This repo contains implementation of deep learning-based steel surface defect segmentation models. Extensive experiments on several deep learning frameworks have been presented with various performance analysis and comparison.

This research enhances early disease diagnosis by analyzing retinal blood vessels in fundus images using deep learning. It employs eight pre-trained CNN models and Explainable AI techniques.

Food image recognition using deep learning in climate change issue

End-to-end deep learning pipeline to classify images of dogs and humans to determine what breed the human or dog most resembles

Solutions to the Advanced CNN course by the Lazy Programmer and all CNN Models I've worked on

A collection of CNN models are trained on Cloud TPU by using PyTorch/XLA

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

CNN - object detection, classification & various model tuning for prediction optimization

Appolo's Ear is an ensemble of three music genre classification neural networks, a nginx server, and a demonstrative react frontend. Its backend is glued together with Docker and Docker compose.

Pretrained ConvNets for pytorch: NASNet, ResNeXt, ResNet, InceptionV4, InceptionResnetV2, Xception, DPN, etc.

This is an image classification project which was carried out during "Applied Machine Learning and Data Science" in Indian Institute of Technology, Kanpur

A wrapper library over several open source cnn models for tensorflow -2.x.

A collection of deep learning models for rice classification using PyTorch, torchvision, and scikit-learn, with performance metrics evaluated using Slurm jobs.

Implementations of popular Convolutional Neural Networks (CNNs) for image classification and learning. Includes AlexNet, VGG, ResNet, and more — ideal for study and experimentation.

Final, award-winning B.Sc. Data Science project (TAU, 2025) with my teammate. Developed Deep Learning models (CNNs, GRUs) for vowel decoding from single-neuron brain signals of epilepsy patients. Done in collaboration with Dr. Ariel Tankus (Ichilov) & Prof. Neta Rabin (TAU). Results show feasibility and per-patient variability in decoding accuracy.