casparwylie/Perceptron
A flexible artificial neural network builder to analyse performance, and optimise the best model.
Implements neural networks from first principles without external ML libraries, exposing lower-level mathematical operations for educational transparency. Provides visual comparison tools for testing different network architectures and hyperparameters side-by-side to understand their impact on training performance. Targets researchers and students seeking hands-on experimentation with ANN design rather than production deployment.
415 stars. No commits in the last 6 months.
Stars
415
Forks
38
Language
Python
License
—
Category
Last pushed
May 21, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/casparwylie/Perceptron"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
Devignitor/Single-neuron-model-Simple-Neural-Network
Single-neuron model trained by gradient descent. It is the most fundamental simple neural network.
TronixLab/Neurona-rgbClassifier
An easy guide for Color classification (RGB) using Multilayer Perceptron (MLP) Neural Networks...
sushantPatrikar/XOR-Gate-With-Neural-Network-Using-Numpy
XOR gate which predicts the output using Neural Network :fire:
pasqal-io/perceptrain
Machine learning training, simplified, at scale
Vinetos/neural-network-xor
Implements a neural network learning XOR gate in your favourite languages !