DOC fix more rst links

Author: lesteve

doc/modules/mixture.rst

@@ -303,7 +303,7 @@ with a finite Gaussian mixture model.
 An important question is how can the Dirichlet process use an infinite,
 unbounded number of clusters and still be consistent. While a full explanation
 doesn't fit this manual, one can think of its `stick breaking process
-<https://en.wikipedia.org/wiki/Dirichlet_process#The_stick-breaking_process>`
+<https://en.wikipedia.org/wiki/Dirichlet_process#The_stick-breaking_process>`_
 analogy to help understanding it. The stick breaking process is a generative
 story for the Dirichlet process. We start with a unit-length stick and in each
 step we break off a portion of the remaining stick. Each time, we associate the

sklearn/metrics/classification.py

@@ -975,7 +975,7 @@ def precision_recall_fscore_support(y_true, y_pred, beta=1.0, labels=None,
     .. [3] `Discriminative Methods for Multi-labeled Classification Advances
            in Knowledge Discovery and Data Mining (2004), pp. 22-30 by Shantanu
            Godbole, Sunita Sarawagi
-           <http://www.godbole.net/shantanu/pubs/multilabelsvm-pakdd04.pdf>`
+           <http://www.godbole.net/shantanu/pubs/multilabelsvm-pakdd04.pdf>`_
 
     Examples
     --------