Deep learning from videography as a tool for measuring infection in poultry (under review)

Code and data for "Deep learning from videography as a tool for measuring infection in poultry". Release of video and physiological data is pending approval from the Department of Veterinary and Animal Sciences at the University of Copenhagen

Setup

Python

Version: 3.10.10

Python dependencies

The dependencies for downstream analyses are listed in env.yml

You can install a virtual environment using conda by running:

conda env create -f env.yml

DeepLabCut data and training

Available soon

Feature extraction from DeepLabCut predictions

python -m dlc4ecoli.dlc.extract --path /path/to/data

Reproducing the figures

You can reproduce most figures by running the plots.ipynb notebook.

The other brms figures are created from the R script (see below).

R

Version: 4.4.0

R dependencies

install.packages(brms, envalysis, ggdist, ggplot2)

Mixed-effects modelling

Simply run the analysis.R script after setting the work directory to this repository

setwd("path/to/dlc4ecoli")