Automatic Regularization Selection (#68)

* doc typo fixes

* small typo fixed in inputs tutorial

* start regularization euler test case

* NullTransformer, ShiftedTransformer

* redo TestBaseTransformer, add TestNullTransformer

* TestShiftScaleTransformer simplified, uses _TestTransformer

* ScaleTransformer implementation and tests, need docs

* ShiftTransformer, ShiftScaleTransformer docs

* TransformerPipeline

* standardize pre __str__()s, update pre api walkthrough

* test coverage

* refactor ROM.fit to prep for fit_regselect

* gridsearch + optimization utility

* gridsearch + optimization utility

* small tweaks for print formatting

* automatic reg selection for nonparametric

* rename unfinished Euler example

* refactor regselect into base

* regularization extra test cases

* fix single-input edge cases

* small typo in parametric tutorial

* add pytest execution clause to all test files

* test / fix RegTest classes

* test and fix nonparametric regselect

* lightly test parametric regselect

* option to buffer TimedBlock messages

* TikhonovSolver.get_operator_regularizer()

* update changelog, version 0.5.8 -> 0.5.9

Author: shanemcq18

.gitignore

@@ -54,3 +54,5 @@ html/
 *.swp
 .markdownlint.json
 Notes/
+tests/add_endclause.py
+docs/use_mathjax_shortcuts.py

docs/_config.yml

@@ -98,6 +98,7 @@ sphinx:
           "B": "\\mathbf{B}"                  # input operator
           "N": "\\mathbf{N}"                  # bilinear operator
           "v": "\\mathbf{v}"                  # basis vector
+          "w": "\\mathbf{w}"                  # weight vector
           "V": "\\mathbf{V}"                  # generic basis matrix
           "W": "\\mathbf{W}"                  # weight matrix
           "Vr": "\\mathbf{V}_{\\!r}"          # trial basis matrix

docs/source/api/lstsq.ipynb

@@ -49,7 +49,7 @@
    "source": [
     ":::{admonition} Example Data\n",
     ":class: tip\n",
-    "The examples on this page use data matrices composed of compressed heat flow data from [tutorial 2](../tutorials/heat_equation.ipynb).\n",
+    "The examples on this page use data matrices composed of compressed heat flow data.\n",
     "You can [download the data here](https://github.com/Willcox-Research-Group/rom-operator-inference-Python3/raw/data/lstsq_example.npz) to repeat the demonstration.\n",
     ":::"
    ]

docs/source/api/main.md

@@ -1,11 +1,9 @@
 # `opinf`
 
-The `opinf` package containts the following submodules.
+The `opinf` package contains the following submodules.
 
 ```{eval-rst}
 
-.. currentmodule:: opinf
-
 .. autosummary::
 
    opinf.lift
@@ -20,4 +18,4 @@ The `opinf` package containts the following submodules.
    opinf.utils
 ```
 
-The main namespace also contains the [`ROM`](opinf.roms.ROM) class from {mod}`opinf.roms`.
+The main namespace also contains the [`ROM`](opinf.roms.ROM) and [`ParametricROM`](opinf.roms.ParametricROM) classes from {mod}`opinf.roms`.

docs/source/api/missing.rst

@@ -28,11 +28,15 @@ pre.ipynb
    :toctree: _autosummaries
    :nosignatures:
 
-   shift
-   scale
+   TransformerTemplate
+   NullTransformer
+   ShiftTransformer
+   ScaleTransformer
    ShiftScaleTransformer
+   TransformerPipeline
    TransformerMulti
-   TransformerTemplate
+   shift
+   scale
 
 basis.ipynb
 -----------