Merge pull request #6 from derb12/development

Merge Development for v0.4.0

.bumpversion.cfg

@@ -0,0 +1,16 @@
+[bumpversion]
+current_version = 0.4.0
+commit = False
+tag = False
+
+[bumpversion:file:setup.cfg]
+search = version = {current_version}
+replace = version = {new_version}
+
+[bumpversion:file:pybaselines/__init__.py]
+search = __version__ = '{current_version}'
+replace = __version__ = '{new_version}'
+
+[bumpversion:file:docs/conf.py]
+search = version = '{current_version}'
+replace = version = '{new_version}'

CHANGELOG.rst

@@ -2,6 +2,72 @@
 Changelog
 =========
 
+Version 0.4.0 (2021-05-30)
+--------------------------
+
+This is a minor version with new features, bug fixes, and deprecations.
+
+New Features
+~~~~~~~~~~~~
+
+* Significantly reduced both the calculation time and memory usage of polynomial.loess.
+  For example, getting the baseline for a dataset with 20,000 points now takes ~12 seconds
+  and ~0.7 Gb of memory compared to ~55 seconds and ~3 Gb of memory in version 0.3.
+* Added a `conserve_memory` parameter to polynomial.loess that will recalculate the distance
+  kernels each iteration, which is slower than the default but uses very little memory. For
+  example, using loess with `conserve_memory` set to True on a dataset with 20,000 points
+  takes ~18 seconds while using ~0 Gb of memory.
+* Allow more user inputs for optimizers.optimize_extended_range to allow specifying the range
+  of `lam`/`poly_order` values to test and to have more control over the added lines and
+  Gaussians on the sides.
+* Added a constant called PERMC_SPEC (accessed from pybaselines.utils.PERMC_SPEC),
+  which is used by SciPy's sparse solver when using Whittaker-smoothing-based algorithms.
+  Changed the default value to "NATURAL", which reduced the computation time of all
+  Whittaker-smoothing-based algorithms by ~5-35% compared to other permc_spec options
+  on the tested system.
+* misc.interp_pts (formerly manual.linear_interp) now allows specifying any interpolation
+  method supported by scipy.interpolate.interp1d, allowing for methods such as spline
+  interpolation.
+
+Bug Fixes
+~~~~~~~~~
+
+* Fixed poly_order calculation for optimizers.adaptive_minmax when poly_order was a
+  single item within a container.
+* Potential fix for namespace error with utils; accessing pybaselines.utils gave an
+  attribute error in very specific envinronments, so changed the import order in
+  pybaselines.__init__ to potentially fix it. Updated the quick start example in case
+  the fix is not correct so that the example will still work.
+* Increased minimum NumPy version to 1.14 to use rcond=None with numpy.linalg.lstsq.
+
+Other Changes
+~~~~~~~~~~~~~
+
+* polynomial.loess now allows inputting weights, specifying a `use_original` keyword for
+  thresholding to match the modpoly and imodpoly functions, and specifying a `return_coef`
+  keyword to allow returning the polynomial coefficients for each x-value to recreate
+  the fitted polynomial, to match all other polynomial functions.
+* Changed the default `smooth_half_window` value in window.noise_median, window.snip, and
+  morphological.mormol to None, rather than being fixed values. Each function sets its default
+  slightly different but still follows the behavior in previous versions, except for
+  window.noise_median as noted below.
+* Changed default `smooth_half_window` value for window.noise_median to match specified
+  `half_window` value rather than 1.
+* Changed default `sigma` value for window.noise_median to scale with the specified
+  `smooth_half_window`, rather than being a fixed value.
+
+Deprecations/Breaking Changes
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+* Renamed pybaselines.manual to pybaselines.misc to allow for adding any future
+  miscellaneous algorithms that will not fit elsewhere.
+* Renamed the manual.linear_interp function to misc.interp_pts to reflect its more
+  general interpolation usage.
+* The parameter dictionary returned from Whittaker-smoothing-based functions
+  no longer includes 'roughness' and 'fidelity' values since the values were not used
+  elsewhere.
+
+
 Version 0.3.0 (2021-04-29)
 --------------------------
 

MANIFEST.in

@@ -1,10 +1,9 @@
 include LICENSE.txt
 include README.rst
 include LICENSES_bundled.txt
+include CHANGELOG.rst
 
-exclude .readthedocs.yaml
-exclude .gitattributes
-exclude .gitignore
+exclude .*
 
 recursive-exclude .git *
 recursive-exclude .github *

README.rst

@@ -19,7 +19,7 @@ pybaselines
     :alt: BSD 3-clause license
 
 
-pybaselines is a collection of baseline algorithms for fitting experimental data.
+pybaselines is a collection of algorithms for estimating the baseline of experimental data.
 
