Skip to content

talmolab/Berrigan_et_al_sleap-roots

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Berrigan_et_al_sleap-roots

This repository contains the code for figures and analyses presented in Berrigan et al. Fast and efficient root phenotyping via pose estimation.

Environment Setup

Before running the analyses, set up the conda environment using the provided env.yaml file. This ensures that you have all the necessary dependencies installed. Execute the following commands in your terminal:

conda env create -f env.yaml
conda activate berrigan_et_al

Figure 5

SLEAP models accurately located root landmarks in 4 species and 3 classes of roots.

Notebook: figure-5/figure_5_error_summary.ipynb

Data:

  • figure-5/average_root_lengths.csv
  • figure-5/plantwise_testsets_metrics.csv

Outputs:

  • Plots showing median localization error as a percentage of average root length.
  • CSV files containing numerical metrics for each model.

Figure 8

Efficient models require fewer labels for accurate predictions and facilitate diverse training sets.

Notebook: figure-8\figure_8_sample_efficiency.ipynb

Data: -figure-8\sample_efficiency_summary.csv

Outputs:

  • Plot showing 90% percentile error vs. number of labeled frame for training per dataset.
  • Plot showing number of labeled frames required in training set for model to plateau to within 3 mm of best.
  • CSV file of number of labeled frames and 90% percentile error value for each trained model that reached plateau.

About

Code for figures and analyses in Berrigan et al.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published