Back to Projects
AI / Machine LearningML FoundationsApr 2026 – Jul 2026
CIFAR-10 Image Classification: PyTorch Benchmark
An end-to-end image-classification project comparing a custom CNN, MobileNetV2 and ResNet-18 under controlled training and transfer-learning conditions. ResNet-18 achieved 87.48% test accuracy. The project includes Grad-CAM interpretability, INT8 quantisation, CLI inference and a live Gradio deployment.
Project Overview
What I Built
An end-to-end image-classification project comparing a custom CNN, MobileNetV2 and ResNet-18 under controlled training and transfer-learning conditions. ResNet-18 achieved 87.48% test accuracy using transfer learning, outperforming the custom CNN baseline. The project includes Grad-CAM visual interpretability, INT8 quantisation, command-line inference tools and a live Gradio demo deployed on Hugging Face Spaces.
Engineering
Tech Stack
Outcome & Current Status
Highlights
- ▸Designed and implemented an end-to-end deep learning pipeline comparing a custom CNN, MobileNetV2 and ResNet-18 under controlled conditions; ResNet-18 achieved 87.48% test accuracy using transfer learning, outperforming the custom CNN baseline
- ▸Extended the project with Grad-CAM interpretability, INT8 quantisation, command-line inference tools and a Gradio demo deployed on Hugging Face Spaces