Welcome to my Portfolio Repository! This repository showcases my expertise in Machine Learning, Artificial Intelligence, and Data Science through a collection of curated projects and a professionally designed portfolio website. It is designed to provide an overview of my skills, experiences, and contributions to real-world AI solutions.
Portfolio/
├── PortfolioWebsite/ # Hugo-based portfolio website
│ ├── archetypes/ # Default archetypes for Hugo content
│ ├── content/ # Markdown files for pages like landing page, resume, and projects
│ ├── static/ # Static assets (images, CSS, JS, etc.)
│ ├── themes/ # Custom Hugo themes (e.g., Adritian)
│ ├── config.toml # Hugo configuration file
│ ├── scripts/ # Scripts for deployment or build automation
│ ├── README.md # Documentation for the portfolio website
│ └── public/ # Generated static site output after running Hugo
│
├── projects/ # Folder containing project implementations
│ ├── RetailDemandForecaster/ # Time-series forecasting project
│ │ ├── notebooks/ # Jupyter notebooks for experiments
│ │ ├── src/ # Python scripts and model code
│ │ ├── data/ # Datasets or links to external datasets
│ │ └── README.md # Documentation for this project
│ ├── PredictiveMaintenanceSystem/ # Predictive maintenance solution
│ │ ├── notebooks/
│ │ ├── src/
│ │ ├── data/
│ │ └── README.md
│ └── EnterpriseKnowledgeSummarizer/ # Document summarization AI tool
│ ├── notebooks/
│ ├── src/
│ ├── data/
│ └── README.md
│
├── README.md # Global repository documentation
├── LICENSE # License for this repository
└── .gitignore # To ignore unnecessary files
The PortfolioWebsite is a Hugo-based static site that presents my work professionally and interactively. It includes:
- Landing Page: An overview of my professional journey and expertise.
- Resume Page: Detailed work experience, education, and technical skills.
- Projects Page: Highlights of my selected AI and data science projects.
Once deployed, my portfolio will be accessible at: [Your Portfolio URL] (Update with the actual link).
- Folder:
projects/RetailDemandForecaster/
- Objective: Predict retail demand using time-series models to improve inventory management.
- Key Features:
- Walmart Store Sales dataset
- Streamlit-based dashboard for interactive analysis
- Deployment-ready pre-trained models
- Tech Stack: Python, Scikit-learn, Streamlit, Plotly
- View Project README: Link to README
- Folder:
projects/PredictiveMaintenanceSystem/
- Objective: Predict equipment failures to minimize downtime and optimize maintenance.
- Key Features:
- NASA Turbofan Engine Degradation dataset
- REST API for real-time predictions
- Feature importance analysis for interpretability
- Tech Stack: Python, Flask/FastAPI, PyTorch, Seaborn
- View Project README: Link to README
- Folder:
projects/EnterpriseKnowledgeSummarizer/
- Objective: Summarize large documents for faster knowledge retrieval in enterprise settings.
- Key Features:
- Fine-tuned Hugging Face transformer models
- Interactive Streamlit/Gradio app for summarization
- Support for large datasets like Wikipedia dumps
- Tech Stack: Python, Hugging Face, Streamlit/Gradio, PyTorch
- View Project README: Link to README
To clone this repository, use the following commands:
git clone https://github.com/Sharma-Pranav/Portfolio.git
cd Portfolio
Create and activate a Conda environment for this portfolio:
conda create --name portfolio python=3.10 -y
conda activate portfolio
Each project folder contains a requirements.txt file. Navigate to the project directory and install dependencies using:
pip install -r requirements.txt
- Python
- Markdown
- Machine Learning: PyTorch, Scikit-learn, Hugging Face Transformers
- Data Visualization: Plotly, Seaborn
- Web Frameworks: Streamlit, Flask, FastAPI, Gradio
- Hugo
- Git
- Docker
- Conda
Contributions are welcome! If you'd like to enhance this portfolio or collaborate on the projects:
- Fork this repository.
- Make your changes in a new branch.
- Submit a pull request with a clear explanation of your changes.
This repository is licensed under the MIT License. See the LICENSE file for more details.
If you'd like to get in touch, feel free to connect:
- Email: topranav@outlook.com
- LinkedIn: https://www.linkedin.com/in/topranav/
- GitHub: Sharma-Pranav