16 = true useNHWC = true 2021-01-16 15:08:24+0100: Cuda backend thread 0: Model name: kata1-b40c256-s5675792640-d1366587029 2021-01-16 17:48:08+0100: GTP Engine starting... 2021-01-16 17:48:08+0100: KataGo v1.8.0 2021-01-16 17:48:08+0100: Using TrompTaylor rules initially, unless GTP/GUI overrides this 2021-01-16 17:48:08+0100: Using 10 CPU thread(s) for search 2021-01-16 17:48:08+0100: nnRandSeed0 = 7549193501443262033 2021-01-16 17:48:08+0100: After dedups: nnModelFile0 = katanetwork.gz useFP16 auto useNHWC auto 2021-01-16 17:48:10+0100: Cuda backend thread 0: Found GPU GeForce RTX 2080 Ti memory 11811160064 compute capability major 7 minor 5 2021-01-16 17:48:10+0100: Cuda backend thread 0: Model version 10 useFP16 = true useNHWC = true 2021-01-16 17:48:10+0100: Cuda backend thread 0: Model name: kata1-b40c256-s5675792640-d1366587029 2021-01-16 17:48:12+0100: Loaded neural net with nnXLen 19 nnYLen 19 2021-01-16 17:48:12+0100: -------------- 2021-01-16 17:48:12+0100: WARNING: Config had unused keys! You may have a typo, an option you specified is being unused from katago-gtp10.cfg 2021-01-16 17:48:12+0100: WARNING: Unused key 'fpuUseParentAverage' in katago-gtp10.cfg 2021-01-16 17:48:12+0100: -------------- 2021-01-16 17:48:12+0100: Loaded config katago-gtp10.cfg 2021-01-16 17:48:12+0100: Loaded model katanetwork.gz 2021-01-16 17:48:12+0100: Model name: kata1-b40c256-s5675792640-d1366587029 2021-01-16 17:48:12+0100: GTP ready, beginning main protocol loop { "leelaz": { "engine-preload-list": [ false, false, false, false, false, false, false, false, false ], "engine-command-list": [ "./katago/katago gtp -model katanetwork.gz -config katago-gtp10.cfg", "", "", "", "", "", "", "", "" ], "max-analyze-time-minutes": 99999, "analyze-update-interval-centisec": 10, "network-file": "lznetwork.gz", "_comment": "note, network-file is obselete in Lizzie 0.7+, ignore network-file, kept for compatibility", "max-game-thinking-time-seconds": 2, "engine-start-location": ".", "avoid-keep-variations": 30, "engine-command": "./leelazero/leelaz --gtp --lagbuffer 0 --weights lznetwork.gz", "print-comms": false, "show-lcb-winrate": false }, "ui": { "comment-font-size": 0, "hold-bestmoves-to-sgf": true, "shadow-size": 100, "show-winrate": true, "autosave-interval-seconds": -1, "limit-best-move-num": 0, "stone-indicator-type": 1, "katago-estimate-blend": true, "win-rate-always-black": false, "board-width": 19, "show-border": false, "show-move-number": false, "winrate-stroke-width": 3, "show-next-moves": true, "comment-node-color": [ 0, 0, 255 ], "show-comment": true, "show-leelaz-variation": true, "show-bestmoves-by-hold": true, "min-playout-ratio-for-stats": 0.1, "fancy-stones": true, "resume-previous-game": false, "comment-font-color": [ 255, 255, 255 ], "show-coordinates": false, "shadows-enabled": true, "show-katago-estimate-onsubboard": false, "show-variation-graph": true, "show-dynamic-komi": true, "gtp-console-style": "body {background:#000000; color:#d0d0d0; font-family:Consolas, Menlo, Monaco, 'Ubuntu Mono', monospace; margin:4px;} .command {color:#ffffff;font-weight:bold;} .winrate {color:#ffffff;font-weight:bold;} .coord {color:#ffffff;font-weight:bold;}", "katago-scoremean-alwaysblack": false, "katago-notshow-winrate": false, "minimum-blunder-bar-width": 3, "large-winrate": false, "confirm-exit": false, "show-katago-estimate-onmainboard": true, "scoremean-line-color": [ 255, 0, 255 ], "show-katago-estimate": true, "show-best-moves": true, "board-color": [ 217, 152, 77 ], "append-winrate-to-comment": false, "fancy-board": true, "color-by-winrate-instead-of-visits": false, "show-captured": true, "replay-branch-interval-seconds": 1, "panel-ui": false, "blunder-bar-color": [ 255, 0, 0, 150 ], "weighted-blunder-bar-height": false, "katago-estimate-mode": "small+dead", "theme": "Default", "show-winrate-in-suggestion": true, "show-scoremean-in-suggestion": true, "new-move-number-in-branch": true, "winrate-line-color": [ 0, 255, 0 ], "blunder-node-colors": [], "minPlayoutRatioForStats": 0.1, "show-katago-boardscoremean": false, "show-playouts-in-suggestion": true, "limit-branch-length": 0, "blunder-winrate-thresholds": [], "board-position-proportion": 4, "show-blunder-bar": false, "only-last-move-number": 0, "board-height": 19, "winrate-miss-line-color": [ 0, 0, 178 ], "show-status": true, "handicap-instead-of-winrate": false, "large-subboard": false, "comment-background-color": [ 0, 0, 0, 200 ], "dynamic-winrate-graph-width": false, "show-subboard": true, "show-katago-scoremean": true, "show-comment-node-color": false, "board-size": 19 } } |
Author: | lightvector [ Sat Jan 16, 2021 12:40 pm ] |
Post subject: | Re: KataGo v1.8.0 |
For visits, ideally you should use a number of visits that is not far from the number of visits you will often use in practice. Same for seconds per move - specify something not far from the number of seconds per move you will often use in practice. Because in general whenever you tune something, you should tune it to be good for how you will use it. --------------- Also, if you are already going to go through the trouble to try to optimize all of this, it's worth to practice the effort of reading log messages and error messages to understand them yourself. The output will tell you important things! (although, I admit, it can be a bit verbose and cluttered sometimes, this is true of a lot of low-level software, reading error and log messages is actually a pretty useful skill). In particular, notice here: Quote: Writing new config file to gtp_custom.cfg (emphasis mine)You should be now able to run KataGo with this config via something like: LG0\Lizzie\katago\katago.exe gtp -model '\LG0\Lizzie\katago\katanetwork.gz' -config 'gtp_custom.cfg' You used KataGo and tuned it and wrote a new config with the results. But then you told Lizzie to run KataGo with a different config. See right here: goame wrote: "engine-command-list": [ "./katago/katago gtp -model katanetwork.gz -config katago-gtp10.cfg", "", And here from KataGo's log about what config is being used (plus a warning about that config). goame wrote: 2021-01-16 17:48:12+0100: -------------- 2021-01-16 17:48:12+0100: WARNING: Config had unused keys! You may have a typo, an option you specified is being unused from katago-gtp10.cfg 2021-01-16 17:48:12+0100: WARNING: Unused key 'fpuUseParentAverage' in katago-gtp10.cfg 2021-01-16 17:48:12+0100: -------------- 2021-01-16 17:48:12+0100: Loaded config katago-gtp10.cfg Fix those up and things will probably make more sense. |
Author: | goame [ Sun Jan 17, 2021 11:42 am ] |
Post subject: | Re: KataGo v1.8.0 |
It works very good now. Thx |
Author: | Maharani [ Fri Feb 26, 2021 2:45 pm ] |
Post subject: | Re: KataGo v1.8.0 |
lightvector wrote: Maharani wrote: Are there any near-, mid- or long-term plans for KataGo to support saving/loading calculation memory? No, not right now. (This might be a lot of work, I think other things take priority). Pretty pweeeease? |
Page 1 of 1 | All times are UTC - 8 hours [ DST ] |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |