This project is a deep learning-based computer vision system designed to classify dog images into their respective breeds. The model was developed using TensorFlow and transfer learning techniques to achieve high classification accuracy on unseen images. It demonstrates practical applications of image preprocessing, model training, evaluation, and prediction in computer vision.
Model url: "https://tfhub.dev/google/imagenet/mobilenet_v2_130_224/feature_vector/4" Pretrained architecture used: MobileNetV2
This was obtained from kaggle which includes test data, train data and labels
Turning images to tensors, Image resizing, batching, norminalization, train/validation split
Python, tensorflow/keras, numpy, matplotlib, scikit-learn, pandas, jupyter notebook, google colab