Run a Task individually on a list of inputs using the MPI process pool
Definition at line 523 of file parallel.py.
◆ __init__()
def lsst.ctrl.pool.parallel.BatchTaskRunner.__init__ |
( |
|
self, |
|
|
|
args, |
|
|
|
kwargs |
|
) |
| |
Constructor
Warn if the user specified multiprocessing.
Definition at line 526 of file parallel.py.
◆ __call__()
def lsst.ctrl.pool.parallel.BatchTaskRunner.__call__ |
( |
|
self, |
|
|
|
cache, |
|
|
|
args |
|
) |
| |
Run the Task on a single target
Strips out the process pool 'cache' argument.
'args' are those arguments provided by the getTargetList method.
Brings down the entire job if an exception is not caught (i.e., --doraise).
Definition at line 566 of file parallel.py.
◆ run()
def lsst.ctrl.pool.parallel.BatchTaskRunner.run |
( |
|
self, |
|
|
|
parsedCmd |
|
) |
| |
Run the task on all targets
Sole input is the result of parsing the command-line with the ArgumentParser.
Output is None if 'precall' failed; otherwise it is a list of calling ourself
on each element of the target list from the 'getTargetList' method.
Definition at line 537 of file parallel.py.
◆ numProcesses
lsst.ctrl.pool.parallel.BatchTaskRunner.numProcesses |
The documentation for this class was generated from the following file: