This GitHub repository contains code that will assist current and future Wildlife Insights data providers to get their data into the format required to be uploaded into Wildlife Insights. The Wildlife Insights Batch Upload Template Data Dictionaries contain the guidelines for creating these templates. All code examples in this repository end up producing these templates. If you need help with any of this please contact us at info@wildlifeinsights.org. We definitely do not want data transformation to be a barrier to getting your data into Wildlife Insights.
Datasets A place to store your dataset that needs to be transformed. This also contains an example dataset.
Docs Overall tracking of migration code from existing camera trap software programs and tools. Our goal is to have data migration code for all the software programs and standards mentioned in the Young et al paper.
Transformation_Code Code (usually in R or Python) that has been used to migrate datasets from some format into the Wildlife Insights Batch Upload templates. All code is in a directory with the name of the: source software, program, format or organization that owned the datasets. The goal is to 1) identify the orginal format or source so that it will be useful to anyone starting from these formats and 2) archive this code so that it can be the basis for new code needed for new data formats.
WI_Global_Taxonomy This directory contains tools that will help you map your taxonomy (the scientific taxonomy or common names you use in your dataset) to our Wildlife Insights Global Taxonomy.
- If you are going to use R, python or another language and want to use the code available here, modify it or use it as psuedo-code for what you want to write, all you need to do is clone or download this repository. Do this by selecting the options in the top right of this page.
- Create a taxonomic mapping between your taxonomic data and the Wildlife Inisghts Global Taxonomy (WIGT). The WI_Global_Taxonomy directory contains tools to help you build a lookup table that maps the names in your dataset (i.e. the taxonomic or common names to describe wildlife and objects seen in your images) to our WIGT. We will be building this out more to add in additional tools to help identify scientific name issues, better handle situations where identification was done using common names and ultimately help you make a mapping between your data and the WIGT.
- Review the R script (
Transformation_Code/Wild_ID) that transforms and the example dataset (Datasests/Example_Dataset/). Transformation code from other standardized formats are also available as well as generic functions in theTransformation_Codedirectory. If you need help or have questions, please contact info@wildlifeinsights.org describing your projects. - Upload your images into the Google Cloud Platform. Please create a Google Cloud Account and read through this documentation. If you have any questions contact us at info@wildlifeinsights.org.
If you want to use R and haven't done so before, initial instructions for installing R, RStudio and Git (if desired) are below.
- Install R: You can download the latest version of R here.
- Install Rstudio: Rstudio Download link
- Install Git: See these installation notes - only needed if you want to share your code with this repository and/or would like your local copy to stay updated as we add and improve code here.
- Clone or download this repository - green button in top-right of this page.
If you have any questions or need help getting your data into the WI Batch Upload Templates please contact us at info@wildlifeinsights.org.