UncertaintyQuantification.jl
UncertaintyQuantification.jl copied to clipboard
Bayesian updating framework
This is where I'll update/improve the framework to better integrate it into the existing code base.
Implemented right now are
- Metropolis Hastings
mh()
- Gibbs Sampler
gibbssample()
- Transitional MCMC
tmcmc()
- Sequential Monte Carlo
smc()
along with some helper methods.
To clean this up I would like to unify the interface somehow. My suggestion:
Have all methods called bayesian_inference()
or some variation of it. These will take the values that are the same among all algorithms while the extra parameters are grouped into structs like
struct TransitionalMarkovChainMonteCarlo begin
...
end
struct SequentialMonteCarlo begin
...
end
For metropolis-hastings as well as tmcmc we would just wrap around TransitionalMCMC.jl
if possible. No need to reinvent the wheel.
Ideally, the method signature would include the arguments that all of the methods have (if they exist). Like
function bayesian_model_updating(likelihood::Function, prior::Function, bmu::AbstractBayesianModelUpdating)
...
end
where the bmu
includes everything else based on the method that should be used (Transitional, Sequential, MH, etc.). If that makes sense.
Having a common interface and grouping the extras in the struct
seems fairly reasonable. Not 100% sure though.
Codecov Report
Attention: 228 lines
in your changes are missing coverage. Please review.
Comparison is base (
e2d4772
) 98.27% compared to head (c182077
) 80.95%. Report is 3 commits behind head on master.
:exclamation: Current head c182077 differs from pull request most recent head 6ffd1ae. Consider uploading reports for the commit 6ffd1ae to get more accurate results
Additional details and impacted files
@@ Coverage Diff @@
## master #63 +/- ##
===========================================
- Coverage 98.27% 80.95% -17.32%
===========================================
Files 27 30 +3
Lines 1045 1292 +247
===========================================
+ Hits 1027 1046 +19
- Misses 18 246 +228
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.