ethology-test-data

ethology-test-data

Test data for the ethology Python package. For a description of the available test data, see the description of available data.

Adding or editing data

Folder structure

When adding data for a specific test module, please include it under the subdirectory with the same name as the test module:

test_data/test_<module-name>/

This will allow us to make efficient use of the pooch library and the pytest fixtures.

⚠️IMPORTANT⚠️:

Please ensure that the filenames are unique across the whole repository. This is because during testing, we currently use the files' names as keys to access the data, and having duplicate names will cause (silent) conflicts.

Steps

Remember to update the file registry at files-registry.txt after adding or editing any data. To do so, run the following from the root of the GIN repository:

1. Ensure all the content is downloaded locally:

gin download --content

2. After editing or adding files, run the provided Python script to re-generate the file-registry.txt file:

python get_sha256_hashes.py

3. Commit and upload the changes to the GIN repository

gin commit -m "Update file and file-registry.txt" <path-to-edited-file> file-registry.txt
gin upload <path-to-edited-file> file-registry.txt

For more details on how to work with GIN data repositories, please see the GIN HowTo guide.

External datasets

Australian Camera Trap Data (ACTD)

We use a bounding boxes annotation file from the Australian Camera Trap Data (ACTD) dataset for testing (ACTD_1_Terrestrial_group_data_CCT.json). This data set is released under the CC BY 4.0. The dataset is described in more detail in the associated manuscript:

Muhammad, Sameeruddin; Mann, Scott; Luke, Callum; Pocknee, Chris; Nair, Supriya; Nair, Jay (2024). Australian Camera Trap Data (ACTD). figshare. Dataset. DOI

Snapshot Serengeti

We use the bounding boxes file from the Snapshot Serengeti dataset for testing (SnapshotSerengetiBboxes_20190903.json). This data set is released under the Community Data License Agreement (permissive variant). The images and species-level labels are described in more detail in the associated manuscript:

Swanson AB, Kosmala M, Lintott CJ, Simpson RJ, Smith A, Packer C (2015) Snapshot Serengeti, high-frequency annotated camera trap images of 40 mammalian species in an African savanna. Scientific Data 2: 150026. (DOI)