Fix readthedocs failing test (#518)

* Empty commit

* Fix spelling mistake and title underline too short

* Fix a spelling mistake

* Empty commit

* Force matplotlib version to be equal to 3.8.0 in requirements_rtd.txt

* Force matplotlib version to be equal to 3.7.0 in requirements_rtd.txt

* Remove specification of matplotlib version in requirements_rtd.txt

* Force pillow version to be equal to 10.2.0 in requirements_rtd.txt

* Remove specification on pillow version in requirements_rtd.txt

* Force sphinx version to be equal to 7.2.6 in requirements_rtd.txt

* Remove specification on sphinx version in requirements_rtd.txt

* Force numpydoc version to be equal to 1.6.0 in requirements_rtd.txt

* Remove specification of the numpydoc version in requirements_rtd.txt

* Force sphinx-gallery version to be equal to 0.15.0 in requirements_rtd.txt

* Remove specification of the sphinx-gallery version in requirements_rtd.txt

* Modifiy sphinx_gallery_conf in the file conf.py in the folder docs to be compatible with sphinx-gallery 0.16.0

* Move matplotlib_svg_scraper in a new file named utils.py in the folder docs

* Move matplotlib_svg_craper back into conf.py and use it in sphinx_gallery_conf as a callable

* Try the value matplotlib for the key image_scrapers in sphinx_gallery_conf

* Remove matplotlib_svg_scraper from conf.py

* Remove file docs/utils.py

* Remove empty line from conf.py

* Use matplotlib_svg_scraper in conf.py and specify sphinx-gallery==0.15.0 in requirements_rtd.txt

Author: YannCabanes

docs/examples/metrics/plot_dtw_custom_metric.py

@@ -5,7 +5,7 @@
 .. currentmodule:: tslearn.metrics
 
 This example illustrates how to use the DTW computation of the optimal
-alignment path [1]_ on an user-defined distance matrix using
+alignment path [1]_ on a user-defined distance matrix using
 :func:`dtw_path_from_metric`.
 
 Left is the DTW of two angular time series using the length of the arc on the

docs/examples/metrics/plot_lcss.py

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 """
 Longest Common Subsequence
-===============
+==========================
 
 This example illustrates LCSS computation between time series and plots the
 alignment path [1]. and its relationship to the DTW.

docs/examples/metrics/plot_lcss_custom_metric.py

@@ -1,11 +1,11 @@
 # -*- coding: utf-8 -*-
 """
-Longest Commom Subsequence with a custom distance metric
-=============================================
+Longest Common Subsequence with a custom distance metric
+========================================================
 .. currentmodule:: tslearn.metrics
 
 This example illustrates how to use the LCSS computation of the
-alignment path [1]_ on an user-defined distance matrix using
+alignment path [1]_ on a user-defined distance matrix using
 :func:`dtw_path_from_metric`.
 
 The example is the LCSS of two angular time series using the length of the arc on the
@@ -117,4 +117,4 @@ def arc_length(angle_1, angle_2, r=1.):
                 horizontalalignment="center", verticalalignment="top")
 
 plt.tight_layout()
-plt.show()
+plt.show()

docs/requirements_rtd.txt

@@ -6,7 +6,7 @@ numba
 sphinx>=1.6.1
 ipykernel
 nbsphinx
-sphinx-gallery
+sphinx-gallery==0.15.0
 pillow
 tensorflow>=2
 Pygments