This C++ version of BAT is still being maintained, but addition of new features is unlikely. Check out our new incarnation, BAT.jl, the Bayesian analysis toolkit in Julia. In addition to Metropolis-Hastings sampling, BAT.jl supports Hamiltonian Monte Carlo (HMC) with automatic differentiation, automatic prior-based parameter space transformations, and much more. See the BAT.jl documentation.

Results of performance testing for BAT version 0.4.3

Back to | overview for 0.4.3 | all versions |

Test "2d_flat"

Results
Status good
CPU time 68.44 s
Real time 68.58 s
Plots 2d_flat.ps
Log 2d_flat.log

Settings
N chains 10
N lag 10
Convergence true
N iterations (pre-run) 2000
N iterations (run) 10000000

Plots
Auto-correlation for the parameter.Auto-correlation for the parameter.The analytic function drawn with contours.
The histogrammed analytic function. Each bin contains the integral of the analytic function over the bin.The distribution from MCMC.The histogrammed analytic function in log-scale. Each bin contains the integral of the analytic function over the bin.
The distribution from MCMC in log-scale.The difference between the distribution from MCMC and the analytic function divided by the square root of the analytic function value in the corresponding bin.The pull between the distribution from MCMC and the analytic function.

Subtest Status Target Test Uncertainty Deviation [%] Deviation [sigma] Tol. (Good) Tol. (Acceptable) Tol. (Bad)
correlation par 0 off 0 0.03708 0.003678 - -10.08 0.3 0.5 0.7
correlation par 1 off 0 0.03716 0.003745 - -9.923 0.3 0.5 0.7
chi2 good 1e+04 9974 141.4 -0.2613 0.1848 424.3 707.1 989.9

Subtest Description
correlation par 0 Calculate the auto-correlation among the points.
correlation par 1 Calculate the auto-correlation among the points.
chi2 Calculate χ2 and compare with prediction for dof=number of bins with an expectation >= 10.
Tolerance good: |χ2-E[χ2]| < 3 · (2 dof)1/2,
Tolerance acceptable: |χ2-E[χ2]| < 5 · (2 dof)1/2,
Tolerance bad: |χ2-E[χ2]| < 7 · (2 dof)1/2.