Skip to content

Latest commit

 

History

History
7 lines (4 loc) · 659 Bytes

File metadata and controls

7 lines (4 loc) · 659 Bytes

Python-Neural-Network-from-scratch

The goal of this project is to implement a functioning neural network in Python without using Deep Learning libraries, and then test how accurately it can predict the curve of a cosine function. The code contains the classes for building neurons, layers and networks, and, also, the instructions to create a simple animation of the network learning the curve mentioned above. Please note that the all the variables and comments in the .py file are written in portuguese, which is my native language.

Animation:

GIF