Preliminary work on installation/doc overhaul

.github/workflows/lint.yml

@@ -20,11 +20,10 @@ jobs:
         python-version: 3.9
         cache: 'pip'  # optional, adds caching for faster builds
 
-    - name: Install dependencies
+    - name: Install UncertainSCI
       run: |
         python -m pip install --upgrade pip
-        pip install flake8
-        pip install -r requirements.txt
+        pip install --editable .["dev"]
 
     - name: Lint with flake8
       run: |

.github/workflows/test.yml

@@ -27,15 +27,10 @@ jobs:
     - name: Display Python version
       run: python -c "import sys; print(sys.version)"
 
-    - name: Install dependencies
-      run: |
-        python -m pip install --upgrade pip
-        pip install pytest pytest-cov
-        pip install -r requirements.txt
-
     - name: Install UncertainSCI
       run: |
-        pip install --editable ./
+        python -m pip install --upgrade pip
+        pip install --editable .["dev"]
 
     - name: Run tests and compute coverage
       run: |

.gitignore

@@ -70,6 +70,7 @@ instance/
 
 # Sphinx documentation
 docs/_build/
+docs/**/generated/
 
 # PyBuilder
 target/

.readthedocs.yaml

@@ -10,6 +10,8 @@ build:
   os: ubuntu-22.04
   tools:
     python: "3.10"
+  apt_packages:
+    - pandoc
 
 # Build documentation in the docs/ directory with Sphinx
 sphinx:
@@ -23,6 +25,7 @@ formats: all
 # Optionally declare the Python requirements required to build your docs
 python:
   install:
-    - requirements: requirements.txt
-    - requirements: docs/requirements.txt
-    - path: .
+    - method: pip
+      path: .
+      extra_requirements:
+        - docs

UncertainSCI/__init__.py

@@ -1 +1,3 @@
-
+"""
+A modern Python-based mathematical uncertainty toolkit.
+"""

UncertainSCI/distributions.py

@@ -1,3 +1,19 @@
+"""
+Classes and tools to define probability distributions.
+
+For PCE (:ref:`pce`), these distributions are evolved through a forward model defining
+the stochastic variation of model parameters.
+
+Tensorizations within a distribution are possible across these families by
+instantiating the distribution appropriately. Tensorizations across distributions
+is also possible but requires individual instantiation of each distribution, followed
+by a constructor call to the :class:`TensorialDistribution` class.
+
+For example, a three-dimensional random variable :math:`Y = (Y_1, Y_2, Y_3)` can have
+independent components, with the distribution of :math:`Y_1` normal, :math:`Y_2` beta,
+and :math:`Y_3` exponential.
+"""
+
 import numpy as np
 import scipy as sp
 from scipy import sparse as sprs
@@ -630,23 +646,19 @@ class BetaDistribution(ProbabilityDistribution):
     \\mu, \\sigma`.
 
     Parameters:
-        alpha (float or iterable of floats, optional): Shape parameter
-        associated to right-hand boundary. Defaults to 1.
-        beta (float or iterable of floats, optional): Shape parameter
-        associated to left-hand boundary. Defaults to 1.
-        mean (float or iterable of floats, optional): Mean of the distribution.
-        Defaults to None.
-        stdev (float or iterable of floats, optional): Standard deviation of
-        the distribution. Defaults to None.
-        dim (int, optional): Dimension of the distribution. Defaults to None.
-        domain (numpy.ndarray or similar, of size 2 x `dim`, optional): Compact
-        hypercube that is the support of the distribution. Defaults to None
+        alpha: Shape parameter associated to right-hand boundary. Defaults to 1.
+        beta: Shape parameter associated to left-hand boundary. Defaults to 1.
+        mean: Mean of the distribution. Defaults to None.
+        stdev: Standard deviation of the distribution. Defaults to None.
+        dim: Dimension of the distribution. Defaults to None.
+        domain: Compact hypercube of size 2 x ``dim`` that is the support of
+            the distribution. Defaults to None.
 
     Attributes:
-        dim (int): Dimension of the distribution.
-        alpha (float or np.ndarray): Shape parameter(s) alpha.
-        beta (float or np.ndarray): Shape parameter(s) beta.
-        polys (:class:`JacobiPolynomials` or list thereof):
+        dim: Dimension of the distribution.
+        alpha: Shape parameter(s) alpha.
+        beta: Shape parameter(s) beta.
+        polys: Jacobi polynomial family or list thereof.
     """
     def __init__(self, alpha=None, beta=None, mean=None, stdev=None,
                  dim=None, domain=None, bounds=None):
@@ -992,17 +1004,15 @@ class UniformDistribution(BetaDistribution):
     \\mu, \\sigma`.
 
     Parameters:
-        mean (float or iterable of floats, optional): Mean of the distribution.
-        Defaults to None.
-        stdev (float or iterable of floats, optional): Standard deviation of
-        the distribution. Defaults to None.
-        dim (int, optional): Dimension of the distribution. Defaults to None.
-        domain (numpy.ndarray or similar, of size 2 x `dim`, optional): Compact
-        hypercube that is the support of the distribution. Defaults to None
+        mean: Mean of the distribution. Defaults to None.
+        stdev: Standard deviation of the distribution. Defaults to None.
+        dim: Dimension of the distribution. Defaults to None.
+        domain: Compact hypercube of size 2 x ``dim`` that is the support of
+            the distribution. Defaults to None.
 
     Attributes:
-        dim (int): Dimension of the distribution.
-        polys (:class:`JacobiPolynomials` or list thereof):
+        dim: Dimension of the distribution.
+        polys: Jacobi polynomial family or list thereof.
     """
 
     def __init__(self, mean=None, stdev=None, dim=None, domain=None, bounds=None):

UncertainSCI/families.py

@@ -1,8 +1,6 @@
 """
-Contains routines that specialize opoly1d things for classical orthogonal polynomial families
-- jacobi polys
-- hermite poly
-- laguerre polys
+Routines that specialize :mod:`UncertainSCI.opoly1d` for classical families of
+orthogonal polynomials.
 """
 import numpy as np
 

UncertainSCI/indexing.py

@@ -1,3 +1,7 @@
+"""
+Module containing indexing tools and families.
+"""
+
 from itertools import combinations
 
 import numpy as np
@@ -463,8 +467,3 @@ def get_indices(self):
             return super().get_indices()
         else:
             return hyperbolic_cross_indices(self.dim, self.order)
-
-
-if __name__ == "__main__":
-
-    pass