Skip to content

Latest commit

 

History

History
36 lines (26 loc) · 1.03 KB

README.md

File metadata and controls

36 lines (26 loc) · 1.03 KB

Binary-Perceptron

Introduction This interactive tool that lets you determine whether a given a color is bright or dim. The input is a rgb value and the output is determined by a single layer perceptron. This is esentially a linear classifier with 2 classes (Bright or Dim).

Options: -You can add you're own training data/set a custom model or use the pretrained model, and evaluate a colors to see if they are bright/dim using the built-in color picker. -A confusion matrix is also included in the app that tells you "how well" ur model does.

Testing Data Format: If you are applying your own testing data then provide a .txt file with with the RGB values on each line. Note: Invalid data formats will result in a failure prompt. Example: 101 20 11 251 01 255 11 14 1 44 11 11

Output: You will get a .txt file with the the prediction result appended to the end of each line for example 101 20 11 1 251 01 255 -1 11 14 1 -1 44 11 11 -1

How the model works: :todo

How the interactive tool works: :todo

API Calls: :todo

Libraries Used: