PDSampler.jl

PDSampler.jl copied to clipboard

Move this repo to the `JuliaAI` organization?

1

@ablaom @tlienart @vollmersj Any thoughts on moving this repo to `JuliaAI` as well?

DilumAluthge

Joris published his code on CRAN (https://cran.r-project.org/web/packages/RZigZag/) - have a look at the code - compare with PDSampler with ZZ - see if any improvements can be made

tlienart

loading time

3

due to dependencies, loading time is quite slow (several seconds). Would be good to check how to strip down dependencies so that only the strict minimum is loaded in order...

tlienart

At the moment we can't build an isotropic gaussian with just a float as covariance matrix, it has to be a vector of float so `s*ones(d)` is ok but not...

tlienart

At the moment this is commented out in https://github.com/alan-turing-institute/PDMP.jl/blob/master/src/kernels.jl#L36-L48 * Need to check that what is coded (by Sebastian) is correct * Need to compress it and make it more...

tlienart

It should work just the same * Add the possibility for `lambdaref=0.0` in the local simulate * Use the `gbps` kernel with standard specular reflection at the boundary * Check...

tlienart

* After #5 fix `ProgressMeter` accordingly * Add the same utility for the global BPS

tlienart

discuss stopping criterion

1

Proper stopping criterion (with theoretical guarantees) is when the clock in BPS or LBPS goes over a pre-specified (large) time `T` such as `1e8`. At the moment we're also stopping...

tlienart

In the helper function `ls_updatepq!`, those lines: ```julia bounce = fg.factors[fidx].nextevent(vcxf, vcvf) acc = bounce.dobounce(g, vcvf) while !acc bounce = fg.factors[fidx].nextevent(vcxf, vcvf) acc = bounce.dobounce(g, vcvf) end tauf = bounce.tau...

tlienart

That is, the number of dimensions of the dimensional vector (Xt,Vt) is independent of the number of data to be modeled. The velocity vector and the position vector will always...

chcolapara