Data and Python scripts for 'Obsidian Sourcing in Mesoamerica and the Isthmo-Colombian Area using Image-Based Machine Learning' (Lyons et al. 2026)

This archive contains all code, data, and computational environment files required to reproduce the analyses presented in the manuscript: Obsidian Sourcing in Mesoamerica and the Isthmo-Colombian Area using Image-Based Machine LearningMike Lyons, Jeannine Langmann, Markus ReindelJournal of Archaeological Science, 2026https://doi.org/10.1016/j.jas.2026.106486 Included materials: full input image dataset used for model training, validation, and evaluation (obsidian_images/) main analysis notebook (Lyons_et_al_2025_obsidian_sourcing_cnn.ipynb) and supporting scripts requirements.txt listing exact Python dependency pins Docker environment definitions (Dockerfile, Dockerfile.gpu, docker-compose.yml, and .dockerignore) enabling reconstruction of the original computational environment Readme.md providing instructions for use and additional information This deposit enables independent reproduction of figures, metrics, and results presented in the associated manuscript.