Neural_Networks_from_Scratch

Neural Networks From Scratch - Using only Numpy

Paper: The Core Mathematics of a Neural Network | Video | Data

Get started

You can set up the environment with all dependencies like so:

conda create --name NN-FS python=3.8
conda activate NN-FS
pip install numpy pandas

High-Level structure

• Paper: Explores in more detail the core mathematics of the neural network.
• data: MNIST Dataset of handwritten digits
• model: Simple 3-Layer Neural Network

How to Run

1. Please download the python or jupyter-notebook file

2. Dowload the dataset - train.csv from (https://www.kaggle.com/competitions/digit-recognizer/data?select=train.csv), and place into a folder named mnist_data

   • At this point your project structure should look like :

├── Neural_Network_from_Scratch.ipynb
├── Neural_Network_from_Scratch.py
└── mnist_data
    └── train.csv

3. Then either run the cells of jupyter notebook or run the python file using

python Neural_Network_from_Scratch.py

Disclaimer(s)

1. The result produced by this code might be slightly different when running on a different GPU.