diff --git a/doc/logos/scikit-learn-logo-thumb.png b/doc/logos/scikit-learn-logo-thumb.png
new file mode 100644
index 0000000000000000000000000000000000000000..e67908194c8f0b012856d2b7bcd08b6a7ba1d72e
Binary files /dev/null and b/doc/logos/scikit-learn-logo-thumb.png differ
diff --git a/doc/sphinxext/gen_rst.py b/doc/sphinxext/gen_rst.py
index c61e336327ebde90a459457fe1a402c3acbbc401..0cf3840302152bf10f5a073401c53e703b9ea150 100644
--- a/doc/sphinxext/gen_rst.py
+++ b/doc/sphinxext/gen_rst.py
@@ -110,6 +110,29 @@ def generate_example_rst(app):
fhindex = file(os.path.join(root_dir, 'index.rst'), 'w')
fhindex.write("""\
+.. raw:: html
+
+ <style type="text/css">
+ .figure {
+ float: left;
+ margin: 10px;
+ height: auto;
+ width: auto;
+ height: 200px;
+ width: 180px;
+ }
+
+ .figure img {
+ display: inline;
+ margin: 3px;
+ }
+
+ .figure .caption {
+ width: 170px;
+ text-align: center !important;
+ }
+ </style>
+
Examples
========
@@ -119,8 +142,6 @@ Examples
# better than nested.
generate_dir_rst('.', fhindex, example_dir, root_dir, plot_gallery)
for dir in sorted(os.listdir(example_dir)):
- if dir == '.svn':
- continue
if os.path.isdir(os.path.join(example_dir, dir)):
generate_dir_rst(dir, fhindex, example_dir, root_dir, plot_gallery)
fhindex.flush()
@@ -144,9 +165,9 @@ def generate_dir_rst(dir, fhindex, example_dir, root_dir, plot_gallery):
return
fhindex.write("""
+
%s
-.. toctree::
""" % file(os.path.join(src_dir, 'README.txt')).read())
if not os.path.exists(target_dir):
@@ -154,7 +175,17 @@ def generate_dir_rst(dir, fhindex, example_dir, root_dir, plot_gallery):
for fname in sorted(os.listdir(src_dir)):
if fname.endswith('py'):
generate_file_rst(fname, target_dir, src_dir, plot_gallery)
- fhindex.write(' %s\n' % (os.path.join(dir, fname[:-3])))
+ thumb = os.path.join(dir, 'images', 'thumb', fname[:-3] + '.png')
+ fhindex.write('.. figure:: %s\n\n' % thumb)
+ link_name = os.path.join(dir, fname).replace(os.path.sep, '_')
+ if link_name.startswith('._'):
+ link_name = link_name[2:]
+ fhindex.write(' :ref:`example_%s`\n\n' % link_name)
+ fhindex.write("""
+.. raw:: html
+
+ <div style="clear: both"></div>
+ """)
def generate_file_rst(fname, target_dir, src_dir, plot_gallery):
@@ -178,15 +209,18 @@ def generate_file_rst(fname, target_dir, src_dir, plot_gallery):
# The following is a list containing all the figure names
figure_list = []
+ image_dir = os.path.join(target_dir, 'images')
+ thumb_dir = os.path.join(image_dir, 'thumb')
+ if not os.path.exists(image_dir):
+ os.makedirs(image_dir)
+ if not os.path.exists(thumb_dir):
+ os.makedirs(thumb_dir)
+ image_path = os.path.join(image_dir, image_fname)
+ thumb_file = os.path.join(thumb_dir, fname[:-3] + '.png')
if plot_gallery and fname.startswith('plot'):
