A continuously evolving collection of end-to-end Machine Learning and Deep Learning case studies built on public datasets (Kaggle & beyond).
This repository serves as an experimentation hub covering data exploration, EDA, feature engineering, modeling, and evaluation across multiple domains.
⚡ The projects listed below are representative examples — the repository is regularly updated with new studies and experiments.
- Heart Disease Prediction
- Tesla Stock Price Forecasting
- Box Office Revenue Prediction
Techniques:
- Scikit-Learn
- XGBoost
- LightGBM
- CatBoost
- Decision models like RandomForest
- Cassava Leaf Disease Classification
- Global Wheat Detection
- EfficientNet Fine-tuning
- YOLOv8 (Keras implementation)
Techniques:
- TensorFlow / Keras
- PyTorch
- Transfer Learning
- CNN Architectures
- Object Detection
- SMS Spam Detection
- Disaster Tweet Classification
Techniques:
- Text Preprocessing
- Embeddings
- Transformers
- Classical ML & Deep Learning models
- End-to-end ML workflows
- Data Cleaning & EDA
- Feature Engineering
- Model Development & Evaluation
- Visualization & Reporting
- Web Scraping for Data Collection
- Python
- TensorFlow / Keras
- PyTorch
- Scikit-Learn
- XGBoost / LightGBM / CatBoost
- Pandas / NumPy
- Matplotlib / Seaborn
This repository acts as a structured ML experimentation portfolio, showcasing applied machine learning across tabular data, computer vision, and NLP tasks.
Some case studies evolve into fully deployable applications in separate dedicated repositories.