Skip to content

In this project, I used Deep Learning with Artificial Neural Networks and also K-nearest Neighbors for image recognition.

Notifications You must be signed in to change notification settings

davidrodriguessp/deep-learning-image-recognition

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 

Repository files navigation

Using Deep Learning for Image Recognition: Building Deep Neural Network Models to Predict Handwritten Digits

neuron image

Photo by Nastya Dulhiier on Unsplash

Author: David Rodrigues | https://github.com/davidrodriguessp | https://www.linkedin.com/in/davidrodrigues/ | davidrodriguessp@hotmail.com

In this project, we created different models to predict handwritten digits. We used a dataset available in the Scikit-learn Python library called load_digits(). The dataset has 1,797 images. Initially, we created a model using the K-nearest neighbors algorithm. Then, we built different models using Artificial Neural Networks.

The conclusion was that the different models reached similar scores, but K-nearest Neighbors (KNN) with k=1 had a slightly better performance, reaching 99% accuracy.

Notebook: Deep Learning for Image Recognition.ipynb

About

In this project, I used Deep Learning with Artificial Neural Networks and also K-nearest Neighbors for image recognition.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published