Usage: <MCCV cannot be invoked directly from the command line>
Wrapper for MCCV (Monte-Carlo Cross Validation).
Brief Algorithm Description:
Required Parameters: (note: the list enclosed in the brakets are possible
values each one of parameters can take ) mccv_parameter_name = x
The parameters name to iterate over when performing
MCCV. Typically k is chosen.
mccv_parameter_values = [list]
A list of values to substitue for mccv_parameter_name.
Since the parameters files are executed as python code,
an easy way to perform MCCV over a sequence is to do
something similar to mccv_parameter_values = range(10)
mccv_test_fraction = x
x is a value between 0.0 and 1.0 which indicates the
percentage of the dataset to use for testing.
mccv_num_trials = x
Number of times to run each algorithm for a particular
value in mccv_parameter_values. An algorithm will be
run a total of mccv_num_trials * len(mccv_parameter_values)
times
Optional / Dependent Parameters:
seed = x
The seed to use for the pseudo-random number generator
when randomly partitioning the training set. Defaults to 42.
mccv_fitness = filename / or yes
The mccv_fitness specifies a file to save the
fitness table computed by MCCV. Without any directory
information the default directory will be the same
as the results file, any provided directory information
will completely override the save location.
if yes, the output files basename is used, but an
extenstion of .fit is used
mccv_save_state = filename
The mccv_save_state option specifies a file to save
a pickled version of the algorithm classes run by mccv.
Like mccv_fitness it defaults to the results directory
when the directory portion of the path is not specified.
Imported modules
|
|
import Numeric
from compClust.mlx.ML_Algorithm import ML_Algorithm
from compClust.util import NaN
from compClust.util.TimeStampedPrintStream import TimeStampedPrintStream
import random
import sys
from types import *
|
Classes
|
|
|
|