delfi is a Python package for density estimation likelihood-free inference. In particular it includes several algorithms for Sequential Neural Posterior Estimation (SNPE).
Several inference algorithms are implemented:
- A basic version of a likelihood-free inference algorithm that uses a mixture-density network to approximate the posterior density
- The SNPE-A algorithm proposed in the paper Fast ε-free Inference of Simulation Models with Bayesian Conditional Density Estimation (Papamakarios & Murray, 2016)
- SNPE-B, as proposed in the paper Flexible statistical inference for mechanistic models of neural dynamics (Lueckmann, Goncalves, Bassetto, Öcal, Nonnenmacher & Macke, 2017)
- APT as proposed in Automatic posterior transformation for likelihood free inference (Greenberg, Nonnenmacher & Macke, 2019)
Please note that the code in this repository is still experimental. An early-stage documentation including installation instructions and a guide on how to get started is available at http://www.mackelab.org/delfi/.