trsvchn / interactive-perceptrons

Tune weights manually.

Interactive Perceptrons

The main aim of this nanoproject is to help you understand intuitively and interactively the role of Weights and Biases. Build logical operators by manually tweaking weights and biases values. Inspired by Lesson 2.8 "Perceptrons as Logical Operators" from Udacity course Intro to Deep Learning with PyTorch.

Perceptrons Included:

• AND Perceptron
• OR Perceptron
• NOT Perceptron
• "XOR Perceptron"
• XOR 2-Layer Perceptron

Requirements

python3 jupyter numpy matplotlib ipywidgets

Installation

1. Clone this repo

git clone https://github.com/trsvchn/interactive-perceptrons.git

2. Install dependencies

pip install -r requirements.txt
jupyter nbextension enable --py widgetsnbextension

Usage

Open perceptrons.ipynb in your jupyter.

Have fun!