diff --git a/examples/svm/plot_svm_regression.py b/examples/svm/plot_svm_regression.py
index 54d2c0b54337b..73654e1ded93c 100644
--- a/examples/svm/plot_svm_regression.py
+++ b/examples/svm/plot_svm_regression.py
@@ -23,9 +23,9 @@
 
 # #############################################################################
 # Fit regression model
-svr_rbf = SVR(kernel='rbf', C=1e3, gamma=0.1)
-svr_lin = SVR(kernel='linear', C=1e3)
-svr_poly = SVR(kernel='poly', C=1e3, degree=2)
+svr_rbf = SVR(kernel='rbf', C=100, gamma=0.1, epsilon=.1)
+svr_lin = SVR(kernel='linear', C=100)
+svr_poly = SVR(kernel='poly', C=100, degree=3, epsilon=.1, coef0=1)
 y_rbf = svr_rbf.fit(X, y).predict(X)
 y_lin = svr_lin.fit(X, y).predict(X)
 y_poly = svr_poly.fit(X, y).predict(X)
@@ -33,12 +33,41 @@
 # #############################################################################
 # Look at the results
 lw = 2
-plt.scatter(X, y, color='darkorange', label='data')
-plt.plot(X, y_rbf, color='navy', lw=lw, label='RBF model')
-plt.plot(X, y_lin, color='c', lw=lw, label='Linear model')
-plt.plot(X, y_poly, color='cornflowerblue', lw=lw, label='Polynomial model')
-plt.xlabel('data')
-plt.ylabel('target')
-plt.title('Support Vector Regression')
-plt.legend()
+fig, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(15, 10), sharey=True)
+
+ax1.plot(X, y_rbf, color='m', lw=lw, label='RBF model')
+ax1.scatter(X[svr_rbf.support_], y[svr_rbf.support_], facecolor="none",
+            edgecolor="m", label='rbf support vectors', s=50)
+ax1.scatter(X[np.setdiff1d(np.arange(len(X)), svr_rbf.support_)],
+            y[np.setdiff1d(np.arange(len(X)), svr_rbf.support_)],
+            facecolor="none",
+            edgecolor="k", label='other training data', s=50)
+ax1.legend(loc='upper center', bbox_to_anchor=(0.5, 1.1),
+           ncol=1, fancybox=True, shadow=True)
+
+
+ax2.plot(X, y_lin, color='c', lw=lw, label='Linear model')
+ax2.scatter(X[svr_lin.support_], y[svr_lin.support_], facecolor="none",
+            edgecolor="c", label='linear support vectors', s=50)
+ax2.scatter(X[np.setdiff1d(np.arange(len(X)), svr_lin.support_)],
+            y[np.setdiff1d(np.arange(len(X)), svr_lin.support_)],
+            facecolor="none",
+            edgecolor="k", label='other training data', s=50)
+ax2.legend(loc='upper center', bbox_to_anchor=(0.5, 1.1),
+           ncol=1, fancybox=True, shadow=True)
+
+
+ax3.plot(X, y_poly, color='g', lw=lw, label='Polynomial model')
+ax3.scatter(X[svr_poly.support_], y[svr_poly.support_], facecolor="none",
+            edgecolor="g", label='poly support vectors', s=50)
+ax3.scatter(X[np.setdiff1d(np.arange(len(X)), svr_poly.support_)],
+            y[np.setdiff1d(np.arange(len(X)), svr_poly.support_)],
+            facecolor="none",
+            edgecolor="k", label='other training data', s=50)
+ax3.legend(loc='upper center', bbox_to_anchor=(0.5, 1.1),
+           ncol=1, fancybox=True, shadow=True)
+
+fig.text(0.5, 0.04, 'data', ha='center', va='center')
+fig.text(0.06, 0.5, 'target', ha='center', va='center', rotation='vertical')
+fig.suptitle("Support Vector Regression", fontsize=14)
 plt.show()