Genetic algorithm which evolves aoe2 DE ai scripts

Overview

AlphaScripter

Use the power of genetic algorithms to evolve AI scripts for Age of Empires II : Definitive Edition. For now this package runs in AOC Userpatch 1.5 to train. In theory, scripts generated this way should be compatible with Age of Empires II : Definitive Edition, but in practice a certain amount of porting will need to take place. This is currently a work in progress.

Dependencies

  • msgpackrpc >= 0.4.1 : A package used to communicate with running AOE processes. (To install, pip install msppack-rpc-python)
  • psutil >= 5.8.0 : A package used to manage running processes.
  • tornado == 4.3.5 : Should be automatically installed with msgpackrpc.

How to install and run

  1. Install 32-bit (!) Python (tested with version 3.9) and the dependencies listed above.
  2. Install Age of Empires II - The Conquerors and install UserPatch 1.5. The UserPatch can be found here: https://userpatch.aiscripters.net/
  3. Download the aoc-auto-game.dll and paste it in the same folder as the AOC executable. You can download this DLL file here.
  4. Open the Python file Main.py, adjust parameters and run the script. (WIP)

What does it do? - Explanation per script

The Main Script Main.py

On run, the script will generate an AI named "Alpha," load it into a game versus the training ai (Alpha must be in the second slot). The script will automatically start new games from the post game menu -- you may want to speed up the game in the first round, and/or end the first round early as it won't count for scoring.

Your game may crash. If it does, you can pause the script with control-v and reset. It will save the progress.

The best script so far will be saved as "best.per" in the .ai directory. It will be overwritten if you restart the script.

The png files are necessary for the auto-load new game function.

I will expand this later -- if you would like to help with this project, you can find me on the AI scripters discord for aoe2 de and dm me. I specifically need help from those knowledgeable about scripting or engine modification.

Run types: You can run vs, run score, or run FFA.

run vs: load alpha into p1 and beta into p2. Game will pick winner as new parent - this is a good adversarial AI but is best for late stage training once the AI is good enough to possibly defeat another player

run score: load training AI (HD, barbarian, extreme) into p1 and alpha into p2 Game will pick all-time-highest scorers as new parent.

run FFA: load alpha-h into slots 1-8, make sure no teams are selected Game will pick two best in each round and crossover their traits. Very good for fast training early on.

Game Launcher game_launcher.py

The game launcher is used to, surprise, launch games. To successfully launch a game, the class Launcher needs to be instantiated. This will take 1 optional argument: path that specifies the path to the AOC executable. If you don't pass this argument, the game will look for the executable in the default path: C:\Program Files\Microsoft Games\age of empires ii\Age2_x1\age2_x1.exe. If the executable cannot be found, the launcher will raise an Exception.

After correctly instantiating this Launcher, you can launch a game using the Launcher.launch_game function. This function takes 2 required and 3 optional arguments:

  • (Required) names : list[str] - This is a list of strings that represent the names of the AI .per files that the game will look for when starting the game. These should be in the ..\age of empires ii\ai folder.
  • (Required) game_settings - This should be an instance of the GameSettings class, which is also declared in game_launcher.py.
  • (Optional) real_time_limit : int - The number of real time seconds after which the game(s) should be automatically quit and closed. If not given, there will not be a real-time limit.
  • (Optional) game_time_limit : int - The number of in-game seconds after the game(s) should be quit and closed.
  • (Optional) instances : int - The number of games to run simultaneously. This currently a little experimental, so keep in mind that your mileage with this setting.

The GameSettings class is used to store settings for the game. To instantiate a GameSettings object, you will have to pass 1 required argument and a lot of optional arguments. See below for a description.

  • (Required) civilisations - A list of strings or int (can also be mixed) that represent the civs that the AI players will use. It must be the same length as the names given to the launcher, otherwise the launcher will raise an Exception. If an element in the given list is not a valid civ, that element will be replaced by 'huns'.
  • (Optional) map_id - A string or int that represents the map. Default 'arabia'
  • (Optional) map_size - A string or int that represent the map size. Default 'medium'
  • (Optional) difficulty - A string or int that represents the difficulty of the game. Default 'hard'
  • (Optional) game_type - Specifies the game type. Default 'random_map'
  • (Optional) resources - Specifies the starting resources of each player. Default 'low'
  • (Optional) reveal_map - Whether the map should be revealed. Default 'normal'
  • (Optional) starting_age - Which age the players should start in. Default 'dark'
  • (Optional) victory_type - The victory type. Doesn't work (yet). Default 'conquest'
The test data, code and detailed description of the AW t-SNE algorithm

AW-t-SNE The test data, code and result of the AW t-SNE algorithm Structure of the folder Datasets: This folder contains two datasets, the MNIST datas

1 Mar 09, 2022
Fedlearn algorithm toolkit for researchers

Fedlearn algorithm toolkit for researchers

89 Nov 14, 2022
Benchmark for Robustness Tests of Control Alrogithms

A gym-like classical control benchmark for evaluating the robustnesses of control and reinforcement learning algorithms.

