Wrappers for the APPL executables. The pomdpsol function solves a model
file and returns the path to the output policy file.
pomdpsol(model, output = tempfile(), precision = 0.001, timeout = NULL, fast = FALSE, randomization = FALSE, memory = NULL, improvementConstant = NULL, timeInterval = NULL, stdout = tempfile(), stderr = tempfile(), spinner = TRUE) polgraph(model, policy, output = tempfile(), max_depth = 3, max_branches = 10, min_prob = 0.001, stdout = "", spinner = TRUE) pomdpsim(model, policy, output = tempfile(), steps = 100, simulations = 3, stdout = "", spinner = TRUE) pomdpeval(model, policy, output = tempfile(), steps = 100, simulations = 3, stdout = "", spinner = TRUE) pomdpconvert(model, stdout = "", spinner = TRUE)
| model | file/path to the |
|---|---|
| output | file/path of the output policy file. This is also returned by the function. |
| precision | targetPrecision. Set targetPrecision as the target precision in solution quality; run ends when target precision is reached. The target precision is 1e-3 by default. |
| timeout | Use timeLimit as the timeout in seconds. If running time exceeds the specified value, pomdpsol writes out a policy and terminates. There is no time limit by default. |
| fast | logical, default FALSE. use fast (but very picky) alternate parser for .pomdp files. |
| randomization | logical, default FALSE. Turn on randomization for the sampling algorithm. |
| memory | Use memoryLimit as the memory limit in MB. No memory limit by default. If memory usage exceeds the specified value, pomdpsol writes out a policy and terminates. Set the value to be less than physical memory to avoid swapping. |
| improvementConstant | Use improvementConstant as the trial improvement factor in the sampling algorithm. At the default of 0.5, a trial terminates at a belief when the gap between its upper and lower bound is 0.5 of the current precision at the initial belief. |
| timeInterval | Use timeInterval as the time interval between two consecutive write-out of policy files. If this is not specified, pomdpsol only writes out a policy file upon termination. |
| stdout | a filename where pomdp run statistics will be stored |
| stderr | currently ignored. |
| spinner | should we show a spinner while sarsop is running? |
| policy | file/path to the policy file |
| max_depth | the maximum horizon of the generated policy graph |
| max_branches | maximum number of branches to show in the policy graph |
| min_prob | the minimum probability threshold for a branch to be shown in the policy graph |
| steps | number of steps for each simulation run |
| simulations | as the number of simulation runs |
model <- system.file("models/example.pomdp", package = "sarsop") policy <- tempfile() pomdpsol(model, output = policy, timeout = 1)#> $load_time_sec #> [1] 0.21 #> #> $init_time_sec #> [1] 1.54 #> #> $run_time_sec #> [1] 1.56 #> #> $final_precision #> [1] 54.1782 #> #> $end_condition #> [1] " Preset timeout reached" #># Other tools evaluation <- pomdpeval(model, policy, stdout = FALSE) graph <- polgraph(model, policy, stdout = FALSE) simulations <- pomdpsim(model, policy, stdout = FALSE)