Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/gnu11111/zoo

Zoo (Genetic Algorithm and Neural Networks Playfield)
https://github.com/gnu11111/zoo

genetic-algorithms kotlin neural-networks simulation

Last synced: 18 days ago
JSON representation

Zoo (Genetic Algorithm and Neural Networks Playfield)

Awesome Lists containing this project

README

        

# Zoo, a play-field for genetic algorithms and neural networks

`Zoo` is a simulation of creatures ('blobs') operating on a rectangular field, using simple neural networks to calculate actions out of neural stimulus from 'sensors'. Inspired by this YouTube video: [I programmed some creatures. They Evolved.](https://www.youtube.com/watch?v=N3tRFayqVtk).

**Note: use `x`, `q` or `` to exit the simulation. Do not use the close-icon from the window-frame or else the programm will continue to run and use up your CPU and RAM!**

---

![Screenshot of the simulation](Zoo.png)

---

Build with Apache Maven:

`mvn clean package`

Run in a Java Runtime:

```
USAGE: java -jar target/zoo-0.8.2-jar-with-dependencies.jar [-c ] [-g ] [-d ] [-q] [-?]

-c ... always use this context
-g ... use this genom as starting-point for every simulation
-d ... delay between steps for first simulation [ms]
-q ... quiet mode, do not ping
```

Use the following keys to control the simulation:
```
Accelerate the simulation
Decelerate the simulation
Only show the result of each generation (toggle)
Rendering on/off (toggle)
S Skip to the end of the current generation
P Pause the simulation (toggle)
L Log world context and genom of the first blob
N, , Start the next (random) simulation
X, Q, Exit the programm
```