grale.inverters

This module provides different ‘inverters’ that can be used to execute the genetic algorithm for lens inversion. Depending on the platform you’re working on, you may be able to use MPI to distribute the needed calculations (typically speeding up the process) for example.

In the inversion module, you can specify the inverter to use as a (case insensitive) string, or as an instance of the available inverter classes. Available are:

  • “Threads”/”Threads:numthreads” or an instance of ThreadsInverter

  • “MPI”/”MPI:numprocesses”/”MPI:numprocesses:numthreadsperprocess:extraoptions” or an instance of MPIProcessInverter

Defaults

grale.inverters.getDefaultInverter()

Returns the default inverter to use when running the genetic algorithm for the lens inversion. Can be changed using setDefaultInverter().

grale.inverters.setDefaultInverter(x)

Sets the default inverter to use when running the genetic algorithm.

InverterException

exception grale.inverters.InverterException

An exception that’s generated if something goes wrong in this module.

Inverters

class grale.inverters.ThreadsInverter(numThreads=0, feedbackObject=None, parametric=False)
__init__(numThreads=0, feedbackObject=None, parametric=False)
class grale.inverters.SingleProcessGdbInverter(feedbackObject=None, gdbExe='gdb', parametric=False)

If this inverter is used, a single process, single core method is used which is started using the debugger GDB. To be able to interact with GDB, it is started in an xterm process. This is meant to make debugging this multi-process architecure somewhat easier.

__init__(feedbackObject=None, gdbExe='gdb', parametric=False)

Initializes an instance of this class; a specific feedback object can be specified for status updates.

class grale.inverters.MPIProcessInverter(numProcesses=None, numThreadsPerProcess=1, extraArgs=[], incomingConnectionTimeout=10, feedbackObject=None, parametric=False)
__init__(numProcesses=None, numThreadsPerProcess=1, extraArgs=[], incomingConnectionTimeout=10, feedbackObject=None, parametric=False)