Python pcmaes_lsの例

コード例 #1

ファイル: lcmaes_interface.py プロジェクト: IloneM/libcmaes

def pcmaes(fitfunc,p):
    has_bounds = isinstance(p,lcmaes.CMAParametersPB) or isinstance(p,lcmaes.CMAParametersPBS)
    has_scaling = isinstance(p,lcmaes.CMAParametersNBS) or isinstance(p,lcmaes.CMAParametersPBS)
    if not has_bounds:
        if not has_scaling:
            return lcmaes.pcmaes(fitfunc,p)
        else:
            return lcmaes.pcmaes_ls(fitfunc,p)
    else:
        if not has_scaling:
            return lcmaes.pcmaes_pwqb(fitfunc,p)
        else:
            return lcmaes.pcmaes_pwqb_ls(fitfunc,p)

コード例 #2

ファイル: lcmaes_interface.py プロジェクト: zxp-proteus/libcmaes

def pcmaes(fitfunc, p):
    has_bounds = isinstance(p, lcmaes.CMAParametersPB) or isinstance(
        p, lcmaes.CMAParametersPBS)
    has_scaling = isinstance(p, lcmaes.CMAParametersNBS) or isinstance(
        p, lcmaes.CMAParametersPBS)
    if not has_bounds:
        if not has_scaling:
            return lcmaes.pcmaes(fitfunc, p)
        else:
            return lcmaes.pcmaes_ls(fitfunc, p)
    else:
        if not has_scaling:
            return lcmaes.pcmaes_pwqb(fitfunc, p)
        else:
            return lcmaes.pcmaes_pwqb_ls(fitfunc, p)

コード例 #3

ファイル: ptests_ls.py プロジェクト: montanier/libcmaes

seed = 0  # 0 for seed auto-generated within the lib.
sigma = 0.1
scaling = [int(1000 * random.random()) for i in xrange(10)]
shift = [int(1000 * random.random()) for i in xrange(10)]
gp = lcmaes.make_genopheno_ls(scaling, shift)
p = lcmaes.make_parameters_ls(x, sigma, gp, lambda_, seed)


# objective function.
def nfitfunc(x, n):
    val = 0.0
    for i in range(0, n):
        val += x[i] * x[i]
    return val


# generate a function object
objfunc = lcmaes.fitfunc_pbf.from_callable(nfitfunc)

# pass the function and parameter to cmaes, run optimization and collect solution object.
cmasols = lcmaes.pcmaes_ls(objfunc, p)

# collect and inspect results
bcand = cmasols.best_candidate()
bx = lcmaes.get_candidate_x(bcand)
print("best x=", bx)
print("distribution mean=", lcmaes.get_solution_xmean(cmasols))
cov = lcmaes.get_solution_cov(cmasols)  # numpy array
print("cov=", cov)
print("elapsed time=", cmasols.elapsed_time(), "ms")

コード例 #4

ファイル: ptests_ls.py プロジェクト: beniz/libcmaes

x = [3]*10
lambda_ = 10 # lambda is a reserved keyword in python, using lambda_ instead.
seed = 0 # 0 for seed auto-generated within the lib.
sigma = 0.1
scaling = [int(1000*random.random()) for i in range(10)]
shift = [int(1000*random.random()) for i in range(10)]
gp = lcmaes.make_genopheno_ls(scaling,shift)
p = lcmaes.make_parameters_ls(x,sigma,gp,lambda_,seed)

# objective function.
def nfitfunc(x,n):
    val = 0.0
    for i in range(0,n):
        val += x[i]*x[i]
    return val

# generate a function object
objfunc = lcmaes.fitfunc_pbf.from_callable(nfitfunc);

# pass the function and parameter to cmaes, run optimization and collect solution object.
cmasols = lcmaes.pcmaes_ls(objfunc,p)

# collect and inspect results
bcand = cmasols.best_candidate()
bx = lcmaes.get_candidate_x(bcand)
print("best x=",bx)
print("distribution mean=",lcmaes.get_solution_xmean(cmasols))
cov = lcmaes.get_solution_cov(cmasols) # numpy array
print("cov=",cov)
print("elapsed time=",cmasols.elapsed_time(),"ms")