Briankim254/Model-to-Predict-Crop-Yields-in-Sub-Saharan-Africa

This repository contains a ML model designed to predict crop yields in Sub-Saharan Africa. By leveraging historical data, weather patterns, and various agronomic factors, the model aims to provide accurate yield forecasts for farmers and policymakers in the region. # Machine Learning Model to Predict Crop Yields in Sub-Saharan Africa Link to the live application - yieldpredictor.streamlit.app ### Devs 1. Brian Kimutai 2. Antoinette Akinyi 3. Peterson Kariuki 4. Francis Mayieka 5. Marcellus Kimutai 6. Ryan Matibo Business Understanding Who is the client? - The target user of this machine learning model is a farmer. Needs of the client 1. To be able to make informed decisions regarding crop planting and harvesting. 2. To improve on food security. 3. To reduce on food waste. 4. To increase income for the farmers. 5. To be able to improve on farming techniques. 6. To be able to choose on better irrigation systems. 7. To be able to make informed decisions on the type of plant based on the soil quality. ## Client engagement process 1. Machine learning team initially meet with the farmers. 2. Data collection to understand the problem statement and the project scope and objectives. 3. Analyzing and evaluating farmers information so as to develop a personalized strategy. 4. Implementing solutions and monitoring appropriate areas where the maize crop can yield. ## Project Objectives * To develop a machine learning model that predicts crop yields in Sub-Saharan Africa based on various factors such as: - Weather patterns - Soil quality - Irrigation systems - Farming techniques Data Acquisition ## Source Systems -The data for this project was obtained and analysed in GYGA. The dataset consists of all the 9 Sub-Saharan countries with three sheets; Country, climate zone and station. ## Data Acquisition process - Data collection: This is the first phase of data acquisition process and the file is in an excel format. - Data Extarction: After uploading the data in our working environment and exported it into dataframes. - Data cleaning: It involves checking of any missing values, duplicates and inconsistencies in our data. - Data preparation: Transforming raw data i.e to remove outliers or anomalies to make accurate predic …