A tool for fitting Gaussian distributions to NMR T2 distributions to allow pore typing in carbonates using Bayesian Gaussian mixture models. Since scikit-learn 0.18, BayesianGaussianMixture is much faster than the PyMC3 implementation

1. Clone the repository
2. In the same folder location as the repository, create a folder called input_files where the LAS files will go
3. Each LAS file should have the TCMR curve and a T2 distribution. T2_MIN & T2_MAX should be 0.3 & 6000ms
4. Run anthills_sklearn.py
5. The outputs will be populated in a folder called anthills_Output
6. The calculated permeability curve can be loaded to Techlog via a CSV file