Skip to content

Latest commit

 

History

History
48 lines (31 loc) · 1020 Bytes

README.md

File metadata and controls

48 lines (31 loc) · 1020 Bytes

Dogs Breeds Classifier Project

The project uses a pre-trained classifier to identify dogs breeds, the project support three main models VGG, AlexNet and ReseNet and gives a comparative statistics about their performance.

Prerequisites

To setup the project you will need to install these packages

python3
torch
torchvision
PIL

Usage

You can run the project from the command line using the following:

ReseNet:

python check_images.py --dir uploaded_images/ --arch resnet  --dogfile dognames.txt

AlexNet:

python check_images.py --dir uploaded_images/ --arch alexnet --dogfile dognames.txt

VGG:

python check_images.py --dir uploaded_images/ --arch vgg  --dogfile dognames.txt

in order to the run the three above command and save their results into a files use the script run_models_batch.sh:

bash ./run_models_batch.sh

Results

After executing the project you should get the below results:

Project results