oxford-iiit-pet

This project uses ResNet50 to classify 37 pet breeds from the Oxford-IIIT Pet dataset, achieving ~92-93% accuracy with transfer learning, data augmentation, and fine-tuning. Includes ROC curve analysis and single-image prediction. Upload your pet pic to test! 🐾📸 Code runs on Colab with GPU support. 🚀README.md

Multi-task pet breed classification and segmentation with ResNet50, MobileNetV3, U-Net, Grad-CAM and Streamlit

Semantic segmentation of pet images with a PyTorch U-Net CNN, featuring Grad-CAM heatmaps and TensorBoard logging

U-Net semantic segmentation study on the Oxford-IIIT Pet Dataset with MobileNetV2 encoder, qualitative mask predictions, and reproducible notebook artifacts.

Fine-grained classification of 37 cat and dog breeds (EfficientNet-B0 transfer learning, 90.9% test top-1) with Grad-CAM explanations, web app and Docker

From-scratch PyTorch U-Net for Oxford-IIIT Pet semantic segmentation with augmentation, Dice+CE loss, AMP, IoU metrics, qualitative grids, curves, and tests.

A hands-on PyTorch semantic segmentation lab for beginners, with runnable training, model and dataset switching, custom extensions, and bilingual guides

ResNet vs ResNeXt on Oxford-IIIT Pet — both implemented from scratch in PyTorch, parameter-matched (11.2M vs 10.6M), trained from random init under identical conditions. Includes misclassification overlap analysis and hardware benchmarks.