fixed some bug in computation of acceptance probability in MetropolisHastings,...
fixed some bug in computation of acceptance probability in MetropolisHastings, now works with restrained value ranges too. Also changed sample_local in continuous node to direct sampling so there is no rejection sampling anymore
Showing
- examples/MCMC_Continuous.py 12 additions, 9 deletionsexamples/MCMC_Continuous.py
- primo/reasoning/ContinuousNode.py 61 additions, 10 deletionsprimo/reasoning/ContinuousNode.py
- primo/reasoning/Evidence.py 10 additions, 11 deletionsprimo/reasoning/Evidence.py
- primo/reasoning/MCMC.py 8 additions, 7 deletionsprimo/reasoning/MCMC.py
- primo/reasoning/MarkovChainSampler.py 3 additions, 3 deletionsprimo/reasoning/MarkovChainSampler.py
- primo/reasoning/__init__.py 1 addition, 1 deletionprimo/reasoning/__init__.py
- primo/reasoning/density/LinearBeta.py 25 additions, 3 deletionsprimo/reasoning/density/LinearBeta.py
- primo/reasoning/density/LinearExponential.py 11 additions, 3 deletionsprimo/reasoning/density/LinearExponential.py
- primo/reasoning/density/LinearGauss.py 13 additions, 2 deletionsprimo/reasoning/density/LinearGauss.py
Loading
Please register or sign in to comment