ViciousSquid/Dosidicus
A cognitive sandbox: Raise digital squids whose brains grow & rewire themselves through Hebbian learning and Neurogenesis
Implements a custom NumPy-based neural simulation engine (STRINg) with fully visible neuron activations, enabling transparent learning mechanics absent in standard deep learning frameworks. Supports both interactive PyQt5 GUI and headless training modes, with persistent cognitive histories saved as exportable brain files that capture how early interactions permanently shape neural structure through Hebbian weight updates and neurogenesis events.
282 stars.
Stars
282
Forks
10
Language
Python
License
GPL-2.0
Category
Last pushed
Mar 11, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ViciousSquid/Dosidicus"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
GoodAI/BrainSimulator
Brain Simulator is a platform for visual prototyping of artificial intelligence architectures.
fargonauts/copycat
Modern port of Melanie Mitchell's and Douglas Hofstadter's Copycat
Paul-G2/letter-spirit-examiner-js
A JavaScript implementation of Letter Spirit's Examiner program
tbaptista/simcx
Simulator for Complex Systems
bjornregnell/scai
How does an AI work? Make your own simple AI using the Scala programming language.