# Predictive Irrigation Models

This repository contains data preprocessing and analytics pipelines for the distribution of Predictive Irrigation Models, integrating data from field sensors, weather, soil, crop, and remote sensing sources.

## Project Structure
```
predictive_irrigation_models/
├── config/
│ ├── params.yml
│ ├── xgcast_params.yml
│ ├── aquacrop_params.yml
│ └── fieldsensor_irrigator_mapping_anonym.yaml.yml
├── pipelines/
│ ├── __init__.py
│ ├── aquacrop_preparation_pipeline.py
│ ├── data_collection_pipeline.py
│ ├── aquacrop_preparation_pipeline.py
│ ├── demo_run.py
│ ├── model_preparation_pipeline.py
│ ├── preprocessing_pipeline.py
│ ├── resample_impute_pipeline.py
│ ├── soilcast_pipeline.py
│ ├── xgcast_pipeline.py
│ └── xgcast_run.py
├── data/
│ ├── 03_primary/
│ ├── 04_model_input/
│ ├── 05_aquacrop_input/
│ ├── 05_xgcast_input/
│ ├── 06_aquacrop_output/
│ └── 06_xgcast_output/
├── tools/
│ └── ...
└── README.md
```

## Features

- **Prefect-based Pipelines:** Modular tasks and flows for data collection, transformation, and saving.
- **Sensor & Weather Data Integration:** Reads and merges raw sensor and weather data for multiple consortia.
- **Soil, Crop, and Remote Sensing Data:** Integrates geospatial and tabular data sources.
- **Automated Testing:** Prefect tasks and flows for validating preprocessing results.
- **Artifact Logging:** Data summary artifacts for monitoring pipeline outputs.

## Getting Started

### Prerequisites

- Python 3.8+
- [Prefect](https://www.prefect.io/)
- uv

Install dependencies:
```uv sync```

Configuration
Edit `config/params.yml` to specify consortia names, data folders, and other parameters.

### Running the Pipeline

From the project root, run:

```python -m main```

### Running Tests
From the project root, run:

```python -m test.test_results```

### Contributing
Feel free to open issues or submit pull requests for improvements or bug fixes.

### License
MIT License

### Contact
For questions or collaboration, please contact the repository owner. ``````