Z2I

The pbrain-Z2I is just a gomoku engine which use the protocol of gomocup. A GUI named qpiskvork can load this engine.

Supported Games

Currently only Gomoku and similar games such as Tic-Tac-Toe.

Welcome other game implementions if you want to become the contributor!

Supported rules

• free-style
• standard
• renju
• caro

Supported OS System

Linux/Windows (tested on Ubuntu 20 + GPU/CPU and Windows 10 + GPU/CPU)

Supported Enviroment

Both GPU and CPU (GPU test on Tesla V100 + Cuda 11 and CPU test on Intel i5/i7)

Language

C++ (for speed!) and python. The model is trained by pytorch (Python) and onnxruntime (C++, for selfplay), and inferenced by onnxruntime (C++).

Dependence

gcc (Linux) or visual studio 19 (windows)

cmake 3.13+

pytorch (tested on 2.20)

onnxruntime-gpu (tested on 1.18)

Installation

Download and install miniconda / python and pip install all the dependent packages such as pytorch

train.sh: convert /data/miniconda3/bin/python to python or python3 or your own python intepreter path

Download onnxruntime: https://github.com/microsoft/onnxruntime/releases/tag/v1.18.1

CMakefiles.txt: convert the onnxruntime path to your own path

git clone --recursive https://github.com/Joker2770/Z2I.git
# git submodule update --init --recursive
cd Z2I
mkdir build
cp ./scripts/*.sh ./build/
cd ./build
cmake ..    # (or 'cmake -A x64 ..')
cmake --build . --config Release   # (or open .sln file through visual Studio 19 and generate for win10)

or build with xmake toolchain(In this way, onnruntime-1.11.1 will be installed with xrepo).

git clone --recursive https://github.com/Joker2770/Z2I.git
# git submodule update --init --recursive
cd Z2I
mkdir build
cp ./scripts/*.sh ./build/
xmake config --mode=release
xmake

Train (Linux)

cd ./build/
bash train.sh
...

If you want to train the model on windows 10, convert train.sh to train.bat and change corresponding commands.

Human play with AI (inference)

1. Run pbrain-Z2I, for example on Linux:

./pbrain-Z2I

There is a better way to run pbrain-Z2I with the GUI program named qpiskvork besides.

Increase or decrease num_mct_sims in config.toml (default 100) to increase the power or speed of AI.

2. Run mcts_test, for example on Linux:

./mcts_test ./weights/1000.onnx 1

Here 1(or 0) = AI play with black(or white) pieces.

Increase or decrease NUM_MCT_SIMS in src/common.h (default 1600) to increase the power or speed of AI.

weight struct

reference

1. https://onnxruntime.ai/docs/execution-providers/CUDA-ExecutionProvider.html
2. https://onnxruntime.ai/docs/execution-providers/OpenVINO-ExecutionProvider.html
3. https://netron.app/