- Greece data converted to wav estimating time from bitrate - might be imprecise.
- Data from Espartar unusable, drop
- Only one recording from Benidorm usable, not representative. Drop.
- Marettimo population only one (pretty bad) recording
A phrase defined as uninterrupted trill or purr + subsequent breath note, the latter included; ends where next purr begins or there is a longer silence.
- Clone the repository:
git clone https://github.com/nilomr/storm-petrel-song.git
. - Install python package (Install in an isolated environment, avoid a dependency hell!):
pip install .
(install) orpip install -e .
(developer install). - To install the R dependencies, use the following command (requires
renv
and R v4.2.1):renv::restore()
Note to self: As of 2024-03-25, had to pass full path to garden. Also had to
conda install -c conda-forge libstdcxx-ng
.
├── .gitignore # Specifies files and directories to be ignored by Git
├── .Rprofile # R configuration file
├── config.yml # Configuration file for the R scripts
├── data/ # Main data folder (not version-tracked)
│ ├── external/ # External data
│ ├── interim/ # Intermediate data
│ ├── processed/ # Processed data
│ └── raw/ # Raw data
├── dependencies.R # R script to manage vscode dependencies
├── LICENSE # License file for the project
├── output/ # Output directory for generated files
│ ├── figures/ # Generated figures
│ └── reports/ # Generated reports
├── pyproject.toml # Python project configuration file
├── R/ # R scripts directory
│ ├── rplot.R # R script for plotting
│ └── utils.R # Utility functions for R
├── README.md # The top-level README file
├── renv/ # R environment management directory
├── renv.lock # Lock file for renv
├── resources/ # Directory for resource files
│ ├── features.csv # Acoustic features under analysis
│ ├── rf_feature_importance.csv # Random forest feature importance
│ └── sbsp_rf_feature_importance.csv # Random forest feature importance (binary task)
├── scripts/ # Directory for Python scripts
│ ├── 0.1-segment-raw-data.py # Script for segmenting raw data
│ ├── 0.2-segment-units.py # Script for segmenting units
│ ├── 0.3-build-dataset.py # Script for building dataset
│ ├── 0.4-extract-measurements.py # Script for extracting measurements
│ ├── 0.5-train-model.py # Script for training model
│ └── ... # Additional scripts
├── tempesta/ # Source code directory for the Tempesta (python) project
└── tempesta.Rproj # RStudio project file for Tempesta