# generate the plot as png image if file name
# starts with plot and if it is more recent than an
# existing image.
- image_dir = os.path.join(target_dir, 'images')
- if not os.path.exists(image_dir):
- os.makedirs(image_dir)
-
- image_path = os.path.join(image_dir, image_fname)
first_image_file = image_path % 1
if (not os.path.exists(first_image_file) or
@@ -228,7 +262,16 @@ def generate_file_rst(fname, target_dir, src_dir, plot_gallery):
figure_list = [f[len(image_dir):]
for f in glob.glob(image_path % '[1-9]')]
#for f in glob.glob(image_path % '*')]
+
+ # generate thumb file
this_template = plot_rst_template
+ from matplotlib import image
+ if os.path.exists(first_image_file):
+ image.thumbnail(first_image_file, thumb_file, 0.2)
+
+ if not os.path.exists(thumb_file):
+ # create something not to replace the thumbnail
+ shutil.copy('logos/scikit-learn-logo-thumb.png', thumb_file)
docstring, short_desc, end_row = extract_docstring(example_file)
diff --git a/examples/covariance/plot_covariance_estimation.py b/examples/covariance/plot_covariance_estimation.py
index c01747f6ebd16e7c92801a140b8461b0b01bd79c..178caceecfe2c6cfd40b0ec2c2cb04d3f1efe31e 100644
--- a/examples/covariance/plot_covariance_estimation.py
+++ b/examples/covariance/plot_covariance_estimation.py
@@ -68,7 +68,6 @@ loglik_real = -log_likelihood(emp_cov, linalg.inv(real_cov))
###############################################################################
# Plot results
-pl.figure(-1)
pl.title("Regularized covariance: likelihood and shrinkage coefficient")
pl.xlabel('Shrinkage')
pl.ylabel('Negative log-likelihood')
diff --git a/examples/manifold/plot_lle_digits.py.prof b/examples/manifold/plot_lle_digits.py.prof
new file mode 100644
index 0000000000000000000000000000000000000000..2d30366716f99aa15baa8284df16ef3f2900d2b0
Binary files /dev/null and b/examples/manifold/plot_lle_digits.py.prof differ
diff --git a/examples/plot_precision_recall.py b/examples/plot_precision_recall.py
index f62b21d0392fb3b7d1e509f087207f2ec25d1ccb..8d4f1a9a0ea5ec5df597b27a3988993f1ae14907 100644
--- a/examples/plot_precision_recall.py
+++ b/examples/plot_precision_recall.py
@@ -40,7 +40,6 @@ precision, recall, thresholds = precision_recall_curve(y[half:], probas_[:,1])
area = auc(recall, precision)
print "Area Under Curve: %0.2f" % area
-pl.figure(-1)
pl.clf()
pl.plot(recall, precision, label='Precision-Recall curve')
pl.xlabel('Recall')
diff --git a/examples/plot_roc.py b/examples/plot_roc.py
index e9863d00055398bfe9f31fb705a230ff23c8ec33..b7c529c041a01c143f12d2a82bd863526a7bbc70 100644
--- a/examples/plot_roc.py
+++ b/examples/plot_roc.py
@@ -44,7 +44,6 @@ roc_auc = auc(fpr, tpr)
print "Area under the ROC curve : %f" % roc_auc
# Plot ROC curve
-pl.figure(-1)
pl.clf()
pl.plot(fpr, tpr, label='ROC curve (area = %0.2f)' % roc_auc)
pl.plot([0, 1], [0, 1], 'k--')
diff --git a/examples/plot_train_error_vs_test_error.py b/examples/plot_train_error_vs_test_error.py
index 26a367fa3d2c1c90cb3935ab2a78e7abbe8422a1..12da0c8edcd50894f3f4132677c5e8cc0077b81c 100644
--- a/examples/plot_train_error_vs_test_error.py
+++ b/examples/plot_train_error_vs_test_error.py
@@ -39,7 +39,8 @@ enet = linear_model.ElasticNet(rho=0.7)
train_errors = list()
test_errors = list()
for alpha in alphas:
- enet.fit(X_train, y_train, alpha=alpha)
+ enet.set_params(alpha=alpha)
+ enet.fit(X_train, y_train)
train_errors.append(enet.score(X_train, y_train))
test_errors.append(enet.score(X_test, y_test))
@@ -48,7 +49,8 @@ alpha_optim = alphas[i_alpha_optim]
print "Optimal regularization parameter : %s" % alpha_optim
# Estimate the coef_ on full data with optimal regularization parameter
-coef_ = enet.fit(X, y, alpha=alpha_optim).coef_
+enet.set_params(alpha=alpha_optim)
+coef_ = enet.fit(X, y).coef_
###############################################################################
# Plot results functions