def main():
runman = RunManager()
# Choose initialState, either from user-inputted parameters or randomly
if len(sys.argv) > 1:
initialState = [eval(xx) for xx in sys.argv[1].split()]
else:
initialState = randUniformPoint(SineModel5.typicalRanges)
runman.explore_dimensions(initialState, SineModel5.typicalRanges, pointsPerDim = 9, repetitions = 2)
def main():
runman = RunManager()
# Choose initialState, either from user-inputted parameters or randomly
if len(sys.argv) > 1:
initialState = [eval(xx) for xx in sys.argv[1].split()]
else:
initialState = randUniformPoint(SineModel5.typicalRanges)
runman.explore_dimensions(initialState,
SineModel5.typicalRanges,
pointsPerDim=9,
repetitions=2)
def doRun():
runman = RunManager()
# Choose initialState, either from user-inputted parameters or randomly
neatFile = None
if len(sys.argv) > 1:
if (len(sys.argv) > 2 and
sys.argv[1] == '-simplex' or sys.argv[1] == '-svm'):
# Simplex filename
import pickle
filename = sys.argv[2]
ff = open(filename, 'r')
strategy = pickle.load(ff)
ff.close()
elif (len(sys.argv) > 2 and sys.argv[1] == '-neat'):
neatFile = sys.argv[2]
currentState = None
elif (len(sys.argv) > 2 and sys.argv[1] == '-filt'):
filtFile = sys.argv[2]
currentState = None
else:
# normal
currentState = [eval(xx) for xx in sys.argv[1].split()]
else:
currentState = randUniformPoint(SineModel5.typicalRanges)
try:
strategy
except:
#strategy = UniformStrategy(currentState, SineModel5.typicalRanges)
#strategy = GaussianStrategy(currentState, SineModel5.typicalRanges)
#strategy = GradientSampleStrategy(currentState)
#strategy = LinearRegressionStrategy(currentState)
#strategy = SimplexStrategy(currentState, SineModel5.typicalRanges)
strategy = RandomStrategy(currentState, SineModel5.typicalRanges)
#strategy = SVMLearningStrategy(currentState, SineModel5.typicalRanges)
#strategy = NEATStrategy(currentState, SineModel5.typicalRanges, neatFile = neatFile) # these args aren't used
#strategy = FileStrategy(filtFile = filtFile)
#runman.do_many_runs(currentState, lambda state: Neighbor.gaussian(SineModel5.typicalRanges, state))
#runman.do_many_runs(currentState, lambda state: gradient_search(SineModel5.typicalRanges, state))
runman.do_many_runs(strategy, SineModel5.typicalRanges)
def doRun():
runman = RunManager()
# Choose initialState, either from user-inputted parameters or randomly
neatFile = None
if len(sys.argv) > 1:
if (len(sys.argv) > 2 and sys.argv[1] == '-simplex'
or sys.argv[1] == '-svm'):
# Simplex filename
import pickle
filename = sys.argv[2]
ff = open(filename, 'r')
strategy = pickle.load(ff)
ff.close()
elif (len(sys.argv) > 2 and sys.argv[1] == '-neat'):
neatFile = sys.argv[2]
currentState = None
elif (len(sys.argv) > 2 and sys.argv[1] == '-filt'):
filtFile = sys.argv[2]
currentState = None
else:
# normal
currentState = [eval(xx) for xx in sys.argv[1].split()]
else:
currentState = randUniformPoint(SineModel5.typicalRanges)
try:
strategy
except:
#strategy = UniformStrategy(currentState, SineModel5.typicalRanges)
strategy = GaussianStrategy(currentState, SineModel5.typicalRanges)
#strategy = GradientSampleStrategy(currentState)
#strategy = LinearRegressionStrategy(currentState)
#strategy = SimplexStrategy(currentState, SineModel5.typicalRanges)
#strategy = RandomStrategy(currentState, SineModel5.typicalRanges)
#strategy = SVMLearningStrategy(currentState, SineModel5.typicalRanges)
#strategy = NEATStrategy(currentState, SineModel5.typicalRanges, neatFile = neatFile) # these args aren't used
#strategy = FileStrategy(filtFile = filtFile)
#runman.do_many_runs(currentState, lambda state: Neighbor.gaussian(SineModel5.typicalRanges, state))
#runman.do_many_runs(currentState, lambda state: gradient_search(SineModel5.typicalRanges, state))
runman.do_many_runs(strategy, SineModel5.typicalRanges)