def __init__(self, points):
self.points = points
self.func = [
self.hardness_field, self.depth_field, self.signal_field,
self.random_field
]
self.func_names = [
'hardness field', 'depth field', 'signal field', 'random field'
]
self.res = 8
kernel = generate_rbfkern(2, 1.0, 1.5)
xs = generate_grid(lb=-2.0, ub=2.0, res=self.res)
ys = np.random.normal(size=(self.res * self.res, 1))
print(xs, ys)
self.true_res = 41
model = GPy.models.GPRegression(xs, ys, kernel, noise_var=1e-10)
x_disc = generate_grid(lb=-2.0, ub=2.0, res=self.true_res)
print(x_disc)
#model.plot()
#self.truesignal = model.predict(x_disc)[0].reshape(true_res, true_res)
#print(self.truesignal)
#kernel = lambda x1, x2: math.exp(- np.linalg.norm(x1 - x2)**2 / 2.0)
#training_inputs = generate_grid(lb=-2.0, ub=2.0, res=20)
#K =
#for i in range(400):
# for j in range(400):
# kernel(training_inputs[i], training_inputs[j])
self.truesignal = model.posterior_samples(x_disc, size=1).reshape(
self.true_res, self.true_res)
kernel = generate_rbfkern(2, 1.0, 1.5)
xs = generate_grid(lb=-2.0, ub=2.0, res=self.res)
ys = np.random.normal(size=(self.res * self.res, 1))
model = GPy.models.GPRegression(xs, ys, kernel, noise_var=1e-10)
x_disc = generate_grid(lb=-2.0, ub=2.0, res=self.true_res)
self.randsignal = model.posterior_samples(x_disc, size=1).reshape(
self.true_res, self.true_res)
ys = np.random.normal(size=(self.res * self.res, 1))
model = GPy.models.GPRegression(xs, ys, kernel, noise_var=1e-10)
x_disc = generate_grid(lb=-2.0, ub=2.0, res=self.true_res)
self.randsignal2 = model.posterior_samples(x_disc, size=1).reshape(
self.true_res, self.true_res)
ys = np.random.normal(size=(self.res * self.res, 1))
model = GPy.models.GPRegression(xs, ys, kernel, noise_var=1e-10)
x_disc = generate_grid(lb=-2.0, ub=2.0, res=self.true_res)
self.randsignal3 = model.posterior_samples(x_disc, size=1).reshape(
self.true_res, self.true_res)
def randomize(self, seed):
np.random.seed(seed)
kernel = generate_rbfkern(2, 1.0, 1.5)
xs = generate_grid(lb=-2.0, ub=2.0, res=self.res)
ys = np.random.normal(size=(self.res * self.res, 1))
model = GPy.models.GPRegression(xs, ys, kernel, noise_var=1e-10)
x_disc = generate_grid(lb=-2.0, ub=2.0, res=self.true_res)
self.truesignal = model.posterior_samples(x_disc, size=1).reshape(
self.true_res, self.true_res)
ys = np.random.normal(size=(self.res * self.res, 1))
model = GPy.models.GPRegression(xs, ys, kernel, noise_var=1e-10)
x_disc = generate_grid(lb=-2.0, ub=2.0, res=self.true_res)
print(x_disc)
self.randsignal = model.posterior_samples(x_disc, size=1).reshape(
self.true_res, self.true_res)
ys = np.random.normal(size=(self.res * self.res, 1))
model = GPy.models.GPRegression(xs, ys, kernel, noise_var=1e-10)
x_disc = generate_grid(lb=-2.0, ub=2.0, res=self.true_res)
self.randsignal2 = model.posterior_samples(x_disc, size=1).reshape(
self.true_res, self.true_res)
ys = np.random.normal(size=(self.res * self.res, 1))
model = GPy.models.GPRegression(xs, ys, kernel, noise_var=1e-10)
x_disc = generate_grid(lb=-2.0, ub=2.0, res=self.true_res)
self.randsignal3 = model.posterior_samples(x_disc, size=1).reshape(
self.true_res, self.true_res)
def __init__(self):
self.kernel = generate_rbfkern(2, 1.0, 0.3)
def __init__(self):
self.res = 20
self.kernel = generate_rbfkern(2, 1.0, 0.3)
self.entropy = 1.0
