/Deep-Learning

This repository serves as a showcase of multiple deep learning projects across domains like healthcare, customer analytics, and computer vision. Each project is implemented in Python using TensorFlow/Keras and is structured in Jupyter Notebooks for reproducibility and learning.

Primary LanguageJupyter Notebook

Deep Learning Projects 🚀

A collection of deep learning projects implemented in TensorFlow and PyTorch, covering computer vision, healthcare applications, customer analytics, and classic datasets. Each notebook is self-contained with code, experiments, and results.

📂 Projects

  1. Customer Churn Prediction

File: Customer_churn_model.ipynb

Description: Built a deep learning model to predict customer churn based on historical behavioral and transactional data. Includes feature preprocessing, model training, and performance evaluation.

Key Concepts: Classification, Tabular Data, Customer Analytics

  1. ECG-based Heart Disease Detection

File: ECG_Based_Heart_Disease_Detection.ipynb

Description: Designed a CNN-based model to analyze ECG signals and detect potential heart disease. Demonstrates how deep learning can support healthcare diagnostics.

Key Concepts: Healthcare AI, CNNs, Time-Series Data

  1. Skin Cancer Classification using MobileNetV2 & TensorFlow Lite

File: Skin_Cancer_Classification_using_MobileNetV2_&_TensorFlow_Lite.ipynb

Description: Trained a lightweight MobileNetV2 model for skin lesion classification and converted it to TensorFlow Lite for deployment on mobile/edge devices.

Key Concepts: Transfer Learning, Mobile AI, Edge Deployment

  1. Hand Gesture Recognition

File: cnn-based-hand-gesture-recognition-model (1).ipynb

Description: Implemented a CNN to recognize hand gestures from images, enabling applications in HCI (Human-Computer Interaction).

Key Concepts: CNNs, Image Classification, Gesture Recognition

  1. MNIST Digit Classification

File: mnist_classification.ipynb

Description: A foundational CNN model trained on the MNIST dataset to classify handwritten digits. Serves as a baseline project for experimenting with deep learning techniques.

Key Concepts: CNNs, Image Classification, Benchmark Dataset