Rename UnivDist to Marginal across codebase

CHANGELOG.md

@@ -23,6 +23,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Changed
 
+- The class `UnivDist` has been renamed to `Marginal`. The name more clearly
+  refers to one-dimensional marginal distributions (of a univariate random
+  variable), which form a `ProbInput`.
 - The property `spatial_dimension` of `ProbInput` and `UQTestFunBareABC` is
   renamed to `input_dimension` for clarity (as opposed to `output_dimension`).
 - The property `name` of UQ test function instances has been renamed to

docs/_toc.yml

@@ -136,27 +136,27 @@ parts:
       - file: prob-input/probabilistic-input
         title: Probabilistic Input Model
       - file: prob-input/available
-        title: Univariate Distributions
+        title: One-Dimensional Marginal Distributions
         sections:
-          - file: prob-input/univariate-distributions/beta
+          - file: prob-input/marginal-distributions/beta
             title: Beta
-          - file: prob-input/univariate-distributions/exponential
+          - file: prob-input/marginal-distributions/exponential
             title: Exponential
-          - file: prob-input/univariate-distributions/gumbel
+          - file: prob-input/marginal-distributions/gumbel
             title: Gumbel (max.)
-          - file: prob-input/univariate-distributions/logitnormal
+          - file: prob-input/marginal-distributions/logitnormal
             title: Logit-Normal
-          - file: prob-input/univariate-distributions/lognormal
+          - file: prob-input/marginal-distributions/lognormal
             title: Log-Normal
-          - file: prob-input/univariate-distributions/normal
+          - file: prob-input/marginal-distributions/normal
             title: Normal (Gaussian)
-          - file: prob-input/univariate-distributions/triangular
+          - file: prob-input/marginal-distributions/triangular
             title: Triangular
-          - file: prob-input/univariate-distributions/trunc-gumbel
+          - file: prob-input/marginal-distributions/trunc-gumbel
             title: Truncated Gumbel (max.)
-          - file: prob-input/univariate-distributions/trunc-normal
+          - file: prob-input/marginal-distributions/trunc-normal
             title: Truncated Normal (Gaussian)
-          - file: prob-input/univariate-distributions/uniform
+          - file: prob-input/marginal-distributions/uniform
             title: Uniform
   - caption: API Reference
     chapters:
@@ -170,8 +170,8 @@ parts:
         title: UQTestFun
       - file: api/probinput
         title: ProbInput
-      - file: api/univdist
-        title: UnivDist
+      - file: api/marginal
+        title: Marginal
       - file: api/input-spec
         title: ProbInputSpec
       - file: api/funparams
@@ -188,8 +188,8 @@ parts:
         title: Test Function Docs
       - file: development/making-a-pull-request
         title: Making a Pull Request
-      - file: development/adding-univariate-distribution
-        title: Univariate Distribution
+      - file: development/adding-marginal-distribution
+        title: Adding a Marginal Distribution
       - file: development/about
         title: About UQTestFuns
       - file: development/code-of-conduct

docs/api/marginal.rst

@@ -0,0 +1,9 @@
+.. _api_reference_marginal_distribution:
+
+One-Dimensional Marginal Distribution
+=====================================
+
+.. automodule:: uqtestfuns.core.prob_input.marginal
+
+.. autoclass:: uqtestfuns.core.prob_input.marginal.Marginal
+   :members:

docs/api/overview.md

@@ -21,14 +21,13 @@ let's start from the top, the {ref}`built-in test functions <test-functions:avai
   represents the input of an uncertainty quantification (UQ) test function.
 - An instance of the ``ProbInput`` class consists mainly of the one-dimensional
   marginals and a copula specification (not yet supported). Each one-dimensional
-  marginal comes from a univariate random variable which in turn
-  is represented by the {ref}`UnivDist <api_reference_univariate_distribution>`
-  class. 
-- An instance of class ``UnivDist`` has a (parametric) probability distribution.
-  Although different instances may have different probability distributions,
-  they are all instances of the same class.
-- As lightweight containers to specify the specifications of a ``ProbInput`` and
-  a ``UnivDist``, {ref}`three custom <api_reference_input_spec>` ``NamedTuple``
+  marginal comes is represented
+  by the {ref}`Marginal <api_reference_marginal_distribution>` class. 
+- An instance of the ``Marginal`` class has a (parametric) probability
+  distribution. Although different instances may have different
+  probability distributions, they are all instances of the same class.
+- As lightweight containers to specify the specifications of a ``ProbInput``
+  and a ``Marginal``, {ref}`three custom <api_reference_input_spec>` ``NamedTuple``
   are defined, namely {ref}`api_reference_input_spec_univdist`, 
   {ref}`api_reference_input_spec_fixdim`, and {ref}`api_reference_input_spec_vardim`.
 

