Skip to content

Building Neuron networks from scratch to better understand the ready made libraries and frameworks.

Notifications You must be signed in to change notification settings

eslamdyab21/NN-from-scratch

Repository files navigation

NN-from-scratch

Building Neuron networks from scratch to better understand the ready made libraries and frameworks.

In this repo, a model is created using only NumPy, i.e forward and backward propagation, loss function, and the mechanics of the layers with weights are implemented from scratch. Starting from P1 (basic neuron) to P6 and NN (the coplete class/ish)

NN.py contains the implemented classes and is used in the NN from scratch - mnist digits2.ipynb notebook in which a model for digits classification is built and trained.

Then some models are made using tensoreflow in Tensoreflow folder.

About

Building Neuron networks from scratch to better understand the ready made libraries and frameworks.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published