 * For Python 3.6+
 * Open Source: BSD 3-Clause License
@@ -83,9 +83,9 @@ e) Optimizers (pybaselines.optimizers)
     2) optimize_extended_range
     3) adaptive_minmax (Adaptive MinMax)
 
-f) Manual methods (pybaselines.manual)
+f) Miscellaneous methods (pybaselines.misc)
 
-    1) linear_interp (Linear interpolation between points)
+    1) interp_pts (Interpolation between points)
 
 
 Installation
@@ -97,8 +97,8 @@ Dependencies
 pybaselines requires `Python <https://python.org>`_ version 3.6 or later
 and the following libraries:
 
-* `NumPy <https://numpy.org>`_ (>= 1.9)
-* `SciPy <https://www.scipy.org/scipylib/index.html>`_
+* `NumPy <https://numpy.org>`_ (>= 1.14)
+* `SciPy <https://www.scipy.org/scipylib/index.html>`_ (>= 0.11)
 
 
 All of the required libraries should be automatically installed when
@@ -136,7 +136,7 @@ Once the repository is downloaded, it can be installed with:
 .. code-block:: console
 
     cd pybaselines
-    python setup.py install
+    pip install .
 
 
 .. _Github repo: https://github.com/derb12/pybaselines
@@ -160,23 +160,24 @@ A simple example is shown below.
 
 .. code-block:: python
 
-    import pybaselines
     import matplotlib.pyplot as plt
     import numpy as np
+    import pybaselines
+    from pybaselines import utils
 
     x = np.linspace(100, 4200, 1000)
     # a measured signal containing several Gaussian peaks
     signal = (
-        pybaselines.utils.gaussian(x, 2, 700, 50)
-        + pybaselines.utils.gaussian(x, 3, 1200, 150)
-        + pybaselines.utils.gaussian(x, 5, 1600, 100)
-        + pybaselines.utils.gaussian(x, 4, 2500, 50)
-        + pybaselines.utils.gaussian(x, 7, 3300, 100)
+        utils.gaussian(x, 2, 700, 50)
+        + utils.gaussian(x, 3, 1200, 150)
+        + utils.gaussian(x, 5, 1600, 100)
+        + utils.gaussian(x, 4, 2500, 50)
+        + utils.gaussian(x, 7, 3300, 100)
     )
     # baseline is a polynomial plus a broad gaussian
     true_baseline = (
         10 + 0.001 * x
-        + pybaselines.utils.gaussian(x, 6, 2000, 2000)
+        + utils.gaussian(x, 6, 2000, 2000)
     )
     noise = np.random.default_rng(1).normal(0, 0.2, x.size)
 

docs/_templates/autoapi/python/module.rst

@@ -7,12 +7,6 @@
 
 .. py:module:: {{ obj.name }}
 
-{% if obj.docstring %}
-.. autoapi-nested-parse::
-
-   {{ obj.docstring|prepare_docstring|indent(3) }}
-
-{% endif %}
 
 {% block subpackages %}
 {% set visible_subpackages = obj.subpackages|selectattr("display")|list %}

docs/_templates/autoapi/python/package.rst

@@ -0,0 +1,100 @@
+{% if not obj.display %}
+:orphan:
+
+{% endif %}
+{{ obj.name }}
+{{ "=" * obj.name|length }}
+
+.. py:module:: {{ obj.name }}
+
+{% if obj.docstring %}
+.. autoapi-nested-parse::
+
+   {{ obj.docstring|prepare_docstring|indent(3) }}
+
+{% endif %}
+
+{% block subpackages %}
+{% set visible_subpackages = obj.subpackages|selectattr("display")|list %}
+{% if visible_subpackages %}
+Subpackages
+-----------
+.. toctree::
+   :titlesonly:
+   :maxdepth: 3
+
+{% for subpackage in visible_subpackages %}
+   {{ subpackage.short_name }}/index.rst
+{% endfor %}
+
+
+{% endif %}
+{% endblock %}
+{% block submodules %}
+{% set visible_submodules = obj.submodules|selectattr("display")|list %}
+{% if visible_submodules %}
+Submodules
+----------
+
+.. toctree::
+   :titlesonly:
+   :maxdepth: 1
+
+{% for submodule in visible_submodules %}
+   {{ submodule.short_name }}/index.rst
+{% endfor %}
+
+
+{% endif %}
+{% endblock %}
+{% block content %}
+{% if obj.all is not none %}
+{% set visible_children = obj.children|selectattr("short_name", "in", obj.all)|list %}
+{% elif obj.type is equalto("package") %}
+{% set visible_children = obj.children|selectattr("display")|list %}
+{% else %}
+{% set visible_children = obj.children|selectattr("display")|rejectattr("imported")|list %}
+{% endif %}
+{% if visible_children %}
+{{ obj.type|title }} Contents
+{{ "-" * obj.type|length }}---------
+
+{% set visible_classes = visible_children|selectattr("type", "equalto", "class")|list %}
+{% set visible_functions = visible_children|selectattr("type", "equalto", "function")|list %}
+{% if "show-module-summary" in autoapi_options and (visible_classes or visible_functions) %}
+{% block classes scoped %}
+{% if visible_classes %}
+Classes
+~~~~~~~
+
+.. autoapisummary::
+
+{% for klass in visible_classes %}
+   {{ klass.id }}
+{% endfor %}
+
+
+{% endif %}
+{% endblock %}
+
+{% block functions scoped %}
+{% if visible_functions %}
+Functions
+~~~~~~~~~
+
+.. autoapisummary::
+   :nosignatures:
+
+{% for function in visible_functions %}
+   {{ function.id }}
+{% endfor %}
+
+
+{% endif %}
+{% endblock %}
+{% endif %}
+{% for obj_item in visible_children %}
+{{ obj_item.rendered|indent(0) }}
+{% endfor %}
+{% endif %}
+{% endblock %}

docs/algorithms/index.rst

@@ -39,4 +39,4 @@ are not fully optimized for the data).
    morphological
    window
    optimizers
-   manual
+   misc

docs/algorithms/misc.rst

@@ -0,0 +1,16 @@
+=======================
+Miscellaneous Baselines
+=======================
+
+The contents of :mod:`pybaselines.misc` contain miscellaneous baseline algorithms
+that do not fit in other categories.
+
+Algorithms
+----------
+
+interp_pts (Interpolation between points)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Interpolates between input points using line segments or splines of
+different orders. Mainly intended for usage with user interfaces and
+is not encouraged otherwise.

docs/algorithms/whittaker.rst

@@ -14,7 +14,7 @@ in pybaselines to distinguish them from polynomial techniques that also take
 advantage of weighted least squares (like :func:`.loess`) and penalized least
 squares (like :func:`.penalized_poly`).
 
-The general idea behnind WSB algorithms is to make the baseline match the measured
+The general idea behind WSB algorithms is to make the baseline match the measured
 data as well as it can while also penalizing the roughness of the baseline. The
 resulting general function that is minimized to determine the baseline is then
 

docs/conf.py

@@ -19,6 +19,8 @@
 #
 import os
 import sys
+
+
 sys.path.insert(0, os.path.abspath('..'))
 
 # -- General configuration ---------------------------------------------
@@ -64,7 +66,7 @@
 # the built documents.
 #
 # The short X.Y version.
-version = '0.3.0'
+version = '0.4.0'
 # The full version, including alpha/beta/rc tags.
 release = version
 

docs/index.rst

@@ -1,7 +1,7 @@
 pybaselines Documentation
 =========================
 
-pybaselines is a collection of algorithms for fitting the baseline of experimental data.
+pybaselines is a collection of algorithms for estimating the baseline of experimental data.
 
 * For Python 3.6+
 * Open Source: BSD 3-Clause License

docs/installation.rst

@@ -10,8 +10,8 @@ Dependencies
 
 pybaselines requires `Python <https://python.org>`_ version 3.6 or later and the following libraries:
 
-* `NumPy <https://numpy.org>`_ (>= 1.9)
-* `SciPy <https://www.scipy.org/scipylib/index.html>`_
+* `NumPy <https://numpy.org>`_ (>= 1.14)
+* `SciPy <https://www.scipy.org/scipylib/index.html>`_ (>= 0.11)
 
 
 All of the required libraries should be automatically installed when installing pybaselines
@@ -51,4 +51,4 @@ Once the repository is downloaded, it can be installed with:
 .. code-block:: console
 
     cd pybaselines
-    python setup.py install
+    pip install .

docs/introduction.rst

@@ -50,6 +50,6 @@ e) Optimizers (:mod:`pybaselines.optimizers`)
     2) optimize_extended_range
     3) adaptive_minmax (Adaptive MinMax)
 
-f) Manual methods (:mod:`pybaselines.manual`)
+f) Miscellaneous methods (:mod:`pybaselines.misc`)
 
-    1) linear_interp (Linear interpolation between points)
+    1) interp_pts (Interpolation between points)