docs/development/adding-univariate-distribution.md → docs/development/adding-marginal-distribution.md

@@ -1,8 +1,9 @@
-(development:adding-univ-dist)=
-# Adding a New Univariate Distribution Type
+(development:adding-marginal-distribution)=
+# Adding a New Marginal Distribution Type
 
-UQTestFuns is delivered with several {ref}`univariate distributions <prob-input:available-univariate-distributions>`
-that are used by the currently available test functions.
+UQTestFuns is delivered with several one-dimensional
+{ref}`marginal distributions <prob-input:available-marginal-distributions>`
+(i.e., univariate distributions) that are used by the currently available test functions.
 In case a new univariate distribution type is required for a new test function,
 the distribution should be added first to the code base.
 
@@ -27,12 +28,12 @@ For example:
 
 ```python
 >>> import uqtestfuns as uqtf
->>> my_var = uqtf.UnivDist(distribution="uniform", parameters=[3, 5])
+>>> my_var = uqtf.Marginal(distribution="uniform", parameters=[3, 5])
 ```
 
 ## Step 0: Putting things in the right place
 
-Univariate random variables in UQTestFuns are represented as instances of the `UnivDist` class.
+Univariate random variables in UQTestFuns are represented as instances of the `Marginal` class.
 The supported distributions for the random variables are implemented in separate modules
 located in the `src/uqtestfuns/core/prob_input/univariate_distributions` directory.
 Here's how the directory look:
@@ -361,11 +362,11 @@ SUPPORTED_MARGINALS = {
 }
 ```
 
-This way you can create a new instance of `UnivDist` class by passing the chosen name as the `distribution` and the corresponding parameters:
+This way you can create a new instance of `Marginal` class by passing the chosen name as the `distribution` and the corresponding parameters:
 
 ```python
 >>> import uqtestfuns as uqtf
->>> my_var = uqtf.UnivDist(distribution="uniform", parameters=[3, 5])
+>>> my_var = uqtf.Marginal(distribution="uniform", parameters=[3, 5])
 ```
 
 Remember that the name `uniform` was chosen as the name of this distribution via the module-level variable `DISTRIBUTION_NAME`.

docs/development/adding-test-function-implementation.md

@@ -124,7 +124,7 @@ Here are some explanations:
   (the one-dimensional marginal specification) and `ProbInputSpecFixDim`
   (the probabilistic input model). These are lightweight _containers_
   (meaning no custom methods) of all the information required to construct,
-  later on, `UnivDist` and `ProbInput` instances, respectively.
+  later on, `Marginal` and `ProbInput` instances, respectively.
 
 ```{note}
 There is also the corresponding `ProbInputSpecVarDim` to store the information
@@ -221,7 +221,7 @@ The keyword is, by convention, the citation key of the particular reference as l
 
 Also notice that in the code above, we store the specifications of the marginals
 in a list of {ref}`api_reference_input_spec_univdist`.
-Each element of this list is used to create an instance of `UnivDist`.
+Each element of this list is used to create an instance of `Marginal`.
 
 With that, we have completed the input specification of the Branin function.
 

docs/development/overview.md

@@ -45,7 +45,8 @@ be sure to check out the guides on:
 - {ref}`adding a new test function implementation <development:adding-test-function-implementation>`
 - {ref}`adding a new test function documentation <development:adding-test-function-docs>`
 - {ref}`making a pull request <development:making-a-pull-request>`
-- In case the univariate distribution types are not yet available: {ref}`adding a new univariate distribution <development:adding-univ-dist>`
+- In case the univariate distribution types are not yet available:
+  {ref}`adding a new univariate distribution <development:adding-marginal-distribution>`
 
