KataGo v1.3.
4
https://github.com/lightvector/KataGo
-----------------------------------------------------
USAGE:
-----------------------------------------------------
FIRST:
Run this to make sure KataGo is working, with a neural net file.
katago.exe benchmark -model <NEURALNET>.bin.gz
(download neural nets here if you don't have one:
https://d3dndmfyhecmj0.cloudfront.net/g170/neuralnets/index.html)
On OpenCL, it should also cause KataGo to tune for your GPU. Then, the benchmark
will report stats about speed and threads. You can configure gtp_example.cfg to use
that many numSearchThreads to get good performance.
ALSO:
You can also run this command to have KataGo generate a gtp config for you, and
automatically tune the number of threads and other parameters
and other settings based on your answers to various questions.
katago.exe genconfig -model <NEURALNET>.bin.gz -output gtp_custom.cfg
NEXT:
This command will run the KataGo engine proper. Feed this command to any program
GUI program to launch KataGo's engine:
katago.exe gtp -model <NEURALNET>.bin.gz
Or if you generated a config yourself:
katago.exe gtp -model <NEURALNET>.bin.gz -config gtp_custom.cfg
KataGo should be able to work with any GUI program that supports GTP, as well as
any analysis program that supports Leela Zero's `lz-analyze` command, such as
Lizzie (https://github.com/featurecat/lizzie) or Sabaki
(https://sabaki.yichuanshen.de/).
NOTE:
If you encounter errors due to a missing "msvcp140.dll" or "msvcp140_1.dll" or
"msvcp140_2.dll" or "vcruntime140.dll" or similar, you need to download and install
the Microsoft Visual C++ Redistributable, here:
https://www.microsoft.com/en-us/download/details.aspx?id=48145
If this is for a 64-bit Windows version of KataGo, these dll files have already
been included for you, otherwise you will need to install them yourself. On a 64-
bit Windows version, there is a rare chance that you may need to delete them if you
already have it installed yourself separately and the pre-included files are
actually causing problems running KataGo.
-----------------------------------------------------
OPENCL VS CUDA:
-----------------------------------------------------
Depending on hardware and settings, in practice the OpenCL version seems to range
from anywhere to several times slower to a little faster than the CUDA version.
More optimization work may happen in the future though - the OpenCL version has
definitely not reached the limit of how well it can be optimized. It also has not
been tested for self-play training with extremely large batch sizes to run hundreds
of games in parallel, all of KataGo's main training runs so far have been performed
with the CUDA implementation.
Extensive testing across different OSs and versions and compilers has not been
done, so if you encounter issues, feel free to open an issue.
-----------------------------------------------------
TUNING FOR PERFORMANCE:
You will very likely want to tune some of the parameters in `default_gtp.cfg` for
your system for good performance, including the number of threads, fp16 usage (CUDA
only), NN cache size, pondering settings, and so on. You can also adjust things
like KataGo's resign threshold or utility function. Most of the relevant parameters
should be be reasonably well documented directly inline in that config.
There are other a few notes about usage and performance at :
https://github.com/lightvector/KataGo
