Package dsc_suite :: Package opt :: Module monte_carlo_adjusted
[hide private]
[frames] | no frames]

Module monte_carlo_adjusted

source code

Created on 08.06.2010


Author: Robert Fischbach, (adepted to GUI by Tobias Heimpold)

Generating solutions randomly distributed over the solution space without any dependencies between them.

Functions [hide private]
 
generate_monte_carlo_data(functions, parameters, file_info, time_check=False)
Sample solution space randomly, evenly distributed and without dependencies.
source code
Variables [hide private]
  __package__ = 'dsc_suite.opt'
Function Details [hide private]

generate_monte_carlo_data(functions, parameters, file_info, time_check=False)

source code 
Sample solution space randomly, evenly distributed and without dependencies.
parameter:
- functions: dictionary with functions from datastructure
  essential keys:
   - "randomSolution" : returns a random representation
   - "costEvaluation" : returns a list of the calculated costs for the given
                        representation
                        
- parameters : dictionary with algorithm parameters
  essential keys:
   - "samples" : number of generated random representations
   
- file_info: dictionary with information about the file names of the data files
  essential keys:
   - "trial_name" : name of trial given in GUI
   - "file_name_list" : additional information included in file name (can be empty string)
   
- time_check: boolean
  True: estimates runtime with reduced samples
  False: full calculation with saving data to files