 ## Updating the documentation
 

docs/getting-started/tutorial-custom-functions.md

@@ -146,8 +146,8 @@ class  and an instance of it can be defined as follows:
 ```{code-cell} ipython3
 # Define a list of marginals
 marginals = [
-    uqtf.UnivDist(distribution="uniform", parameters=[-5, 10], name="x1"),
-    uqtf.UnivDist(distribution="uniform", parameters=[0, 15], name="x2"),
+    uqtf.Marginal(distribution="uniform", parameters=[-5, 10], name="x1"),
+    uqtf.Marginal(distribution="uniform", parameters=[0, 15], name="x2"),
 ]
 # Create a probabilistic input
 my_input = uqtf.ProbInput(marginals=marginals, name="Branin-Input")

docs/prob-input/available.md

@@ -1,19 +1,20 @@
-(prob-input:available-univariate-distributions)=
-# Available Univariate Distributions
+(prob-input:available-marginal-distributions)=
+# Available One-Dimensional Marginal Distributions
 
-The table below lists all the available univariate distribution types used
-to construct ``UnivDist`` instances. ``UnivDist`` are used to represent the
-univariate marginals of a (possibly, multivariate) probabilistic input model.
+The table below lists all the available one-dimensional marginal distribution
+types used to construct ``Marginal`` instances.
+``Marginal`` instances are used to represent the one-dimensional marginals
+of a (possibly, multivariate) probabilistic input model.
 
-|                                         Name                                          | Keyword value for `distribution` |                     Notation                      |             Support              | Number of parameters |
-|:-------------------------------------------------------------------------------------:|:--------------------------------:|:-------------------------------------------------:|:--------------------------------:|:--------------------:|
-|                {ref}`Beta <prob-input:univariate-distributions:beta>`                 |             `"beta"`             |       $\mathrm{Beta}(\alpha, \beta, a, b)$        | $[a, b], \; a, b \in \mathbb{R}$ |          4           |
-|         {ref}`Exponential <prob-input:univariate-distributions:exponential>`          |         `"exponential"`          |              $\mathcal{E}(\lambda)$               |          $[0, \infty)$           |          1           |
-|           {ref}`Gumbel (max.) <prob-input:univariate-distributions:gumbel>`           |            `"gumbel"`            |           $\mathrm{Gumbel}(\mu, \beta)$           |       $(-\infty, \infty)$        |          2           |
-|         {ref}`Logit-Normal <prob-input:univariate-distributions:logitnormal>`         |         `"logitnormal"`          |    $\mathcal{N}_{\mathrm{logit}}(\mu, \sigma)$    |             $(0, 1)$             |          2           |
-|           {ref}`Log-Normal <prob-input:univariate-distributions:lognormal>`           |          `"lognormal"`           |    $\mathcal{N}_{\mathrm{log}} (\mu, \sigma)$     |          $(0, \infty)$           |          2           |
-|         {ref}`Normal (Gaussian) <prob-input:univariate-distributions:normal>`         |            `"normal"`            |            $\mathcal{N}(\mu, \sigma)$             |       $(-\infty, \infty)$        |          2           |
-|          {ref}`Triangular <prob-input:univariate-distributions:triangular>`           |          `"triangular"`          |             $\mathcal{T}_r(a, b, c)$              | $[a, b], \; a, b \in \mathbb{R}$ |          3           |
-|   {ref}`Truncated Gumbel (max.) <prob-input:univariate-distributions:trunc-gumbel>`   |         `"trunc-gumbel"`         | $\mathrm{Gumbel}_{\mathrm{Tr}}(\mu, \beta, a, b)$ | $[a, b], \; a, b \in \mathbb{R}$ |          4           |
-| {ref}`Truncated Normal (Gaussian) <prob-input:univariate-distributions:trunc-normal>` |         `"trunc-normal"`         |  $\mathcal{N}_{\mathrm{Tr}}(\mu, \sigma, a, b)$   | $[a, b], \; a, b \in \mathbb{R}$ |          4           |
