I have recently attempted to use the Bayesian inference tool. To start, I simply ran the first example problem, i.e., uq_Example_Inversion_01_Beam. I get this error with uq_display which is the same as https://uqworld.org/t/error-with-uq-display/351 . The solution in this post is to reinstall Matlab, which I do not want to do.
I ran another inference problem, this time, uq_Example_Inversion_04_PredPrey.m. The script breaks at uq_display(myBayesianAnalysis, ‘trace’, 1) line. And, I get the following error,
Data must be a single matrix Y or a list of pairs X,Y.
Error in uq_traceplot (line 125)
Error in uq_AIES (line 208)
plotHandle = uq_traceplot(Sample(1:ii,plotParameters,:),…
Error in uq_inversion (line 91)
Results.LogLikeliEval] = uq_AIES(…
Error in uq_initialize_uq_inversion (line 708)
Further digging into the problem, I realize there is some problem with uq_plot because the following breaks too.
InputOpts.Marginals(1).Type = ‘Uniform’;
InputOpts.Marginals(1).Parameters = [0 1];
InputOpts.Marginals(2).Type = ‘Uniform’;
InputOpts.Marginals(2).Parameters = [0 1];
myInput = uq_createInput(InputOpts);
X_MC = uq_getSample(80,‘MC’);
uq_plot(ax, X_MC(:,1), X_MC(:,2), ‘.’, ‘MarkerSize’, 10)
It seems like uq_plot is not able to handle the axes handle. Please help me to rectify this.
Moreover, the axis labels need the latex interpreter option, which I found missing in several codes (e.g., uq_traceplot.m).