Options for run/1

The following are all optional. If an option term is not present the indicated default value is used.

copy(Copy) copies some of the files used to the results directory. The idea is that results directories can be made self explanatory. Copy should be one of `none' or `all' or it should be a list of : `data',`runfile',`script'(Scr) and `slp'. The default value is [runfile]. Note that you should substitute the scriptname used in your run of experiments, for Scr.

display(DMethod) method controlling display of models. Use `next_display_at'. If option is absent no models will be displayed.

id(ID) a number of instances of this option are allowed. Each $ID$ should correspond to the first argument of a run_data/11 clause. If absent, then all experiments described by run_data/11 will be pursued.

initial(M) registers initial model, M. If it is not present the initial model is sampled from the prior. This option is intended for advanced users only. You need to be very careful that your SLP will be able to reach efficiently to the/a proof of the model. (mcmcms/carts/bcw/edit/irun.pl demonstrates the option's usage.)

print(PointsList) points to be recorded to output files. If absent the default value of [24,28,29] is used. Other numbers are undocumented. If more detail is needed, please look at the source files for dbg/2. Point 24 prints the initial model, point 28 prints all subsequent visited models in the chain while 29 prints proposed models. The latter is printed in the form of b(Model) terms, while current models are deposited as c(Model). On how $Models$ are extracted from the top query, see Appendix B, 5th argument ($Call$).

progress(Atom,Percentile) if present, a progress bar will be displayed for every experiment within the run. $Atom$ will be written once for every completed $Percentile$ of the number of repetitions in each experiment.

results_dir(ResDir) is the directory for the output files from the experiments. Currently if the directory exists, there will be no complaints, and no clean-up. So unless you know what you are doing it is better if the directory does not exist. If $ResDir$ is a path, e.g. of the form dir1/sub1/test, which does not exist, then it will be created. Default value: `test'.

runs_file(RFile) file containing the run_data/11 definitions. Default: `run_test'.

source_dir(SDir) a directory relative to which runs/ slps/ and data/ are looked in for. Default value: `.'.