This project was part of the AI-Programming with Python Nanodregree by Udacity.
This project aims to evaluate and compare different image classification algorithms using Python. The primary objectives are to determine the most effective algorithm for classifying images as "dogs" or "not dogs," assess their accuracy in identifying dog breeds, and consider the computational time required for each algorithm. Determine which CNN model architecture (ResNet, AlexNet, or VGG) gives best results
- Time your program Use Time Module to compute program runtime
- Get program Inputs from the user Use command line arguments to get user inputs 3.Create Pet Images Labels Use the pet images filenames to create labels Store the pet image labels in a data structure (e.g. dictionary)
- Create Classifier Labels and Compare Labels Use the Classifier function to classify the images and create the classifier labels Compare * Classifier Labels to Pet Image Labels Store Pet Labels, Classifier Labels, and their comparison in a complex data structure (e.g. dictionary of lists)
- Classifying Labels as "Dogs" or "Not Dogs" Classify all Labels as "Dogs" or "Not Dogs" using dognames.txt file Store new classifications in the complex data structure (e.g. dictionary of lists)
- Calculate the Results Use Labels and their classifications to determine how well the algorithm worked on classifying images
- Print the Results