Merge pull request #459 from ArnoStrouwen/docs

LanguageTool

Project.toml

@@ -19,7 +19,7 @@ Aqua = "0.8"
 Cubature = "1.5"
 Distributions = "0.25.71"
 ExtendableSparse = "1"
-Flux = "0.14"
+Flux = "0.13, 0.14"
 ForwardDiff = "0.10.19"
 GLM = "1.5"
 IterativeSolvers = "0.9"

docs/src/BraninFunction.md

@@ -1,13 +1,13 @@
 # Branin Function
 
-The Branin Function is commonly used as a test function for metamodelling in computer experiments, especially in the context of optimization.
+The Branin function is commonly used as a test function for metamodelling in computer experiments, especially in the context of optimization.
 
 The expression of the Branin Function is given as:
 ``f(x) = (x_2 - \frac{5.1}{4\pi^2}x_1^{2} + \frac{5}{\pi}x_1 - 6)^2 + 10(1-\frac{1}{8\pi})\cos(x_1) + 10``
 
 where ``x = (x_1, x_2)`` with ``-5\leq x_1 \leq 10, 0 \leq x_2 \leq 15``
 
-First of all we will import these two packages `Surrogates` and `Plots`.
+First of all, we will import these two packages: `Surrogates` and `Plots`.
 
 ```@example BraninFunction
 using Surrogates
@@ -50,7 +50,7 @@ scatter!(xs, ys)
 plot(p1, p2, title="True function")
 ```
 
-Now it's time to try fitting different surrogates and then we will plot them.
+Now it's time to try fitting different surrogates, and then we will plot them.
 We will have a look at the radial basis surrogate `Radial Basis Surrogate`. :
 
 ```@example BraninFunction
@@ -65,7 +65,7 @@ scatter!(xs, ys, marker_z=zs)
 plot(p1, p2, title="Radial Surrogate")
 ```
 
-Now, we will have a look on `Inverse Distance Surrogate`:
+Now, we will have a look at `Inverse Distance Surrogate`:
 ```@example BraninFunction
 InverseDistance = InverseDistanceSurrogate(xys, zs,  lower_bound, upper_bound)
 ```

docs/src/InverseDistance.md

@@ -1,4 +1,4 @@
-The **Inverse Distance Surrogate** is an interpolating method and in this method the unknown points are calculated with a weighted average of the sampling points. This model uses the inverse distance between the unknown and training points to predict the unknown point. We do not need to fit this model because the response of an unknown point x is computed with respect to the distance between x and the training points.
+The **Inverse Distance Surrogate** is an interpolating method, and in this method, the unknown points are calculated with a weighted average of the sampling points. This model uses the inverse distance between the unknown and training points to predict the unknown point. We do not need to fit this model because the response of an unknown point x is computed with respect to the distance between x and the training points.
 
 Let's optimize the following function to use Inverse Distance Surrogate:
 
@@ -53,7 +53,7 @@ plot!(InverseDistance, label="Surrogate function",  xlims=(lower_bound, upper_bo
 
 Having built a surrogate, we can now use it to search for minima in our original function `f`.
 
-To optimize using our surrogate we call `surrogate_optimize` method. We choose to use Stochastic RBF as optimization technique and again Sobol sampling as sampling technique.
+To optimize using our surrogate we call `surrogate_optimize` method. We choose to use Stochastic RBF as the optimization technique and again Sobol sampling as the sampling technique.
 
 ```@example Inverse_Distance1D
 @show surrogate_optimize(f, SRBF(), lower_bound, upper_bound, InverseDistance, SobolSample())
@@ -65,7 +65,7 @@ plot!(InverseDistance, label="Surrogate function",  xlims=(lower_bound, upper_bo
 
 ## Inverse Distance Surrogate Tutorial (ND):
 
-First of all we will define the `Schaffer` function we are going to build surrogate for. Notice, one how its argument is a vector of numbers, one for each coordinate, and its output is a scalar.
+First of all we will define the `Schaffer` function we are going to build a surrogate for. Notice, how its argument is a vector of numbers, one for each coordinate, and its output is a scalar.
 
