Skip to content

mrtnzram/Nester

Repository files navigation

Nester

created by Ramiel Martinez as an extension of AVGN (Sainburg, et al., 2020) with the assistance of Ximena Leon, Kate Snyder, and Nicole Creanza

Nester is a software compiled inside JupyterLab with matplotlib interactive widgets which utilizes machine-learning algorithms for semi-automated syntaxing of birdsong syllables. It is the intended step after using Chipper. It mainly utilizes UMAP and HDBSCAN algorithms curated by the avgn package for automatic clustering of bird song syllables using spectrogram features. This software extends the capabilities of the algorithm by introducing human supervision. It features real time dataframe corrections, UMAP and HDBSCAN clustering changes, and visualizations.

Background

Nester utilized the existing code:

AVGN

Animal Vocalization Generative Network (AVGN) is a repository of python tools created by Tim Sainburg et al. (2020) centered around latent models used to generate, visualize, and characterize animal vocalizations. Nester pulls code from AVGN for spectrogram processing, UMAP and HDBSCAN methods, and visualizations.

Chipper output (GZIPs) to JSON

Nester utilizes modified code created by Kate Snyder to process chipper output files (GZIPs) to JSON as preprocessing for Nester's algorithm. It takes chipper syllable onset and offset's for accurate syllable segmentation.

Getting Started

To get Nester started on your device, follow the installation steps below and refer to the Nester Manual.

  1. Open terminal/powershell.
  2. Set path to directory where you want to clone the directory: cd filepath
  3. Clone directory: git clone https://github.com/mrtnzram/Nester.git
  4. Create conda environment: conda create -n env_name python==3.10.16 pip
  5. Activate environment: conda activate env_name
  6. Install interactive widget dependencies: conda install -c conda-forge ipympl jupyterlab
  7. If Jupyter is not already installed, install Jupyter: pip install jupyter
  8. Run Jupyter lab: jupyter lab

Contributing to Nester

Nester is currently in its early stages of development. In the near future, Nester would ideally become a software with a GUI. If you would like to contribute and create a GUI for Nester feel free to use the code. If you would like to suggest any changes, please feel free to submit a pull request.

If you find any bugs or would like to suggest changes or improvements to Nester, please create a new issue at github/mrtnzram/Nester/issues.

About

Semi-automatic syntaxing of bird syllables and animal sounds using machine learning algorithms in JupyterLab

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors