From 6e56aefbb84d4fe9416e073e54a510031d122d45 Mon Sep 17 00:00:00 2001
From: faridm69 <faridmohammadi69@gmail.com>
Date: Fri, 18 Sep 2020 17:36:24 +0200
Subject: [PATCH] [surrogate][GPE] model selection activated and y_normalize
has been set to True.
---
BayesValidRox/BayesInference/BayesInference.py | 2 +-
.../surrogate_models/surrogate_models.py | 16 ++++++++--------
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/BayesValidRox/BayesInference/BayesInference.py b/BayesValidRox/BayesInference/BayesInference.py
index a1c214226..950e16ce9 100644
--- a/BayesValidRox/BayesInference/BayesInference.py
+++ b/BayesValidRox/BayesInference/BayesInference.py
@@ -178,7 +178,7 @@ class BayesInference:
logL = multivariate_normal.logpdf(TotalOutputs, mean=Data, cov=covMatrix)
except:
logL = -np.inf
-
+
return logL
#--------------------------------------------------------------------------------------------------------
diff --git a/BayesValidRox/surrogate_models/surrogate_models.py b/BayesValidRox/surrogate_models/surrogate_models.py
index b1508bb70..101887435 100644
--- a/BayesValidRox/surrogate_models/surrogate_models.py
+++ b/BayesValidRox/surrogate_models/surrogate_models.py
@@ -885,7 +885,7 @@ class Metamodel:
return pca, OutputMatrix
#--------------------------------------------------------------------------------------------------------
- def GaussianProcessEmulator(self, X, y, autoSelect=False, varIdx=None):
+ def GaussianProcessEmulator(self, X, y, autoSelect=True, varIdx=None):
from sklearn.gaussian_process import GaussianProcessRegressor
from sklearn.gaussian_process.kernels import (RBF, Matern, RationalQuadratic,
@@ -893,11 +893,10 @@ class Metamodel:
ConstantKernel)
- kernels = [1.0 * RBF(length_scale=1.0, length_scale_bounds=(1e-1, 10.0)),
- 1.0 * RationalQuadratic(length_scale=1.0, alpha=0.1),
- # ConstantKernel(0.1, (0.01, 10.0))
- # * (DotProduct(sigma_0=1.0, sigma_0_bounds=(0.1, 10.0)) ** 2),
- 1.0 * Matern(length_scale=1.0, length_scale_bounds=(1e-1, 10.0),
+ kernels = [1.0 * RBF(length_scale=1.0, length_scale_bounds=(1e-5, 1e5)),
+ 1.0 * RationalQuadratic(length_scale=0.2, alpha=0.1),
+ # 1.0 * ExpSineSquared(length_scale=1.0, length_scale_bounds=(1e-05, 1e05)),
+ 1.0 * Matern(length_scale=1.0, length_scale_bounds=(1e-5, 1e5),
nu=1.5)]
if autoSelect:# Automatic selection of the kernel
@@ -905,7 +904,7 @@ class Metamodel:
BME = []
for i, kernel in enumerate(kernels):
gp[i] = GaussianProcessRegressor(kernel=kernel, n_restarts_optimizer=2,
- normalize_y=False)
+ normalize_y=True)
# Fit to data using Maximum Likelihood Estimation of the parameters
gp[i].fit(X, y)
@@ -915,7 +914,8 @@ class Metamodel:
gp = gp[np.argmax(BME)]
else:
- gp = GaussianProcessRegressor(kernel=kernels[-1], n_restarts_optimizer=2)
+ gp = GaussianProcessRegressor(kernel=kernels[-1], n_restarts_optimizer=2,
+ normalize_y=True)
gp.fit(X, y)
# Compute score
--
GitLab