Creating Site Specific Synthetic ML Training Datasets for Conservation: Sample Synthetic Training Dataset [Resized]

These files support the paper "Creating Site Specific Synthetic Machine Learning Training Datasets for Conservation" by providing a sample of the synthetic data generated using the described methodology. Additionally, a sample of this dataset could and has been used to train different image classifier machine learning models. 

The provided ZIP file contains four folders of synthetic training data JPEG images, separated by species, then background-type within the subsequent sub-folders. All of these folders and sub-folders are clearly labeled. The folder marked "Blank Backgrounds" is the only folder with images that DO NOT contain any herpetofauna. Instead, this folder consists of only background images, which were used to train an image classifier model to recognize the presence of (or lack there-of) a herpetofauna specimen within an image.  

*NOTE - These images were resized for the purposes of efficient upload to Github//Zenodo. There may be same quality/detail loss compared to the original images...*