This project is a Dog Breed Classifier built using transfer learning with the VGG19 model. The classifier is capable of identifying different breeds of dogs from images. Custom dataloaders are utilized for efficient data handling and processing
The dataset used for this project consists of images of various dog breeds taken from the Kaggle competition Dog Breed Identification: https://www.kaggle.com/c/dog-breed-identification.
VGG19 was used for the Dog Breed Classifier because its pre-trained deep architecture, with 19 layers, excels in feature extraction and offers strong performance for image classification tasks, making it an ideal choice for leveraging transfer learning to achieve high accuracy in identifying various dog breeds.
The model was trained for 10 epochs reaching a Cross entropy Loss of around 0.4