Dog-Breed-Classification

We are conducting dog breed classification using a dataset from Kaggle, which includes 120 dog breeds. This task presents significant challenges due to the limited number of samples available for each breed. To address this, we have developed a model using the TensorFlow framework and have utilized two highly regarded pre-trained image classifiers: ResNet-101 and Inception V3. We have also experimented with Vision Transformer. Our project notebook is powered by a Nvidia A100 GPU to enhance processing speed. This project is particularly valuable not only for its ability to deepen our understanding of leveraging and refining image classification models using pre-trained networks but also for its practical implications in areas such as dog license registration, animal shelter adoption processes, and educational programs about dog breeds.