stateline
stateline copied to clipboard
NICTA
Distributed Markov Chain Monte Carlo
The refactor of the stateline code is great! However, it seems a bug may have crept in in the handling of Inf likelihoods. Our model has a strong boundary for...
This will be helpful in the long run. Currently, Google Benchmark (https://github.com/google/benchmark) looks like a good candidate for performing the benchmarking. We just need a way to record benchmarking results...
Replace chain flat arrays with 2d arrays through code, simplify the complex object structures.
At CppCon 2015, the C++ standards committee open sourced a [guideline](https://github.com/isocpp/CppCoreGuidelines/blob/master/CppCoreGuidelines.md) for writing good modern C++. I think it would be a good idea for stateline to follow it. As...
Decimation and swap interval are the same parameter, and could be estimated heuristically?
O(N^2) is slow for high dimensional proposals (thousands of dimensions). Instead we could use a diagonal anisotropic covariance.
Can we estimate burn in length automatically? - Save it in the database - automatically stop adaption of proposals and betas - The adaption of proposals and betas is a...