manjunathsirur1994/NDVI-forecasting-Kenya

This project forecasts NDVI values using MODIS13Q1 data from Google Earth Engine. It employs Python and Jupyter notebooks to automate data download, preprocessing, and modeling with ML/DL/statistical techniques. It helps a tractor/equipment lending company gauge vegetation and optimize business decisions. **NDVI Forecasting Project for Agricultural Equipment Lending.** This project forecasts NDVI (Normalized Difference Vegetation Index) values to assist a tractor and equipment lending company in gauging vegetation health and expanding their business based on vegetation cover. Utilizing MODIS13Q1 data from Google Earth Engine, the project is implemented in Python with Jupyter notebooks and includes the following components: **Data Acquisition:** Utilizes Google Earth Engine to download MODIS13Q1 raster data. **Data Processing:** Converts raster data into readable dataframes for further analysis. **Modeling:** Employs various machine learning, deep learning, and statistical models to forecast NDVI values. **Pipeline:** An end-to-end pipeline that automates data download, preprocessing, model training, and forecasting. By providing accurate NDVI predictions, this project enables the company to make informed decisions about lending equipment based on the current and forecasted vegetation cover, optimizing resource allocation and business expansion.