This Streamlit application allows users to build, train, validate, and use a Random Forest model to predict one spectral reflectance band based on others from a given dataset. The interface is user-friendly, providing easy selections for input and output variables, adjusting test data rate, and visualizing model performance and feature importance.
Ensure you have the following libraries installed:
streamlit
pandas
matplotlib
seaborn
scikit-learn
numpy
Install the required libraries using:
pip install streamlit pandas matplotlib seaborn scikit-learn numpy
- Run the Streamlit application:
streamlit run streamlit_app.py
- Variable Selection: Select one output variable (target) and at least one input variable (features).
- Test Data Rate: Adjust the rate of test data using a slider.
- Title and Description: Provides an overview of the application.
- Dataset Information: Details about the dataset used.
- Model Information: Displays selected output variable, input variables, training data shape, and test data shape.
- Scatter Plot: Scatter plot comparing predicted vs. observed values for the test data, showing R-squared and RMSE values.
- Feature Importance Chart: Bar chart showing the importance of each feature in the model.
- Model Prediction:
- User input fields for entering values of input variables.
- Displays the predicted value of the output variable.
- Bar chart showing the predicted output value.
This project is licensed under the MIT License. See the LICENSE file for details.