KataGo
KataGo copied to clipboard
lightvector
GTP engine and self-play learning in Go
# Issue When evaluating an input using the TensorRT backend in `match` or `selfplay`, I sometimes hit a `Got nonfinite for policy sum` error that I never hit when using...
Hello, thank you for creating Katago. I have benefited tremendously. I heard that Katago evaluates an estimate of how uncertain or inaccurate the winrates and scores are. I was just...
Hey, I thought that it could really accelerate NN architecture research if we could create such a dataset. The biggest win is that wide array of researchers would be able...
This is the training result in my project. The ```fixup``` is Fixup Initialization (with gradient clipping). The ```renorm``` is Batch Renormalization. The two network both use the same training data...
Hi, I'd like to run KataGo on my M1 macbook with native support for CoreML but sadly, I don't know the framework plus I have no experience with macOS development....
Has anyone verified the pytorch-retwrite branch? I want to use the pytorch code, is it neccessary to test it with the tf codes ?
I am building conda packages for katago and the current builds requires the CMake config to be patched to work on osx. Here is the patch: https://github.com/conda-forge/staged-recipes/pull/20777#discussion_r997638970
Probably not optimal, but it's a start :) Linking to a libzip.a compiled with ``` cmake .. -DLIBZIP_DO_INSTALL=0 -DBUILD_SHARED_LIBS=0 -DENABLE_BZIP2=0 -DENABLE_LZMA=0 -DENABLE_COMMONCRYPTO=0 -DENABLE_GNUTLS=0 -DENABLE_MBEDTLS=0 -DENABLE_OPENSSL=0 -DENABLE_WINDOWS_CRYPTO=0 ``` appears to work....
So that could use CUDA 11.0 + tensorflow_gpu 2.4.0, makes synchronous training on a single machine possible.
Tuning hGemmWmma for convolutions Testing 146 different configs LLVM ERROR: Couldn't allocate input reg for constraint 'l'!
