Skip to content
#

from-scratch-in-python

Here are 9 public repositories matching this topic...

Language: Python
Filter by language

Over 4 weeks, I manually developed a fully connected neural network using only NumPy—without any ML frameworks. It uses Xavier and He initialization, Leaky ReLU activation, L2 regularization, and learning rate decay. The model achieves up to 91% test accuracy and visualizes training, validation, and test losses for early stopping and evaluation.

  • Updated Jun 30, 2025
  • Python

Improve this page

Add a description, image, and links to the from-scratch-in-python topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the from-scratch-in-python topic, visit your repo's landing page and select "manage topics."

Learn more