Skip to content

README.md

Latest commit

 

History

History
116 lines (80 loc) · 4.92 KB

README.md

File metadata and controls

116 lines (80 loc) · 4.92 KB

commits forks stars issues MIT license

Face and Digit Classification

A implementation of Naive Bayes algorithm and perceptron algorithm
Explore the docs »

View Demo · Report Bug · Request Feature

Table of Contents
  1. About The Project
  2. Usage
  3. License
  4. Contact

About The Project

This is a course project of Rutgers CS440 Intro to Artificial Intelligence.

Built with

  • Language: Python

Dataset

The data folder contains two different types of images: digits from 0-9 and human faces.

Each type of data is splited to three parts:

  • Training dataset
  • Validation dataset
  • Test dataset

Size of a digit image is 28 x 28.

Size of a face image is 60 x 70.

Test result

Average training time

  • Digit

    Digit Average Training Time

  • Face

    Face Average Training Time

Accuracy and standard deviation

Digit using Naive Bayes

Accuracy Digit(Naive Bayes)

Digit using Perceptron

Accuracy Digit(Perceptron)

Face using Naive Bayes

Accuracy Face(Naive Bayes)

Face using Perceptron

Accuracy Face(Perceptron)

Weight visualization examples

  • Digit

    Weight for digit 1 Weight for Digit 3

  • Face

    Weight for face

Usage

Classify face and digit by using naive bayes algorithm and perceptron algorithm.

License

Distributed under the MIT License.

See LICENSE for more information.

Contact

Feiyu Zheng - [email protected]

Project Link: https://github.com/ChaserZ98/Face-and-Digit-Classification

(back to top)