Neural Network in Python

This project was created to test different ideas and approaches in the field of neural networks. Here, I implemented my first simple neural network from scratch and trained it on the classic Iris Dataset.

In addition to the main model, I also wrote a small custom library — originally tailored for the Iris dataset, but designed to be easily scalable and adaptable to other tasks.

Feature

• A fully from-scratch neural network written in pure Python
• Flexible configuration of the architecture and activation functions
• Training and testing on the Iris Dataset
• Clean and readable code structure for experiments

Goal

This is an educational and experimental project — I use it to explore new ideas and better understand how neural networks work under the hood.

Install and run

• Git clone

  git clone https://github.com/ownNurzh/neural-network-in-python.git

• Poetry install
  Pip

  pip install poetry

  
  Windows Powershell

  (Invoke-WebRequest -Uri https://install.python-poetry.org -UseBasicParsing).Content | py -

  
  Linux

  curl -sSL https://install.python-poetry.org | python3 -

• Install dependencies

  poetry install

• Run

  poetry run main.py