Skip to content

This project was my first project at the Udacity's Nanodegree program "AI Programming with Python)

License

Notifications You must be signed in to change notification settings

KaiKrah79/Project.identify_dog_breads_with_pretrained_image_classifier

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Project 'Use a pretrained image classifier to identify dog breeds'

This project was my first project at the Udacity's Nanodegree program "AI Programming with Python"

Information to the project

The story behind is a dog show where i have volunteered organizing the event. every participant has to register and submit some information about their dogs. And some people are planning to register pets, that aren't dogs. So i had to analyse which participants are dogs. My tasks were to check which image classifier is the best to classify the registration images, how well the best classifier works and to check how long each algorithm takes to solve the classification problem. Therefore i explored the three archtiectures AlexNet, VGG and ResNet.

Starting Point

Starting file is the "check_images.py".

  • To run this project, it is required you've already installed Python 3 on your operating system.
  • Download the workspace files and keep all in one folder. There are two folders of interest in which we're classifying, the pet_images folder and the uploaded_images folder.
  • Use terminal/command prompt to run the project.

Acknowledgements

This project was my first project at the Udacity Nanodegree program "AI Programming with Python". I am super thankful for this experience.

License

This project is open source and available under the [Udacity License]

About

This project was my first project at the Udacity's Nanodegree program "AI Programming with Python)

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published