trasfer-learning

Implementation of the model-agnostic meta-learning framework on CWRU bearing fault dataset to address cross-domain few-shot fault diagnosis problem.

ODIR-2019: Ocular Disease Intelligent Recognition is a project leveraging state-of-the-art deep learning architectures to analyze and classify ocular diseases based on medical imaging data. This repository implements advanced machine learning techniques and modern neural network architectures to push the boundaries of intelligent recognition

Deep learning for brain tumor MRI: detection (99.4%) and 4-class typing (94.75%) via EfficientNet transfer learning + Grad-CAM.

A machine learning tool built with TensorFlow and the VGG16 model. It classifies waste items from images, assisting in efficient recycling. Users upload waste images, and the system identifies the waste type.

Thesis work, University of Groningen : Lifelong 3D Object Recognition and Grasp Synthesis using Dual Memory Recurrent Self-Organization Networks

Göz görüntülerinde diyabetik retinopati belirtilerinin tespiti

Incremental Learning using MobileNetV2 of Logo Dataset

Classifying various food using transfer learning.

A deep learning system for medical image diagnosis that detects diseases from chest X-ray images using CNN and transfer learning. The project includes a full training pipeline, model inference API, and a Streamlit dashboard for real-time predictions.

A deep learning technique that leverages transfer learning to classify images into different food categories.

An image classifier developed for ImageNet dataset

Production-grade deep learning pipeline for 4-class kidney CT scan (cyst/normal/stone/tumor) classification (VGG16 transfer learning, 97.95% val accuracy). The project includes config-driven training, MLflow experiment tracking, and a FastAPI with a Gradio UI, and is Dockerized with automated CI/CD deployment to AWS ECR/EC2 using GitHub Actions.

An end-to-end CNN Image Classification Model which identifies the food in your image

Deep learning-based Indian food image classifier built with PyTorch and Streamlit, capable of classifying images into five Indian food categories.

This Repository consists of all Deep Learning related projects

EN3150 – Pattern Recognition | University of Moratuwa A PyTorch-based Convolutional Neural Network (CNN) project for handwritten digit recognition using the MNIST dataset. Includes optimizer comparison (Adam, SGD, SGD+Momentum), momentum analysis, and transfer learning with pretrained ResNet18 and VGG16 models.