def kernel(self, points1, points2=None, degree=0, depth=1):
arc_cosine = kernels.ArcCosine(degree, depth, white=0.0)
if points2 is not None:
return tf.Session().run(arc_cosine.kernel(np.array(points1, dtype=np.float32),
np.array(points2, dtype=np.float32)))
else:
return tf.Session().run(arc_cosine.kernel(np.array(points1, dtype=np.float32)))
def kernel(cls, points1, points2=None, degree=0, depth=1):
arc_cosine = kernels.ArcCosine(degree, depth, white=0.0)
cls.session.run(tf.global_variables_initializer())
if points2 is not None:
return cls.session.run(
arc_cosine.kernel(np.array(points1, dtype=np.float32),
np.array(points2, dtype=np.float32)))
else:
return cls.session.run(
arc_cosine.kernel(np.array(points1, dtype=np.float32)))
def diag_kernel(self, points, degree=0, depth=1):
arc_cosine = kernels.ArcCosine(degree, depth, white=0.0)
return tf.Session().run(arc_cosine.diag_kernel(np.array(points, dtype=np.float32)))
test_data = pd.read_csv(TEST_PATH, sep=r"\s+", header=None)
train_X = train_data.values[:, :-1]
train_Y = train_data.values[:, -1:]
test_X = test_data.values[:, :-1]
test_Y = test_data.values[:, -1:]
data = datasets.DataSet(train_X, train_Y)
test = datasets.DataSet(test_X, test_Y)
Z = init_z(data.X, NUM_INDUCING)
likelihood = likelihoods.Logistic() # Setup initial values for the model.
if KERNEL == 'arccosine':
kern = [
kernels.ArcCosine(data.X.shape[1],
degree=DEGREE,
depth=DEPTH,
lengthscale=LENGTHSCALE,
std_dev=1.0,
input_scaling=IS_ARD) for i in range(1)
]
else:
kern = [
kernels.RadialBasis(data.X.shape[1],
lengthscale=LENGTHSCALE,
input_scaling=IS_ARD) for i in range(1)
]
print("Using Kernel " + KERNEL)
m = autogp.GaussianProcess(likelihood,
kern,
Z,
def diag_kernel(cls, points, degree=0, depth=1):
arc_cosine = kernels.ArcCosine(degree, depth, white=0.0)
cls.session.run(tf.global_variables_initializer())
return cls.session.run(
arc_cosine.diag_kernel(np.array(points, dtype=np.float32)))