Tasimudzwa-V-Makoni/Agri-Bot-Price-Predictor

Developed a predictive model using Python and TensorFlow to forecast agricultural market trends for small-scale farmers in Zimbabwe. The project involves data preprocessing of historical price sets and training a regression model to optimize harvest timing. Demonstrates proficiency in Machine Learning, data analysis, and documentation. --Project Overview Agri-Bot is a full-stack agricultural solution designed to empower small-scale farmers in Zimbabwe. It combines an autonomous hardware rover for real-time soil monitoring with a Machine Learning-powered web application to predict market crop prices. By bridging the gap between IoT (Internet of Things) and Data Science, this project aims to reduce crop failure caused by unpredictable environmental factors and minimise financial loss due to market volatility. --Key Features Autonomous Soil Sensing: Mobile rover equipped with pH and Soil Temperature sensors for field diagnostics. Predictive Analytics: A Decision Tree Regressor model that forecasts crop prices with 93-95% accuracy. Farmer Dashboard: A responsive web interface built with Flask for data visualisation. Market Intelligence: Real-time tracking of "Top Gainer" and "Top Loser" crops to guide planting decisions. --Tech Stack Software Languages: Python (Backend & ML), C++ (Hardware Logic) ML Libraries: Scikit-learn, Pandas, NumPy Web Framework: Flask, Jinja2 Frontend: HTML5, CSS3, Chart.js Hardware Microcontroller: Arduino Uno/Mega Communication: I2C Protocol, Serial (PySerial) Sensors: DS18B20 (Temperature), Analogue pH Probe --Machine Learning Implementation The core of the predictive engine uses a Decision Tree Regressor. The model was trained on historical agricultural datasets, focusing on: 1. Data cleaning and preprocessing via Pandas. 2. Feature scaling and selection. 3. Model validation to ensure reliability in real-world Zimbabwean market conditions. --Project Structure Plaintext ├── app.py # Flask Application logic ├── model.py # Machine Learning model training ├── static/ # CSS and JavaScript (Chart.js) ├── templates/ # HTML Dashboards (Jinja2) ├── hardware/ # Arduino (.ino) sketches └── data/ # Historical crop price datasets --Impact & Vision This project addresses UN Sustainable Development Goal 2: Zero …