doc: fix intersphinx links

doc/source/conf.py

@@ -136,6 +136,7 @@ intersphinx_mapping = {
         "http://py4cfd-sce-floss-0f4cd01cbe2305292bb4264806746195f04ffc2d3a1863.pages.gitlab.ost.ch/doc/",
         None,
     ),
+    "sklearn": ("http://scikit-learn.org/stable", None),
 }
 # -- Options for Gallery extension ----------------------------------------------
 examples_dir = "../examples"

doc/source/introduction.rst

@@ -75,15 +75,15 @@ For any method that uses the geometry rather than the point data of the bubbles,
    The constant part is where :math:`F_\text{bouancy} = F_\text{resistance}` the state of equilibrium means that acceleration is also 0 which results in a constant velocity.
    As a result the module returns the mean velocity for the middle 60% of the domain (assumed to be the constant part), and its standard deviation, the module also returns the velocity on the created uniform scale.
 
-These five post precessing modules are gathered together in the module :mod:`bubbles.pp_main.post_all` with which all postprocessing routines can be applied to a study.
+These five post precessing modules are gathered together in the module :func:`bubbles.pp_main.post_all` with which all postprocessing routines can be applied to a study.
 It returns a pandas.dataFrame with the postprocessing data and stores a pickle file for each simulation with the raw postprocessing data, which can be used for further calculations or plotting.
 
 ML model
 ----------
 
-To implement machine learning models the :mod:`sklearn.gaussian_process.GausianPrecessRegressor` (GPR) form scikit learn, as well as custom models where used.
+To implement machine learning models the :obj:`sklearn.gaussian_process.GaussianProcessRegressor` (GPR) form scikit learn, as well as custom models where used.
 Detailed information on how this model works can be found either in the scikit learn documentation or for the custom models a simplified explanation is available here. :ref:`ML Theory`.
-In cases where GPR from scikit learn was used, a combination of kernels was chosen to represent the behavior of the data. In special cases the :mod:`sklearn.linear_model.LinearRegression` was used.
+In cases where GPR from scikit learn was used, a combination of kernels was chosen to represent the behavior of the data. In special cases the :obj:`sklearn.linear_model.LinearRegression` was used.
 For custom made models the bases function was chosen to represent the known states of extrapolation, for example for a data set that goes towards infinity at zero a function that goes to infinity at zero was chosen.
 This method ensures decent extrapolation qualities for the model, also the extrapolation qualities are discussed with each model.