Merge pull request scikit-learn#4723 from amueller/backlinks

[MRG] website: adding backlinks to docstrings

doc/modules/clustering.rst

@@ -879,13 +879,11 @@ classes according to some similarity metric.
 
 .. currentmodule:: sklearn.metrics
 
+.. _adjusted_rand_score:
 
 Adjusted Rand index
 -------------------
 
-Presentation and usage
-~~~~~~~~~~~~~~~~~~~~~~
-
 Given the knowledge of the ground truth class assignments ``labels_true``
 and our clustering algorithm assignments of the same samples
 ``labels_pred``, the **adjusted Rand index** is a function that measures
@@ -1000,13 +998,11 @@ random labelings by defining the adjusted Rand index as follows:
  * `Wikipedia entry for the adjusted Rand index
    <http://en.wikipedia.org/wiki/Rand_index#Adjusted_Rand_index>`_
 
+.. _mutual_info_score:
 
 Mutual Information based scores
 -------------------------------
 
-Presentation and usage
-~~~~~~~~~~~~~~~~~~~~~~
-
 Given the knowledge of the ground truth class assignments ``labels_true`` and
 our clustering algorithm assignments of the same samples ``labels_pred``, the
 **Mutual Information** is a function that measures the **agreement** of the two
@@ -1168,12 +1164,11 @@ calculated using a similar form to that of the adjusted Rand index:
  * `Wikipedia entry for the Adjusted Mutual Information
    <http://en.wikipedia.org/wiki/Adjusted_Mutual_Information>`_
 
+.. _homogeneity_completeness:
+
 Homogeneity, completeness and V-measure
 ---------------------------------------
 
-Presentation and usage
-~~~~~~~~~~~~~~~~~~~~~~
-
 Given the knowledge of the ground truth class assignments of the samples,
 it is possible to define some intuitive metric using conditional entropy
 analysis.
@@ -1329,9 +1324,6 @@ mean of homogeneity and completeness**:
 Silhouette Coefficient
 ----------------------
 
-Presentation and usage
-~~~~~~~~~~~~~~~~~~~~~~
-
 If the ground truth labels are not known, evaluation must be performed using
 the model itself. The Silhouette Coefficient
 (:func:`sklearn.metrics.silhouette_score`)

doc/modules/covariance.rst

@@ -248,6 +248,7 @@ paper. It is the same algorithm as in the R ``glasso`` package.
      graphical lasso" <http://biostatistics.oxfordjournals.org/content/9/3/432.short>`_,
      Biostatistics 9, pp 432, 2008
 
+.. _robust_covariance:
 
 Robust Covariance Estimation
 ============================

doc/modules/decomposition.rst

@@ -554,9 +554,9 @@ structure of the error covariance :math:`\Psi`:
 * :math:`\Psi = \sigma^2 \mathbf{I}`: This assumption leads to
   the probabilistic model of :class:`PCA`.
 
-* :math:`\Psi = diag(\psi_1, \psi_2, \dots, \psi_n)`: This model is called Factor
-  Analysis, a classical statistical model. The matrix W is sometimes called
-  the "factor loading matrix".
+* :math:`\Psi = diag(\psi_1, \psi_2, \dots, \psi_n)`: This model is called
+  :class:`FactorAnalysis`, a classical statistical model. The matrix W is
+  sometimes called the "factor loading matrix".
 
 Both model essentially estimate a Gaussian with a low-rank covariance matrix.
 Because both models are probabilistic they can be integrated in more complex

doc/modules/ensemble.rst

@@ -780,6 +780,8 @@ accessed via the ``feature_importances_`` property::
 
 .. currentmodule:: sklearn.ensemble.partial_dependence
 
+.. _partial_dependence:
+
 Partial dependence
 ..................
 
@@ -989,10 +991,10 @@ calculated as follows:
 ================  ==========    ==========      ==========
 classifier        class 1       class 2         class 3
 ================  ==========    ==========      ==========
-classifier 1	  w1 * 0.2     w1 * 0.5       w1 * 0.3
-classifier 2	  w2 * 0.6     w2 * 0.3       w2 * 0.1
+classifier 1	  w1 * 0.2      w1 * 0.5        w1 * 0.3
+classifier 2	  w2 * 0.6      w2 * 0.3        w2 * 0.1
 classifier 3      w3 * 0.3      w3 * 0.4        w3 * 0.3
-weighted average  0.37	        0.4            0.3
+weighted average  0.37	        0.4             0.3
 ================  ==========    ==========      ==========
 
 Here, the predicted class label is 2, since it has the
@@ -1031,7 +1033,7 @@ Vector Machine, a Decision Tree, and a K-nearest neighbor classifier::
     :scale: 75%
 
 Using the `VotingClassifier` with `GridSearch`
----------------------------------------------
+----------------------------------------------
 
 The `VotingClassifier` can also be used together with `GridSearch` in order
 to tune the hyperparameters of the individual estimators::

doc/modules/feature_extraction.rst

@@ -826,6 +826,7 @@ Some tips and tricks:
 Customizing the vectorizer can also be useful when handling Asian languages
 that do not use an explicit word separator such as whitespace.
 
+.. _image_feature_extraction:
 
 Image feature extraction
 ========================

doc/modules/feature_selection.rst

@@ -13,6 +13,8 @@ improve estimators' accuracy scores or to boost their performance on very
 high-dimensional datasets.
 
