Skip to content

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Latest commit

 

History

10 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 

Repository files navigation

First Neural Network

Google colab link:

Python Files

  • NeuralNetwork.py: This file hold the Neural Network class that is used in other files
  • NNExample.py: This file trains and plots the loss of the NN throughout training process (should be first file to be ran)
  • NNTests.py: This file contains tests for the NN

Packages used:

  • Numpy, matplotlib, and unittest

Class Neural Network:

  • Contains 7 class attributes
  • Contains 8 class methods

Info:

  • Creates a 3-6-2 Neural Network
  • 3 nodes for input layer
  • 6 nodes for hidden layer
  • 2 nodes for output layer
  • Activation function used is a sigmoid
  • Uses back propergation to update the weights
  • Trains the NN 1,000 times
  • Plots the loss over number of training iterations
  • Prints out the shape of both weight matrices
  • Prints out the predicted value for the test case and the rounded value of the test case

Questions:

  • What are the dimensions of weight matrix 1 and weight matrix 2?
    • The dimension of weight matrix 1 is: 3x6
    • The dimension of weight matrix 2 is: 6x2
  • Test [1,1,1]. What are the predicted y values for it?
    • The predicted y values for this sample are: [0.98363427, 0.0157898] (subject to change)
    • The rounded y values for this sample are: [1,0]
  • Could you guess what is the meaning of y1 and y2?
    • My guess what the meaning of y1 and y2 mean is that the outcome is [0,1] when the input is even in binary and [1,0] when the outcome is odd in binary. For example 000, 010, 100 are all [0,1] while the sequences 001, 011, 111 are all [1,0]. This makes me believe that this is determining whether the binary sequence is even or odd.

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages