All projects
Book On Rails
Tech Stack: Spring Boot, Thymeleaf, MySQL, Design Principles and Pattern
- Led the development of a train ticket booking system using Spring Boot and Thymeleaf, managing a team of 8 developers.
- Implemented design patterns and principles for scalability and performance optimization.
- Integrated backend services with frontend components, ensuring a seamless user experience.
Campus Compass
Tech Stack: Django, MySQL
- Developed a comprehensive online platform for student engagement using Django and MySQL.
- Collaborated with team members to handle backend development and documentation.
- Incorporated software engineering techniques for code quality and future scalability.
Parkinson Prediction Model
Tech Stack: TensorFlow, Scikit-Learn, Python
- Created a binary classification model with 98.31% accuracy for early detection of Parkinson's disease.
- Optimized model performance through meticulous data preprocessing and feature engineering.
- Applied TensorFlow and Scikit-Learn for advanced machine learning analysis and validation.
Stock Price Prediction
Tech Stack: Python, Pandas, Matplotlib, Statsmodels
- Conducted exploratory data analysis to identify underlying patterns in time series data.
- Utilized ARIMA and SARIMAX models to analyze stock prices and forecast trends.
- Developed a generalized prediction model with a SMAPE value of 0.033, aiding decision-making in financial markets.
Dog vs Cat Classification Model
Tech Stack: Python, TensorFlow, MobileNetV2, Tkinter
- Developed a binary classification model using MobileNetV2 architecture to distinguish between images of dogs and cats.
- Achieved an impressive accuracy of 98.75% on the classification task.
- Implemented a user-friendly GUI using tkinter for seamless interaction with the model.
MNIST Digit Recognition Model
Tech Stack: Python, TensorFlow, OpenCV, NumPy
- Developed and compared various deep learning models to classify digits from the MNIST dataset.
- Implemented and evaluated five different architectures and incorporated techniques like batch normalization, dropout, pooling, and early stopping.
- Achieved 99.46% accuracy by integrating Encoder-Decoder Architechture integrating CNN, early stopping, batch normalization, pooling, and dropout, outperforming traditional FCNN and RNN models
Car Price Prediction
Tech Stack: Python, Scikit-Learn, Matplotlib, Pandas, NumPy
- Developed a model to predict car prices using Lasso Regression, achieving an R-squared error of 0.87.
- Utilized libraries including matplotlib, scikit-learn, pandas, and numpy for data visualization, preprocessing, and model building.
- Applied regression analysis to accurately predict car prices, aiding in decision-making for buyers and sellers.