OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Alpha Gobang Zero

Alpha Gobang Zero is a Python desktop Gomoku game with a reinforcement-learning opponent, a policy-value neural network, and a script for self-play training.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Alpha Gobang Zero — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a documented Python and PyTorch example that combines an interactive Gomoku game with a self-play training entry point.

Who it is for

  • Gomoku players looking for a desktop game against a reinforcement-learning opponent.
  • Developers exploring policy-value networks and self-play training in a board-game project.

Documented uses

  • Play Gomoku through the desktop game interface.
  • Launch the project's reinforcement-learning opponent.
  • Start the included self-play training script.
  • Inspect a network composed of convolutional and residual blocks with policy and value heads.
Before you choose

Requirements & limits

The repository is public and identifies the project as GPL-3.0 licensed. The README lists Windows 10 and Windows 11, Python 3.8, PyTorch 1.8.1 and PyQt 5.15.2.

  • The documented network has one ConvBlock, four ResidueBlocks, one PolicyHead and one ValueHead.
  • The README documents separate game and training entry points: game.py and train.py.
  • PyTorch installation is a separate setup step referenced to an accompanying blog post.
  • Some Windows 10 versions may exhibit a window-dragging issue associated with the acrylic interface background.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Start self-play model training

Launch the project’s model-training script from its Conda environment.

Before you start

  • The Alpha_Gobang_Zero Conda environment created for the project

Steps

  1. Activate the environment with conda activate Alpha_Gobang_Zero.

  2. Start training with the command below.

    Documented command
    python train.py

Expected outcome

The project’s training script is launched for model training.

Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
zhiyiYo/Alpha-Gobang-Zero · Official homepage title · retained in its source language
How we review sourcesSuggest a correction