Kim Taekyung 4 Jan 18, 2022
Python Sorted Container Types: Sorted List, Sorted Dict, and Sorted Set

Python Sorted Containers Sorted Containers is an Apache2 licensed sorted collections library, written in pure-Python, and fast as C-extensions. Python

Grant Jenks 2.8k Jan 04, 2023
Official implementation of "Path Planning using Neural A* Search" (ICML-21)

Path Planning using Neural A* Search (ICML 2021) This is a repository for the following paper: Ryo Yonetani*, Tatsunori Taniai*, Mohammadamin Barekata

OMRON SINIC X 82 Jan 07, 2023
🧬 Training the car to do self-parking using a genetic algorithm

🧬 Training the car to do self-parking using a genetic algorithm

Oleksii Trekhleb 652 Jan 03, 2023
FingerPy is a algorithm to measure, analyse and monitor heart-beat using only a video of the user's finger on a mobile cellphone camera.

FingerPy is a algorithm using python, scipy and fft to measure, analyse and monitor heart-beat using only a video of the user's finger on a m

Thiago S. Brasil 37 Oct 21, 2022
A Python library for simulating finite automata, pushdown automata, and Turing machines

Automata Copyright 2016-2021 Caleb Evans Released under the MIT license Automata is a Python 3 library which implements the structures and algorithms

Caleb Evans 219 Dec 12, 2022
CLI Eight Puzzle mini-game featuring BFS, DFS, Greedy and A* searches as solver algorithms.

🕹 Eight Puzzle CLI Jogo do quebra-cabeças de 8 peças em linha de comando desenvolvido para a disciplina de Inteligência Artificial. Escrito em python

Lucas Nakahara 1 Jun 30, 2021
Tic-tac-toe with minmax algorithm.

Tic-tac-toe Tic-tac-toe game with minmax algorithm which is a research algorithm his objective is to find the best move to play by going through all t

5 Jan 27, 2022
A simple library for implementing common design patterns.

PyPattyrn from pypattyrn.creational.singleton import Singleton class DummyClass(object, metaclass=Singleton): # DummyClass is now a Singleton!

1.7k Jan 01, 2023
A library for benchmarking, developing and deploying deep learning anomaly detection algorithms

A library for benchmarking, developing and deploying deep learning anomaly detection algorithms Key Features • Getting Started • Docs • License Introd

OpenVINO Toolkit 1.5k Jan 04, 2023
Algorithms-in-Python - Programs related to DSA in Python for placement practice

Algorithms-in-Python Programs related to DSA in Python for placement practice CO

MAINAK CHAUDHURI 2 Feb 02, 2022
How on earth can I ever think of a solution like that in an interview?!

fuck-coding-interviews This repository is created by an awkward programmer who always struggles with coding problems on LeetCode, even with some Easy

Vinta Chen 613 Jan 08, 2023
Nature-inspired algorithms are a very popular tool for solving optimization problems.

Nature-inspired algorithms are a very popular tool for solving optimization problems. Numerous variants of nature-inspired algorithms have been develo

NiaOrg 215 Dec 28, 2022
An implementation of ordered dithering algorithm in python as multimedia course project

One way of minimizing the size of an image is to simply reduce the number of bits you use to represent each pixel.

7 Dec 02, 2022
A Python program to easily solve the n-queens problem using min-conflicts algorithm

QueensProblem A program to easily solve the n-queens problem using min-conflicts algorithm Performances estimated with a sample of 1000 different rand

0 Oct 21, 2022
FPE - Format Preserving Encryption with FF3 in Python

ff3 - Format Preserving Encryption in Python An implementation of the NIST approved FF3 and FF3-1 Format Preserving Encryption (FPE) algorithms in Pyt

Privacy Logistics 42 Dec 16, 2022
Python Package for Reflection Ultrasound Computed Tomography (RUCT) Delay And Sum (DAS) Algorithm

pyruct Python Package for Reflection Ultrasound Computed Tomography (RUCT) Delay And Sum (DAS) Algorithm The imaging setup is explained in these paper

Berkan Lafci 21 Dec 12, 2022
Implementation of an ordered dithering algorithm used in computer graphics

Ordered Dithering Project In this project, we use an ordered dithering method to turn an RGB image, first to a gray scale image and then, turn the gra

1 Oct 26, 2021