 ```@example Inverse_DistanceND
 using Plots # hide
@@ -84,7 +84,7 @@ end
 
 ### Sampling
 
-Let's define our bounds, this time we are working in two dimensions. In particular we want our first dimension `x` to have bounds `-5, 10`, and `0, 15` for the second dimension. We are taking 60 samples of the space using Sobol Sequences. We then evaluate our function on all of the sampling points.
+Let's define our bounds, this time we are working in two dimensions. In particular we want our first dimension `x` to have bounds `-5, 10`, and `0, 15` for the second dimension. We are taking 60 samples of the space using Sobol Sequences. We then evaluate our function on all the sampling points.
 
 ```@example Inverse_DistanceND
 n_samples = 60
@@ -124,7 +124,7 @@ plot(p1, p2, title="Surrogate") # hide
 
 
 ### Optimizing
-With our surrogate we can now search for the minima of the function.
+With our surrogate, we can now search for the minima of the function.
 
 Notice how the new sampled points, which were created during the optimization process, are appended to the `xys` array.
 This is why its size changes.

docs/src/LinearSurrogate.md

@@ -28,7 +28,7 @@ plot!(f, label="True function", xlims=(lower_bound, upper_bound))
 
 ## Building a Surrogate
 
-With our sampled points we can build the **Linear Surrogate** using the `LinearSurrogate` function.
+With our sampled points, we can build the **Linear Surrogate** using the `LinearSurrogate` function.
 
 We can simply calculate `linear_surrogate` for any value.
 
@@ -51,7 +51,7 @@ plot!(my_linear_surr_1D, label="Surrogate function",  xlims=(lower_bound, upper_
 
 Having built a surrogate, we can now use it to search for minima in our original function `f`.
 
-To optimize using our surrogate we call `surrogate_optimize` method. We choose to use Stochastic RBF as optimization technique and again Sobol sampling as sampling technique.
+To optimize using our surrogate we call `surrogate_optimize` method. We choose to use Stochastic RBF as the optimization technique and again Sobol sampling as the sampling technique.
 
 ```@example linear_surrogate1D
 @show surrogate_optimize(f, SRBF(), lower_bound, upper_bound, my_linear_surr_1D, SobolSample())
@@ -63,7 +63,7 @@ plot!(my_linear_surr_1D, label="Surrogate function",  xlims=(lower_bound, upper_
 
 ## Linear Surrogate tutorial (ND)
 
-First of all we will define the `Egg Holder` function we are going to build surrogate for. Notice, one how its argument is a vector of numbers, one for each coordinate, and its output is a scalar.
+First of all we will define the `Egg Holder` function we are going to build a surrogate for. Notice, one how its argument is a vector of numbers, one for each coordinate, and its output is a scalar.
 
 ```@example linear_surrogateND
 using Plots # hide
@@ -104,7 +104,7 @@ plot(p1, p2, title="True function") # hide
 ```
 
 ### Building a surrogate
-Using the sampled points we build the surrogate, the steps are analogous to the 1-dimensional case.
+Using the sampled points, we build the surrogate, the steps are analogous to the 1-dimensional case.
 
 ```@example linear_surrogateND
 my_linear_ND = LinearSurrogate(xys, zs,  lower_bound, upper_bound)
@@ -119,7 +119,7 @@ plot(p1, p2, title="Surrogate") # hide
 ```
 
 ### Optimizing
-With our surrogate we can now search for the minima of the function.
+With our surrogate, we can now search for the minima of the function.
 
 Notice how the new sampled points, which were created during the optimization process, are appended to the `xys` array.
 This is why its size changes.

docs/src/Salustowicz.md

@@ -39,7 +39,7 @@ scatter(x, y, label="Sampled points", xlims=(lower_bound, upper_bound), legend=:
 plot!(xs, salustowicz.(xs), label="True function", legend=:top)
 ```
 
-Now, let's fit Salustowicz Function with different Surrogates:
+Now, let's fit the Salustowicz function with different surrogates:
 
 ```@example salustowicz1D
 InverseDistance = InverseDistanceSurrogate(x, y, lower_bound, upper_bound)

docs/src/abstractgps.md

@@ -1,12 +1,12 @@
 # Gaussian Process Surrogate Tutorial
 
 !!! note
-    This surrogate requires the 'SurrogatesAbstractGPs' module which can be added by inputting "]add SurrogatesAbstractGPs" from the Julia command line. 
+    This surrogate requires the 'SurrogatesAbstractGPs' module, which can be added by inputting "]add SurrogatesAbstractGPs" from the Julia command line. 
 
 Gaussian Process regression in Surrogates.jl is implemented as a simple wrapper around the [AbstractGPs.jl](https://github.com/JuliaGaussianProcesses/AbstractGPs.jl) package. AbstractGPs comes with a variety of covariance functions (kernels). See [KernelFunctions.jl](https://github.com/JuliaGaussianProcesses/KernelFunctions.jl/) for examples.
 
