mentoramytutor-rgb/AGRICULTURE_ANN_DASHBOARD

This project presents the design and implementation of a National Framework for Agriculture based on Artificial Neural Networks (ANN) in Nigeria. The framework addresses critical challenges in agricultural planning, including limited data integration, low adoption of precision farming techniques. # AGRICULTURE_ANN_DASHBOARD This project presents the design and implementation of a National Framework for Agriculture based on Artificial Neural Networks (ANN) in Nigeria. The framework addresses critical challenges in agricultural planning, including limited data integration, low adoption of precision farming techniques. # 🌾 Nigeria National Agriculture ANN Framework A production-standard Artificial Neural Network (ANN) framework for agricultural analysis across all 36 Nigerian states + FCT, implementing Remote Sensing, GIS, and a Streamlit dashboard. --- ## Project Structure ``` nigeria_ann_agriculture/ │ ├── phase1_data_generator.py ← Synthetic data (3,800 records, 37 states) ├── phase2_ann_model.py ← TensorFlow ANN training + all plots ├── phase3_gis_maps.py ← GIS maps (static + interactive Folium) ├── phase4_dashboard.py ← Streamlit dashboard ├── run_all.py ← Master script (runs all phases) ├── requirements.txt ← Python dependencies │ ├── data/ │ ├── raw/ ← nigeria_agriculture_raw.csv (3,800 rows) │ ├── processed/ ← state_aggregates.csv, predictions_full.csv │ └── shapefiles/ ← .shp, .geojson files │ ├── models/ │ ├── ann_final_model.keras ← Trained ANN model │ ├── ann_best_model.keras ← Best checkpoint (early stopping) │ ├── feature_scaler.pkl ← StandardScaler for features │ ├── target_scaler.pkl ← StandardScaler for targets │ ├── le_crop.pkl ← LabelEncoder for crops │ ├── le_season.pkl ← LabelEncoder for season │ ├── le_zone.pkl ← LabelEncoder for geo zone │ ├── training_history.csv ← Epoch-by-epoch loss & MAE │ └── test_metrics.json ← MAE, RMSE, R² per target │ └── outputs/ ├── plots/ │ ├── 01_feature_correlation_heatmap.png │ ├── 02_training_loss_curve.png │ ├── 03_mae_curve.png │ ├── 04_actual_vs_predicted.png │ ├── 05_residuals_distribution.png │ └── 06_feature_importan …