An interactive machine learning & data visualization web application built with Python and Streamlit that allows users to upload datasets, clean data automatically, train ML models, and explore insights using 2D and 3D visualizations.
This project focuses on Exploratory Data Analysis (EDA) and model understanding, not on prediction generation.
- Upload any CSV dataset
- Automatic data cleaning:
- Handling missing values
- Basic preprocessing
- Select target column
- Choose problem type:
- Classification
- Regression
- Select machine learning model:
- Random Forest
- Decision Tree
- Train model on cleaned data
- Visualize insights using:
- 2D charts (scatter, distribution, comparisons)
- 3D interactive visualizations
- Simple and user-friendly web interface
The goal of this project is to help users:
- Understand their datasets visually
- Explore feature relationships
- Analyze data before deploying ML models
- Learn machine learning workflows without writing code
This makes the app useful for students, beginners, and rapid data exploration.
- Python
- Streamlit
- Pandas
- NumPy
- Scikit-learn
- Matplotlib / Plotly
- Exploratory Data Analysis (EDA)
- Machine Learning education
- Dataset inspection before model deployment
- Rapid ML prototyping
git clone https://github.com/UzairArain2008/data_science_model_app.git
cd data_science_model_app
pip install -r requirements.txt
streamlit run app.py