General Context

This repository contains the development of a technical test described here.

In this repository you will find:

• Dockerized environment configuration and usage instructions.
• Set of scripts for data management and database loading.
• Interactive notebooks (Jupyter and R Markdown) for preprocessing and postprocessing spatial data.

---

Part 1. Infrastructure Setup with Docker and Data Loading

Infrastructure

• A Docker Compose based environment was created for project development. This environment consists of:
  • A PostgreSQL (v16) database with installed extensions PostGIS (v3.5) and h3-pg (Uber).
  • A Python 3.9 container with its main spatial dependencies.
• Data loading into the database was performed through a Python script that runs via console.

---

Part 2. Data Analysis: Pre and Postprocessing

Preprocessing

• Outside the Dockerized environment, spatial data exploration and preparation was performed.
• At this stage, QGIS tools were used for visual inspection and R Markdown for data validation and cleaning.

Postprocessing

• Data loaded in the database is utilized.
• Jupyter Notebook is used to analyze datasets, perform advanced spatial analysis, and create visualizations.

Repository File Structure:

pt_agtech_Ramiro_Manzo/
├── docker-compose.yml
├── README.md

├── postgis_h3/
│   ├── Dockerfile

├── python_geo/
│   ├── Dockerfile
│   └── preprocessing/
│       ├── altimetry_kriging.geojson
│       ├── ec_subsurface_kriging.geojson
│       ├── ec_surface_kriging.geojson
│       ├── Grid_final.geojson
│       ├── preprocessing.pdf
│       └── preprocessing.Rmd

├── workspace/
│   ├── notebooks/
│   │   ├── postprocessing.ipynb
│   │   └── final_questions.ipynb
│   │
│   └── scripts/
│       └── load_data.py
│   └── README.md 

└── data/
    ├── Grid_final.geojson
    ├── la_magdalena_L4.geojson
    ├── soy_performance_2019_2021_2023.parquet
    └── veris_data.gpkg