 !!! tip
-    The examples below demonstrate the use of AbstractGPs with out-of-the-box settings without hyperparameter optimization (i.e. without changing parameters like lengthscale, signal variance and noise variance.) Beyond hyperparameter optimization, careful initialization of hyperparameters and priors on the parameters is required for this surrogate to work properly. For more details on how to fit GPs in practice, check out [A Practical Guide to Gaussian Processes](https://infallible-thompson-49de36.netlify.app/).
+    The examples below demonstrate the use of AbstractGPs with out-of-the-box settings without hyperparameter optimization (i.e. without changing parameters like lengthscale, signal variance, and noise variance). Beyond hyperparameter optimization, careful initialization of hyperparameters and priors on the parameters is required for this surrogate to work properly. For more details on how to fit GPs in practice, check out [A Practical Guide to Gaussian Processes](https://infallible-thompson-49de36.netlify.app/).
 
     Also see this [example](https://juliagaussianprocesses.github.io/AbstractGPs.jl/stable/examples/1-mauna-loa/#Hyperparameter-Optimization) to understand hyperparameter optimization with AbstractGPs.
 ## 1D Example 

docs/src/ackley.md

@@ -64,4 +64,4 @@ plot!(xs, ackley.(xs), label="True function", legend=:top)
 plot!(xs, my_rad.(xs), label="Radial basis optimized", legend=:top)
 ```
 
-The DYCORS methods successfully finds the minimum.
+The DYCORS method successfully finds the minimum.

docs/src/cantilever.md

@@ -42,7 +42,7 @@ plot(p1, p2, title="True function")
 ```
 
 
-Fitting different Surrogates:
+Fitting different surrogates:
 ```@example beam
 mypoly = PolynomialChaosSurrogate(xys, zs,  lb, ub)
 loba = LobachevskySurrogate(xys, zs,  lb, ub)

docs/src/gek.md

@@ -1,6 +1,6 @@
 ## Gradient Enhanced Kriging
 
-Gradient-enhanced Kriging is an extension of kriging which supports gradient information. GEK is usually more accurate than kriging, however, it is not computationally efficient when the number of inputs, the number of sampling points, or both, are high. This is mainly due to the size of the corresponding correlation matrix that increases proportionally with both the number of inputs and the number of sampling points.
+Gradient-enhanced Kriging is an extension of kriging which supports gradient information. GEK is usually more accurate than kriging. However, it is not computationally efficient when the number of inputs, the number of sampling points, or both, are high. This is mainly due to the size of the corresponding correlation matrix, which increases proportionally with both the number of inputs and the number of sampling points.
 
 Let's have a look at the following function to use Gradient Enhanced Surrogate:
 ``f(x) = sin(x) + 2*x^2``
@@ -15,7 +15,7 @@ default()
 
 ### Sampling
 
-We choose to sample f in 8 points between 0 to 1 using the `sample` function. The sampling points are chosen using a Sobol sequence, this can be done by passing `SobolSample()` to the `sample` function.
+We choose to sample f in 8 points between 0 and 1 using the `sample` function. The sampling points are chosen using a Sobol sequence, this can be done by passing `SobolSample()` to the `sample` function.
 
 ```@example GEK1D
 n_samples = 10
@@ -34,7 +34,7 @@ plot!(f, label="True function", xlims=(lower_bound, upper_bound), legend=:top)
 
 ### Building a surrogate
 
-With our sampled points we can build the Gradient Enhanced Kriging surrogate using the `GEK` function.
+With our sampled points, we can build the Gradient Enhanced Kriging surrogate using the `GEK` function.
 
 ```@example GEK1D
 
@@ -47,7 +47,7 @@ plot!(my_gek, label="Surrogate function", ribbon=p->std_error_at_point(my_gek, p
 
 ## Gradient Enhanced Kriging Surrogate Tutorial (ND)
 
-First of all let's define the function we are going to build a surrogate for.
+First of all, let's define the function we are going to build a surrogate for.
 