-|             {ref}`Uniform <prob-input:univariate-distributions:uniform>`              |           `"uniform"`            |                $\mathcal{U}(a, b)$                | $[a, b], \; a, b \in \mathbb{R}$ |          2           |
+|                                        Name                                         | Keyword value for `distribution` |                     Notation                      |             Support              | Number of parameters |
+|:-----------------------------------------------------------------------------------:|:--------------------------------:|:-------------------------------------------------:|:--------------------------------:|:--------------------:|
+|                {ref}`Beta <prob-input:marginal-distributions:beta>`                 |             `"beta"`             |       $\mathrm{Beta}(\alpha, \beta, a, b)$        | $[a, b], \; a, b \in \mathbb{R}$ |          4           |
+|         {ref}`Exponential <prob-input:marginal-distributions:exponential>`          |         `"exponential"`          |              $\mathcal{E}(\lambda)$               |          $[0, \infty)$           |          1           |
+|           {ref}`Gumbel (max.) <prob-input:marginal-distributions:gumbel>`           |            `"gumbel"`            |           $\mathrm{Gumbel}(\mu, \beta)$           |       $(-\infty, \infty)$        |          2           |
+|         {ref}`Logit-Normal <prob-input:marginal-distributions:logitnormal>`         |         `"logitnormal"`          |    $\mathcal{N}_{\mathrm{logit}}(\mu, \sigma)$    |             $(0, 1)$             |          2           |
+|           {ref}`Log-Normal <prob-input:marginal-distributions:lognormal>`           |          `"lognormal"`           |    $\mathcal{N}_{\mathrm{log}} (\mu, \sigma)$     |          $(0, \infty)$           |          2           |
+|         {ref}`Normal (Gaussian) <prob-input:marginal-distributions:normal>`         |            `"normal"`            |            $\mathcal{N}(\mu, \sigma)$             |       $(-\infty, \infty)$        |          2           |
+|          {ref}`Triangular <prob-input:marginal-distributions:triangular>`           |          `"triangular"`          |             $\mathcal{T}_r(a, b, c)$              | $[a, b], \; a, b \in \mathbb{R}$ |          3           |
+|   {ref}`Truncated Gumbel (max.) <prob-input:marginal-distributions:trunc-gumbel>`   |         `"trunc-gumbel"`         | $\mathrm{Gumbel}_{\mathrm{Tr}}(\mu, \beta, a, b)$ | $[a, b], \; a, b \in \mathbb{R}$ |          4           |
+| {ref}`Truncated Normal (Gaussian) <prob-input:marginal-distributions:trunc-normal>` |         `"trunc-normal"`         |  $\mathcal{N}_{\mathrm{Tr}}(\mu, \sigma, a, b)$   | $[a, b], \; a, b \in \mathbb{R}$ |          4           |
+|             {ref}`Uniform <prob-input:marginal-distributions:uniform>`              |           `"uniform"`            |                $\mathcal{U}(a, b)$                | $[a, b], \; a, b \in \mathbb{R}$ |          2           |

docs/prob-input/univariate-distributions/beta.md → docs/prob-input/marginal-distributions/beta.md

@@ -12,7 +12,7 @@ kernelspec:
   name: python3
 ---
 
-(prob-input:univariate-distributions:beta)=
+(prob-input:marginal-distributions:beta)=
 # Beta Distribution
 
 The Beta distribution is a four-parameter continuous probability distribution.
@@ -83,7 +83,7 @@ colors = ["#a6cee3", "#1f78b4", "#b2df8a", "#33a02c"]
 
 univ_dists = []
 for parameter in parameters:
-    univ_dists.append(uqtf.UnivDist(distribution="beta", parameters=parameter))
+    univ_dists.append(uqtf.Marginal(distribution="beta", parameters=parameter))
 
 fig, axs = plt.subplots(2, 2, figsize=(10,10))
 

docs/prob-input/univariate-distributions/exponential.md → docs/prob-input/marginal-distributions/exponential.md

@@ -20,7 +20,7 @@ import matplotlib.pyplot as plt
 import numpy as np
 ```
 
-(prob-input:univariate-distributions:exponential)=
+(prob-input:marginal-distributions:exponential)=
 # Exponential Distribution
 
 The exponential distribution is a single-parameter continuous probability
@@ -55,7 +55,7 @@ colors = ["#a6cee3", "#1f78b4", "#b2df8a", "#33a02c"]
 
 univ_dists = []
 for parameter in parameters:
-    univ_dists.append(uqtf.UnivDist(distribution="exponential", parameters=parameter))
+    univ_dists.append(uqtf.Marginal(distribution="exponential", parameters=parameter))
 
 fig, axs = plt.subplots(2, 2, figsize=(10,10))
 

docs/prob-input/univariate-distributions/gumbel.md → docs/prob-input/marginal-distributions/gumbel.md

@@ -20,7 +20,7 @@ import matplotlib.pyplot as plt
 import numpy as np
 ```
 
-(prob-input:univariate-distributions:gumbel)=
+(prob-input:marginal-distributions:gumbel)=
 # Gumbel (max.) Distribution
 
 The Gumbel (max.) distribution is a two-parameter continuous probability distribution.
@@ -54,7 +54,7 @@ colors = ["#a6cee3", "#1f78b4", "#b2df8a", "#33a02c"]
 
 univ_dists = []
 for parameter in parameters:
-    univ_dists.append(uqtf.UnivDist(distribution="gumbel", parameters=parameter))
+    univ_dists.append(uqtf.Marginal(distribution="gumbel", parameters=parameter))
 
 fig, axs = plt.subplots(2, 2, figsize=(10,10))
 

docs/prob-input/univariate-distributions/logitnormal.md → docs/prob-input/marginal-distributions/logitnormal.md

@@ -20,12 +20,12 @@ import matplotlib.pyplot as plt
 import numpy as np
 ```
 
-(prob-input:univariate-distributions:logitnormal)=
+(prob-input:marginal-distributions:logitnormal)=
 # Logit-Normal Distribution
 
 The logit-normal distribution is a two-parameter continuous probability distribution.
 A logit-normal random variable is a variable whose _logit_ is
-a {ref}`normally distributed <prob-input:univariate-distributions:normal>` 
+a {ref}`normally distributed <prob-input:marginal-distributions:normal>` 
 random variable.
 
 ```{admonition} Logit and logistic function
@@ -85,7 +85,7 @@ colors = ["#a6cee3", "#1f78b4", "#b2df8a", "#33a02c"]
 univ_dists = []
 for parameter in parameters:
     univ_dists.append(
-    uqtf.UnivDist(distribution="logitnormal", parameters=parameter)
+    uqtf.Marginal(distribution="logitnormal", parameters=parameter)
     )
 
 fig, axs = plt.subplots(2, 2, figsize=(10,10))

docs/prob-input/univariate-distributions/lognormal.md → docs/prob-input/marginal-distributions/lognormal.md

@@ -12,12 +12,12 @@ kernelspec:
   name: python3
 ---
 
-(prob-input:univariate-distributions:lognormal)=
+(prob-input:marginal-distributions:lognormal)=
 # Log-Normal Distribution
 
 The log-normal distribution is a two-parameter continuous probability distribution.
 A log-normal random variable is a variable whose (natural) logarithm is 
-a {ref}`normally distributed <prob-input:univariate-distributions:normal>`
+a {ref}`normally distributed <prob-input:marginal-distributions:normal>`
 random variable.
 The table below summarizes some important aspects of the distribution.
 
@@ -60,7 +60,7 @@ colors = ["#a6cee3", "#1f78b4", "#b2df8a", "#33a02c"]
 univ_dists = []
 for parameter in parameters:
     univ_dists.append(
-        uqtf.UnivDist(distribution="lognormal", parameters=parameter)
+        uqtf.Marginal(distribution="lognormal", parameters=parameter)
     )
 
 fig, axs = plt.subplots(2, 2, figsize=(10,10))

docs/prob-input/univariate-distributions/normal.md → docs/prob-input/marginal-distributions/normal.md

@@ -20,7 +20,7 @@ import matplotlib.pyplot as plt
 import numpy as np
 ```
 
-(prob-input:univariate-distributions:normal)=
+(prob-input:marginal-distributions:normal)=
 # Normal (Gaussian) Distribution
 
 The normal (or Gaussian) distribution is a two-parameter continuous probability
@@ -66,7 +66,7 @@ colors = ["#a6cee3", "#1f78b4", "#b2df8a", "#33a02c"]
 
 univ_dists = []
 for parameter in parameters:
-    univ_dists.append(uqtf.UnivDist(distribution="normal", parameters=parameter))
+    univ_dists.append(uqtf.Marginal(distribution="normal", parameters=parameter))
 
 fig, axs = plt.subplots(2, 2, figsize=(10,10))