DeepDive is a method for estimating biodiversity from fossil data using simulation-trained deep learning models.
The repositories here contain:
-
The DeepDive Python programme along with installation instructions.
-
The DeepDiveR R package that provides an interface to setup DeepDive analyses from fossil occurrence data.
-
For worked examples using DeepDiveR see carnivoran_analysis. For Python scripts compatible with older versions of DeepDive see marine_analysis and proboscidean_analysis.
Read more about the programme in our open access paper and application note.