From 26a7de371960a92a24f14f5da760e63438b566ca Mon Sep 17 00:00:00 2001
From: odunbar <odunbar@caltech.edu>
Date: Mon, 26 Sep 2022 17:55:12 -0700
Subject: [PATCH] example to produce comparable figs

---
 examples/Emulator/GaussianProcess/plot_GP.jl         | 12 ++++++------
 .../Emulator/RandomFeature/optimize_and_plot_RF.jl   |  6 +++---
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/examples/Emulator/GaussianProcess/plot_GP.jl b/examples/Emulator/GaussianProcess/plot_GP.jl
index 5a815969c..09c822158 100644
--- a/examples/Emulator/GaussianProcess/plot_GP.jl
+++ b/examples/Emulator/GaussianProcess/plot_GP.jl
@@ -68,9 +68,9 @@ if !isdir(output_directory)
 end
 
 #create the machine learning tools: Gaussian Process
-gppackage = GPJL()
+gppackage = SKLJL()
 pred_type = YType()
-gaussian_process = GaussianProcess(gppackage, noise_learn = true)
+gaussian_process = GaussianProcess(gppackage, noise_learn = false)
 
 # Generate training data (x-y pairs, where x ∈ ℝ ᵖ, y ∈ ℝ ᵈ)
 # x = [x1, x2]: inputs/predictors/features/parameters
@@ -92,7 +92,7 @@ gx[2, :] = g2x
 
 # Add noise η
 μ = zeros(d)
-Σ = 0.1 * [[0.8, 0.0] [0.0, 0.5]] # d x d
+Σ = 0.1 * [[0.8, 0.1] [0.1, 0.5]] # d x d
 noise_samples = rand(MvNormal(μ, Σ), n)
 # y = G(x) + η
 Y = gx .+ noise_samples
@@ -182,9 +182,9 @@ println("GP trained")
 
 # Plot mean and variance of the predicted observables y1 and y2
 # For this, we generate test points on a x1-x2 grid.
-n_pts = 50
-x1 = range(0.0, stop = 2 * π, length = n_pts)
-x2 = range(0.0, stop = 2 * π, length = n_pts)
+n_pts = 200
+x1 = range(0.0, stop = (4.0/5.0) * 2 * π, length = n_pts)
+x2 = range(0.0, stop = (4.0/5.0) * 2 * π, length = n_pts)
 X1, X2 = meshgrid(x1, x2)
 # Input for predict has to be of size N_samples x input_dim
 inputs = permutedims(hcat(X1[:], X2[:]), (2, 1))
diff --git a/examples/Emulator/RandomFeature/optimize_and_plot_RF.jl b/examples/Emulator/RandomFeature/optimize_and_plot_RF.jl
index ed9b088fb..5abd2792a 100644
--- a/examples/Emulator/RandomFeature/optimize_and_plot_RF.jl
+++ b/examples/Emulator/RandomFeature/optimize_and_plot_RF.jl
@@ -125,7 +125,7 @@ n_features = 200
 
 # hyperparameter prior
 μ_l = 5.0
-σ_l = 10.0
+σ_l = 5.0
 prior_lengthscale = constrained_gaussian("lengthscale", μ_l, σ_l, 0.0, Inf, repeats = p)
 
 srfi = ScalarRandomFeatureInterface(n_features,prior_lengthscale)
@@ -138,8 +138,8 @@ optimize_hyperparameters!(emulator) # although RF already optimized
 # Plot mean and variance of the predicted observables y1 and y2
 # For this, we generate test points on a x1-x2 grid.
 n_pts = 200
-x1 = range(0.0, stop = 2 * π, length = n_pts)
-x2 = range(0.0, stop = 2 * π, length = n_pts)
+x1 = range(0.0, stop = 4.0/5.0* 2 * π, length = n_pts)
+x2 = range(0.0, stop = 4.0/5.0* 2 * π, length = n_pts)
 X1, X2 = meshgrid(x1, x2)
 # Input for predict has to be of size N_samples x input_dim
 inputs = permutedims(hcat(X1[:], X2[:]), (2, 1))