Skip to content

This program enables the user to draw a number on the screen, and then the program attempts to guess which number it represents. It utilizes a simple neural network model.

Notifications You must be signed in to change notification settings

masanbasa3k/Number_Guessing_Neural_Network

Repository files navigation

Number Guessing Neural Network

This program is an interactive and user-friendly application that harnesses the power of a simple neural network model to tackle the MNIST Digit data set. Its primary functionality lies in the ability to allow users to draw a number directly on the screen, providing an intuitive and engaging experience. Once the user completes their hand-drawn representation of a digit, the program employs its neural network model to make an educated guess on which number the drawing corresponds to.

Requirements

  • Python 3.6 or below
  • Pygame
  • TKinter
  • Tensroflow
  • Keras
  • Numpy
  • Matplotlib

numbers

Model loss and accuracy

Model loss is 0.02%

Model accuracy is %99

About

This program enables the user to draw a number on the screen, and then the program attempts to guess which number it represents. It utilizes a simple neural network model.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published