Package: petrographer
Title: Petrographic Thin Section Analysis with Deep Learning
Version: 0.0.0.9000
Authors@R: c(
        person("Nicolas", "Gauthier", , "nicolas.gauthier@ufl.edu", role = c("aut", "cre"),
           comment = c(ORCID = "0000-0002-2225-5827")),
        person("Ashley", "Rutkoski", , "arutkoski@ufl.edu", role = c("aut")))
Description: Automated petrographic analysis of thin section images
        using deep learning models. Provides tools for training RF-DETR
        detection and segmentation models, running predictions with
        SAHI (Slicing Aided Hyper Inference), calculating morphological
        properties, and analyzing results. Supports both local and HPC
        training workflows with seamless R-Python integration via
        reticulate.
License: MIT + file LICENSE
Encoding: UTF-8
Roxygen: list(markdown = TRUE)
RoxygenNote: 7.3.3
Depends: R (>= 4.1)
Imports: cli, dplyr, fs, glue, jsonlite, pins, processx, purrr, readr,
        reticulate, stringr, tibble, tidyr,
Suggests: grid, ggplot2, hipergator, here, htmltools, knitr, patchwork,
        png, quarto, reactable, scico, testthat (>= 3.0.0), tools,
        withr
VignetteBuilder: quarto
Config/testthat/edition: 3
Remotes: flmnh-ai/hipergator
URL: https://flmnh-ai.github.io/petrographer/
BugReports: https://github.com/flmnh-ai/petrographer/issues
SystemRequirements: Python (>= 3.8), PyTorch, rfdetr, SAHI,
        supervision, opencv-python, scikit-image, pycocotools
Config/pak/sysreqs: cmake make libicu-dev libpng-dev libuv1-dev
        libopencv-dev libssl-dev python3 libx11-dev
Repository: https://flmnh-ai.r-universe.dev
Date/Publication: 2026-04-21 17:27:24 UTC
RemoteUrl: https://github.com/flmnh-ai/petrographer
RemoteRef: HEAD
RemoteSha: cdaf5711967f2c2016b52ff728a364e441166f8e
NeedsCompilation: no
Packaged: 2026-10-18 07:13:40 UTC; root
Author: Nicolas Gauthier [aut, cre] (ORCID:
    <https://orcid.org/0000-0002-2225-5827>),
  Ashley Rutkoski [aut]
Maintainer: Nicolas Gauthier <nicolas.gauthier@ufl.edu>