algo = MipMRSort
while options.random_model_type != "default" \
and options.random_model_type != "choquet":
print("1. Default MR-Sort model")
print("2. Choquet MR-Sort model")
i = raw_input("Type of random model to initialize? ")
if i == '1':
options.random_model_type = 'default'
elif i == '2':
options.random_model_type = 'choquet'
random_model_type = options.random_model_type
options.na = read_multiple_integer(options.na,
"Number of assignment examples")
options.nc = read_multiple_integer(options.nc, "Number of criteria")
options.ncat = read_multiple_integer(options.ncat, "Number of categories")
options.na_gen = read_multiple_integer(options.na_gen, "Number of " \
"generalization alternatives")
options.pcerrors = read_multiple_integer(options.pcerrors, "Percentage " \
"of errors")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_%s-%s.csv" % (DATADIR, algo.__name__, dt)
options.filename = read_csv_filename(options.filename, default_filename)
directory = options.filename + "-data"
if not os.path.exists(directory):
os.makedirs(directory)
parser.add_option("-s",
"--nseeds",
action="store",
type="string",
dest="nseeds",
help="number of seeds")
parser.add_option("-f",
"--filename",
action="store",
type="string",
dest="filename",
help="filename to save csv output")
(options, args) = parser.parse_args()
options.na = read_multiple_integer(options.na, "Number of " \
"assignment examples")
options.nc = read_multiple_integer(options.nc, "Number of criteria")
options.ncat = read_multiple_integer(options.ncat, "Number of " \
"categories")
options.na_gen = read_multiple_integer(options.na_gen, "Number of " \
"generalization alternatives")
options.pcerrors = read_multiple_integer(options.pcerrors,
"Ratio of errors")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
options.ns = read_multiple_integer(options.ns, "Number of function " \
"segments")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_lp_avfsort-%s.csv" % (DATADIR, dt)
options.filename = read_csv_filename(options.filename, default_filename)
parser.add_option("-f",
"--filename",
action="store",
type="string",
dest="filename",
help="filename to save csv output")
(options, args) = parser.parse_args()
while not options.csvfile:
options.csvfile = raw_input("CSV file containing data ? ")
data = load_mcda_input_data(options.csvfile)
if data is None:
exit(1)
options.pclearning = read_multiple_integer(options.pclearning,
"Percentage of data to " \
"use in the learning set")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_mip_mrsort-%s-%s.csv" \
% (DATADIR, data.name, dt)
options.filename = read_csv_filename(options.filename, default_filename)
directory = options.filename + "-data"
if not os.path.exists(directory):
os.makedirs(directory)
run_tests(options.nseeds, data, options.pclearning, options.filename)
print("Results saved in '%s'" % options.filename)
print("2. PROPORTIONAL veto")
i = raw_input("What mode of veto? ")
if i == '1':
options.vetom = 'absolute'
elif i == '2':
options.vetom = 'proportional'
vetom = options.vetom
if vetom == 'absolute':
veto_func = 1
elif vetom == 'proportional':
veto_func = 2
options.vparam = read_multiple_float(options.vparam,
"Value of the veto param")
options.na = read_multiple_integer(options.na,
"Number of assignment examples")
options.nc = read_multiple_integer(options.nc, "Number of criteria")
options.ncat = read_multiple_integer(options.ncat, "Number of categories")
options.na_gen = read_multiple_integer(options.na_gen, "Number of " \
"generalization alternatives")
options.pcerrors = read_multiple_integer(options.pcerrors, "Ratio of " \
"errors")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_%s-%s.csv" % (DATADIR, "mip_mrsort_veto", dt)
options.filename = read_csv_filename(options.filename, default_filename)
directory = "%s/test_%s-%s" % (DATADIR, "mip_mrsort_veto", dt)
if not os.path.exists(directory):
os.makedirs(directory)
dest="filename",
help="filename to save csv output")
(options, args) = parser.parse_args()
while not options.csvfile:
options.csvfile = raw_input("CSV file containing data ? ")
data = load_mcda_input_data(options.csvfile)
if data is None:
exit(1)
if options.choquet is True:
lp_weights = LpMRSortMobius
heur_profiles = MetaMRSortProfilesChoquet
options.pclearning = read_multiple_integer(options.pclearning,
"Percentage of data to " \
"use in the learning set")
options.max_oloops = read_multiple_integer(options.max_oloops, "Max " \
"number of loops for " \
"profiles' metaheuristic")
options.nmodels = read_multiple_integer(options.nmodels,
"Population size (models)")
options.max_loops = read_multiple_integer(options.max_loops, "Max " \
"number of loops for the " \
"whole metaheuristic")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_meta_mrsort3-%s-%s.csv" \
% (DATADIR, data.name, dt)
options.filename = read_csv_filename(options.filename, default_filename)
parser.add_option("-s", "--nseeds", action = "store", type="string",
dest = "nseeds",
help = "number of seeds")
parser.add_option("-f", "--filename", action = "store", type="string",
dest = "filename",
help = "filename to save csv output")
(options, args) = parser.parse_args()
while not options.csvfile:
options.csvfile = raw_input("CSV file containing data ? ")
data = load_mcda_input_data(options.csvfile)
if data is None:
exit(1)
options.pclearning = read_multiple_integer(options.pclearning,
"Percentage of data to " \
"use in the learning set")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_mip_mrsort-%s-%s.csv" \
% (DATADIR, data.name, dt)
options.filename = read_csv_filename(options.filename, default_filename)
directory = options.filename + "-data"
