diff --git a/doc/conf.py b/doc/conf.py
index 78529fe9..ed03944c 100644
--- a/doc/conf.py
+++ b/doc/conf.py
@@ -90,9 +90,11 @@ extensions = ['sphinx.ext.autodoc',
               'ipython_directive',
               'ipython_console_highlighting',
               #'mathmpl',
-              'only_directives',
-              'plot_directive',
+              #'only_directives',
+              'matplotlib.sphinxext.plot_directive'
+              #'plot_directive'
              ]
+plot_formats = [('png', 80), ('pdf', 50)]
 
 print "finished importing"
 
@@ -122,7 +124,7 @@ class Mock(object):
 #import mock
 
 print "Mocking"
-MOCK_MODULES = ['pylab', 'matplotlib', 'sympy', 'sympy.utilities', 'sympy.utilities.codegen', 'sympy.core.cache', 'sympy.core', 'sympy.parsing', 'sympy.parsing.sympy_parser']
+MOCK_MODULES = ['pylab', 'sympy', 'sympy.utilities', 'sympy.utilities.codegen', 'sympy.core.cache', 'sympy.core', 'sympy.parsing', 'sympy.parsing.sympy_parser']
 #'matplotlib', 'matplotlib.color', 'matplotlib.pyplot', 'pylab' ]
 for mod_name in MOCK_MODULES:
     sys.modules[mod_name] = Mock()
diff --git a/doc/tuto_GP_regression.rst b/doc/tuto_GP_regression.rst
index 1a6e245a..93f5a02f 100644
--- a/doc/tuto_GP_regression.rst
+++ b/doc/tuto_GP_regression.rst
@@ -9,14 +9,13 @@ Gaussian process regression tutorial
 
 .. plot::
 
-   import matplotlib.pyplot as plt
-   import numpy as np
-   x = np.random.randn(1000)
-   plt.hist( x, 20)
-   plt.grid()
-   plt.title(r'Normal: $\mu=%.2f, \sigma=%.2f$'%(x.mean(), x.std()))
-   plt.show()
-
+    import matplotlib.pyplot as plt
+    import numpy as np
+    x = np.random.randn(1000)
+    plt.hist( x, 20)
+    plt.grid()
+    plt.title(r'Normal: $\mu=%.2f, \sigma=%.2f$'%(x.mean(), x.std()))
+    plt.show()
 
 
 We will see in this tutorial the basics for building a 1 dimensional and a 2 dimensional Gaussian process regression model, also known as a kriging model.