This is the repo for TraitBlender, a Blender addon and workflow for simulating raw datasets.
Major work on this project began at Image Datapalooza 2023, a workshop hosted by the Imageomics Institute. A major goal of the institute is to use Big Data and modern machine learning methods to better understand traits, such as phenotypes. However, many machine learning methods, such as neural networks, operate as "Black Boxes", making it difficult to understand their results and interpret their decisions. This problem is compounded by the fact that phenotypes themselves are often quite complex, and are often constructed and used differently by different researchers.
The decouple these different sources of confusion, we introduce TraitBlender, and Blender addon and workflow for simulating raw datasets under known processes. By raw data, we mean images and 3D models of simulated taxa.
There have long-been established software ecosystems for simulating tabular data under imaging evolutionary processes, such as the ape or phytools packages in R. These packages have empowered many researchers who develop models to test those models under common assumptions. As far as we know, however, no such accessible ecosystem previously existed for generating images are 3D models under these processes.
####Rest in progress
The following commands need to be run from the base directory of this repo.
docker build -t traitblender .
On Linux or Mac
mkdir results
docker run -v $(pwd)/results:/src/TraitBlender/results -it traitblender
On Windows you will need to replace $(pwd) with your current directory.
When the above command completes check the results directory for images created.