 ```@example GEK_ND
 using Plots # hide
@@ -69,7 +69,7 @@ end
 
 ### Sampling
 
-Let's define our bounds, this time we are working in two dimensions. In particular we want our first dimension `x` to have bounds `0, 10`, and `0, 10` for the second dimension. We are taking 80 samples of the space using Sobol Sequences. We then evaluate our function on all of the sampling points.
+Let's define our bounds, this time we are working in two dimensions. In particular, we want our first dimension `x` to have bounds `0, 10`, and `0, 10` for the second dimension. We are taking 80 samples of the space using Sobol Sequences. We then evaluate our function on all the sampling points.
 
 ```@example GEK_ND
 n_samples = 45
@@ -91,7 +91,7 @@ plot(p1, p2, title="True function") # hide
 ```
 
 ### Building a surrogate
-Using the sampled points we build the surrogate, the steps are analogous to the 1-dimensional case.
+Using the sampled points, we build the surrogate, the steps are analogous to the 1-dimensional case.
 
 ```@example GEK_ND
 grad1 = x1 -> 2*(300*(x[1])^5 - 300*(x[1])^2*x[2] + x[1] -1)

docs/src/gekpls.md

@@ -1,6 +1,6 @@
 ## GEKPLS Surrogate Tutorial
 
-Gradient Enhanced Kriging with Partial Least Squares Method (GEKPLS) is a surrogate modelling technique that brings down computation time and returns improved accuracy for high-dimensional problems. The Julia implementation of GEKPLS is adapted from the Python version by [SMT](https://github.com/SMTorg) which is based on this [paper](https://arxiv.org/pdf/1708.02663.pdf).  
+Gradient Enhanced Kriging with Partial Least Squares Method (GEKPLS) is a surrogate modeling technique that brings down computation time and returns improved accuracy for high-dimensional problems. The Julia implementation of GEKPLS is adapted from the Python version by [SMT](https://github.com/SMTorg) which is based on this [paper](https://arxiv.org/pdf/1708.02663.pdf).  
 
 The following are the inputs when building a GEKPLS surrogate: 
 

docs/src/gramacylee.md

@@ -1,6 +1,6 @@
 ## Gramacy & Lee Function
 
-Gramacy & Lee Function is a continuous function. It is not convex. The function is defined on 1-dimensional space. It is an unimodal. The function can be defined on any input domain but it is usually evaluated on
+the Gramacy & Lee function is a continuous function. It is not convex. The function is defined on a 1-dimensional space. It is unimodal. The function can be defined on any input domain, but it is usually evaluated on
 ``x \in [-0.5, 2.5]``.
 
 The Gramacy & Lee is as follows:
@@ -25,7 +25,7 @@ function gramacylee(x)
 end
 ```
 
-Let's sample f in 25 points between -0.5 and 2.5 using the `sample` function. The sampling points are chosen using a Sobol Sample, this can be done by passing `SobolSample()` to the `sample` function.
+Let's sample f in 25 points between -0.5 and 2.5 using the `sample` function. The sampling points are chosen using a Sobol sample, this can be done by passing `SobolSample()` to the `sample` function.
 
 ```@example gramacylee1D
 n = 25
@@ -38,7 +38,7 @@ scatter(x, y, label="Sampled points", xlims=(lower_bound, upper_bound), ylims=(-
 plot!(xs, gramacylee.(xs), label="True function", legend=:top)
 ```
 
-Now, let's fit Gramacy & Lee Function with different Surrogates:
+Now, let's fit Gramacy & Lee function with different surrogates:
 
 ```@example gramacylee1D
 my_pol = PolynomialChaosSurrogate(x, y, lower_bound, upper_bound)

docs/src/index.md

@@ -8,7 +8,7 @@ The construction of a surrogate model can be seen as a three-step process:
 2. Construction of the surrogate model
 3. Surrogate optimization
 
-The sampling methods are super important for the behavior of the Surrogate. Sampling can be done through [QuasiMonteCarlo.jl](https://github.com/SciML/QuasiMonteCarlo.jl), all the functions available there can be used in Surrogates.jl.
+The sampling methods are super important for the behavior of the surrogate. Sampling can be done through [QuasiMonteCarlo.jl](https://github.com/SciML/QuasiMonteCarlo.jl), all the functions available there can be used in Surrogates.jl.
 
 The available surrogates are:
 
@@ -27,7 +27,7 @@ That is, simultaneously looking for a minimum **and** sampling the most unknown
 The available optimization methods are:
 
 - Stochastic RBF (SRBF)
-- Lower confidence bound strategy (LCBS)
+- Lower confidence-bound strategy (LCBS)
 - Expected improvement (EI)
 - Dynamic coordinate search (DYCORS)