if not os.path.exists(directory):
os.makedirs(directory)
run_tests(options.nseeds, data, options.pclearning, options.filename)
print("Results saved in '%s'" % options.filename)
parser.add_option("-o", "--max_oloops", action = "store", type="string",
dest = "max_oloops",
help = "max number of loops for the metaheuristic " \
"used to find the profiles")
parser.add_option("-f",
"--filename",
action="store",
type="string",
dest="filename",
help="filename to save csv output")
(options, args) = parser.parse_args()
algo = MetaMRSortPop3
options.na = read_multiple_integer(options.na,
"Number of assignment examples")
options.nc = read_multiple_integer(options.nc, "Number of criteria")
options.ncat = read_multiple_integer(options.ncat, "Number of categories")
options.ns = read_multiple_integer(options.ns, "Number of function " \
"segments")
options.na_gen = read_multiple_integer(options.na_gen, "Number of " \
"generalization alternatives")
options.pcerrors = read_multiple_integer(options.pcerrors, "Ratio of " \
"errors")
options.max_oloops = read_multiple_integer(options.max_loops, "Max " \
"number of loops for " \
"profiles' metaheuristic")
options.nmodels = read_multiple_integer(options.nmodels,
"Population size (models)")
options.max_loops = read_multiple_integer(options.max_loops, "Max " \
"number of loops for the " \
parser.add_option("-s", "--nseeds", action = "store", type="string",
dest = "nseeds",
help = "number of seeds")
parser.add_option("-f", "--filename", action = "store", type="string",
dest = "filename",
help = "filename to save csv output")
(options, args) = parser.parse_args()
while not options.csvfile:
options.csvfile = raw_input("CSV file containing data ? ")
data = load_mcda_input_data(options.csvfile)
if data is None:
exit(1)
options.pclearning = read_multiple_integer(options.pclearning,
"Percentage of data to " \
"use in the learning set")
options.ns = read_multiple_integer(options.ns, "Number of function " \
"segments")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_lp_avfsort-%s-%s.csv" \
% (DATADIR, data.name, dt)
options.filename = read_csv_filename(options.filename, default_filename)
directory = options.filename + "-data"
if not os.path.exists(directory):
os.makedirs(directory)
run_tests(options.nseeds, data, options.pclearning, options.ns,
dest = "filename",
help = "filename to save csv output")
(options, args) = parser.parse_args()
while not options.csvfile:
options.csvfile = raw_input("CSV file containing data ? ")
data = load_mcda_input_data(options.csvfile)
if data is None:
exit(1)
if options.choquet is True:
lp_weights = LpMRSortMobius
heur_profiles = MetaMRSortProfilesChoquet
options.pclearning = read_multiple_integer(options.pclearning,
"Percentage of data to " \
"use in the learning set")
options.max_oloops = read_multiple_integer(options.max_oloops, "Max " \
"number of loops for " \
"profiles' metaheuristic")
options.nmodels = read_multiple_integer(options.nmodels,
"Population size (models)")
options.max_loops = read_multiple_integer(options.max_loops, "Max " \
"number of loops for the " \
"whole metaheuristic")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_meta_mrsort2-%s-%s.csv" \
% (DATADIR, data.name, dt)
options.filename = read_csv_filename(options.filename, default_filename)
parser.add_option("-m", "--nmodels", action = "store", type="string",
dest = "nmodels",
help = "Size of the population (of models)")
parser.add_option("-o", "--max_oloops", action = "store", type="string",
dest = "max_oloops",
help = "max number of loops for the metaheuristic " \
"used to find the profiles")
parser.add_option("-f", "--filename", action = "store", type="string",
dest = "filename",
help = "filename to save csv output")
(options, args) = parser.parse_args()
algo = MetaMRSortPop3
options.na = read_multiple_integer(options.na,
"Number of assignment examples")
options.nc = read_multiple_integer(options.nc, "Number of criteria")
options.ncat = read_multiple_integer(options.ncat, "Number of categories")
options.ns = read_multiple_integer(options.ns, "Number of function " \
"segments")
options.na_gen = read_multiple_integer(options.na_gen, "Number of " \
"generalization alternatives")
options.pcerrors = read_multiple_integer(options.pcerrors, "Ratio of " \
"errors")
options.max_oloops = read_multiple_integer(options.max_loops, "Max " \
"number of loops for " \
"profiles' metaheuristic")
options.nmodels = read_multiple_integer(options.nmodels,
"Population size (models)")
options.max_loops = read_multiple_integer(options.max_loops, "Max " \
"number of loops for the " \
options.third = True
elif i == '4':
options.fourth = True
i = 0
if options.third is True:
algo = MetaMRSortProfiles3
i += 1
if options.fourth is True:
algo = MetaMRSortProfiles4
i += 1
if i > 1:
print("Cannot select multiple algorithms at the same time")
sys.exit(1)
options.na = read_multiple_integer(options.na, "Number of " \
"assignment examples")
options.nc = read_multiple_integer(options.nc, "Number of criteria")
options.ncat = read_multiple_integer(options.ncat, "Number of " \
"categories")
options.na_gen = read_multiple_integer(options.na_gen, "Number of " \
"generalization alternatives")
options.pcerrors = read_multiple_integer(options.pcerrors,
"Ratio of errors")
options.max_loops = read_multiple_integer(options.max_loops,
"Max number of loops")
options.nseeds = read_single_integer(options.nseeds, "Number of seeds")
dt = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
default_filename = "%s/test_%s-%s.csv" \
% (DATADIR, algo.__name__, dt)
options.filename = read_csv_filename(options.filename, default_filename)