 
+.. _variance_threshold:
+
 Removing features with low variance
 ===================================
 
@@ -45,6 +47,8 @@ so we can select using the threshold ``.8 * (1 - .8)``::
 As expected, ``VarianceThreshold`` has removed the first column,
 which has a probability :math:`p = 5/6 > .8` of containing a zero.
 
+.. _univariate_feature_selection:
+
 Univariate feature selection
 ============================
 
@@ -101,6 +105,7 @@ univariate p-values:
 
     :ref:`example_feature_selection_plot_feature_selection.py`
 
+.. _rfe:
 
 Recursive feature elimination
 =============================

doc/modules/grid_search.rst

@@ -72,6 +72,8 @@ evaluated and the best combination is retained.
       classifier (here a linear SVM trained with SGD with either elastic
       net or L2 penalty) using a :class:`pipeline.Pipeline` instance.
 
+.. _randomized_parameter_search:
+
 Randomized Parameter Optimization
 =================================
 While using a grid of parameter settings is currently the most widely used

doc/modules/kernel_approximation.rst

@@ -43,6 +43,7 @@ kernel function or a precomputed kernel matrix.
 The number of samples used - which is also the dimensionality of the features computed -
 is given by the parameter ``n_components``.
 
+.. _rbf_kernel_approx:
 
 Radial Basis Function Kernel
 ----------------------------
@@ -98,6 +99,7 @@ use of larger feature spaces more efficient.
 
     * :ref:`example_plot_kernel_approximation.py`
 
+.. _additive_chi_kernel_approx:
 
 Additive Chi Squared Kernel
 ---------------------------
@@ -130,6 +132,7 @@ with the approximate feature map provided by :class:`RBFSampler` to yield an app
 feature map for the exponentiated chi squared kernel.
 See the [VZ2010]_ for details and [VVZ2010]_ for combination with the :class:`RBFSampler`.
 
+.. _skewed_chi_kernel_approx:
 
 Skewed Chi Squared Kernel
 -------------------------

doc/modules/linear_model.rst

@@ -266,6 +266,7 @@ They also tend to break when the problem is badly conditioned
 
   * :ref:`example_linear_model_plot_lasso_model_selection.py`
 
+.. _elastic_net:
 
 Elastic Net
 ===========
@@ -486,6 +487,9 @@ previously chosen dictionary elements.
    <http://blanche.polytechnique.fr/~mallat/papiers/MallatPursuit93.pdf>`_,
    S. G. Mallat, Z. Zhang,
 
+
+.. _bayesian_regression:
+
 Bayesian Regression
 ===================
 
@@ -752,6 +756,8 @@ while with ``loss="hinge"`` it fits a linear support vector machine (SVM).
 
  * :ref:`sgd`
 
+.. _perceptron:
+
 Perceptron
 ==========
 

doc/modules/manifold.rst

@@ -149,6 +149,7 @@ The overall complexity of Isomap is
      <http://www.sciencemag.org/content/290/5500/2319.full>`_
      Tenenbaum, J.B.; De Silva, V.; & Langford, J.C.  Science 290 (5500)
 
+.. _locally_linear_embedding:
 
 Locally Linear Embedding
 ========================

doc/modules/metrics.rst

@@ -35,6 +35,8 @@ the kernel:
 
 .. currentmodule:: sklearn.metrics.pairwise
 
+.. _cosine_similarity:
+
 Cosine similarity
 -----------------
 :func:`cosine_similarity` computes the L2-normalized dot product of vectors.
@@ -63,6 +65,8 @@ is equivalent to :func:`linear_kernel`, only slower.)
       Information Retrieval. Cambridge University Press.
       http://nlp.stanford.edu/IR-book/html/htmledition/the-vector-space-model-for-scoring-1.html
 
+.. _linear_kernel:
+
 Linear kernel
 -------------
 The function :func:`linear_kernel` computes the linear kernel, that is, a
@@ -73,6 +77,8 @@ If ``x`` and ``y`` are column vectors, their linear kernel is:
 
     k(x, y) = x^\top y
 
+.. _polynomial_kernel:
+
 Polynomial kernel
 -----------------
 The function :func:`polynomial_kernel` computes the degree-d polynomial kernel
@@ -94,6 +100,8 @@ where:
 
 If :math:`c_0 = 0` the kernel is said to be homogeneous.
 
+.. _sigmoid_kernel:
+
 Sigmoid kernel
 --------------
 The function :func:`sigmoid_kernel` computes the sigmoid kernel between two
@@ -111,6 +119,8 @@ where:
     * :math:`\gamma` is known as slope
     * :math:`c_0` is known as intercept
 
+.. _rbf_kernel:
+
 RBF kernel
 ----------
 The function :func:`rbf_kernel` computes the radial basis function (RBF) kernel
@@ -123,6 +133,8 @@ between two vectors. This kernel is defined as:
 where ``x`` and ``y`` are the input vectors. If :math:`\gamma = \sigma^{-2}`
 the kernel is known as the Gaussian kernel of variance :math:`\sigma^2`.
 
+.. _chi2_kernel:
+
 Chi-squared kernel
 ------------------
 The chi-squared kernel is a very popular choice for training non-linear SVMs in

doc/modules/mixture.rst

@@ -133,6 +133,7 @@ parameters to maximize the likelihood of the data given those
 assignments. Repeating this process is guaranteed to always converge
 to a local optimum. 
 
+.. _vbgmm:
 
 VBGMM classifier: variational Gaussian mixtures
 ================================================