• Search by Module
• Search by Words
• Search Projects

• Java
• Python
• JavaScript
• TypeScript
• C++
• Scala
• Blog

• pysc2-tutorial-master
  • Building a Basic Agent
    • simple_agent.py
    • simple_agent_step2.py
    • simple_agent_step1.py
    • simple_agent_step3.py
    • simple_agent_step4.py
    • simple_agent_step2a.py
  • Battle Predictor
    • tvt.csv
    • train_tvt.py
    • terran_agent.py
    • plot_tvt.py
    • battle_tvt.py
  • Building an Attack Agent
    • attack_agent_step4.py
    • attack_agent_step1.py
    • attack_agent.py
    • attack_agent_step6.py
    • attack_agent_step5.py
    • attack_agent_step2.py
    • attack_agent_step3.py
  • Reinforcement Learning Terran Bot
    • learning_agent.py
  • Build a Raw Protoss Agent
    • raw_agent_step2.py
    • raw_agent_step5.py
    • raw_agent_step3.py
    • raw_agent_step6.py
    • raw_agent_step4.py
    • raw_agent_step7.py
  • LICENSE
  • Building a Sparse Reward Agent
    • sparse_agent_step4.py
    • sparse_agent_step3.py
    • sparse_agent.py
    • sparse_agent_step5.py
    • sparse_agent_step6.py
    • sparse_agent_step2.py
    • sparse_agent_step1.py
    • sparse_agent_step7.py
  • Building a Smart Agent
    • smart_agent_step5.py
    • smart_agent_step1.py
    • smart_agent_step2.py
    • smart_agent_step4.py
    • smart_agent.py
    • smart_agent_step3.py
  • Refining the Sparse Reward Agent
    • refined_agent.py
  • Build a Zerg Bot
    • zerg_agent_step7.py
    • zerg_agent_step2.py
    • zerg_agent_step5.py
    • zerg_agent_step1.py
    • zerg_agent_step6.py
    • zerg_agent_step3.py
    • zerg_agent_step4.py
  • README.md
  • .gitignore

How To: PySC2

Preface

As I explore PySC2 I'm going to document my experiences here in the form of tutorials so you can follow along and build a StarCraft II bot of your own.

In order to keep things simple I will restrict the bot's race to Terran, at least for the first draft. I may expand on the tutorial if it becomes financially viable to do so.

If you see a mistake or some code that could be improved, please let me know.

Table of Contents

1. Building a Basic Agent
2. Building a Simple Reinforcement Learning Agent
3. Add Smart Attacking to Your Agent
4. Building a Sparse Reward Agent
5. Refine Your Sparse PySC2 Agent