Fraetor29 / How-NN-Works

How Neural Networks work inside

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

How-NN-Works

This is a simple script where the user can see how Neural Networks work inside.

In "networks.py" are defined few types of Neural Network.

In "train.py" is defined "train()" method, that will train our choosen Neural Network.

In How_NN_Works the user can play with parameters and see on plots how Neural Networks work inside.

abc1 abc2

Training Neural Network with 1 Hidden Layer and Tanh Activator; The size of Hidden Layer is: 2

Last value of Loss Function: 0.13; Total training time: 0.61 [s]

abc3 abc4

Training Neural Network with 1 Hidden Layer and ReLU Activator; The size of Hidden Layer is: 2

Last value of Loss Function: 0.18; Total training time: 0.39 [s]

About

How Neural Networks work inside


Languages

Language:Python 100.0%