Hulk의 개인 공부용 블로그

GPyOpt - 처음 tutorial 진행

17 Oct 2017 | bayesian inference 
from GPyOpt.methods import BayesianOptimization
import numpy as np

# --- Define your problem
def f(x):
    return (6*x - 2)**2 * np.sin(12 * x - 4)
bounds = [(0,1)]

# --- Solve your problem
myBopt = BayesianOptimization(f=f, bounds=bounds)
myBopt.run_optimization(max_iter=16)
myBopt.plot_acquisition()

output_0_0.png

print(myBopt.x_opt)
print(myBopt.Y_new)

[ 0.75506832]
[[-5.66412991]]

domain 여러개 확인..

tensorflow에 적용하기 전에, discrete한게 되는지 봐야함.

from GPyOpt.methods import BayesianOptimization
import numpy as np

# --- Define your problem
def f(x):
    return (6*x[0,0] - 2)**2 * np.sin(12 * x[0, 1] - 4)

domain = [{'name': 'a',
           'type': 'continuous',
           'domain':(0,10)},
          {'name': 'test',
           'type': 'discrete',
           'domain':(0,1,2,3,4,5,6,7,8,9,10),
           'dimensionality': 1}]

# --- Solve your problem
myBopt = BayesianOptimization(f=f, domain=domain)
myBopt.run_optimization(max_iter=5)
myBopt.plot_acquisition()

myBopt.X

array([[  2.0503028 ,   6.        ],
       [  2.44130349,   4.        ],
       [  5.74897004,   7.        ],
       [  2.66912911,   9.        ],
       [  6.31059492,   8.        ],
       [  7.10601309,   7.        ],
       [  8.53670111,   6.        ],
       [ 10.        ,   5.        ],
       [  9.49270161,   7.        ],
       [ 10.        ,   8.        ]])

myBopt.x_opt

array([ 9.49270161,  7.        ])

myBopt.Y

array([[ -9.52947568e+01],
       [  2.83173048e+00],
       [ -1.04939570e+03],
       [ -6.31711134e+01],
       [ -1.00254584e+03],
       [ -1.64119927e+03],
       [ -2.17534541e+03],
       [ -1.75449757e+03],
       [ -3.00172758e+03],
       [ -2.62212386e+03]])

Related Posts