the l1 logreg example now works with l1_min_c

examples/linear_model/plot_logistic_path.py

@@ -18,6 +18,7 @@
 
 from scikits.learn import linear_model
 from scikits.learn import datasets
+from scikits.learn.svm import l1_min_c
 
 iris = datasets.load_iris()
 X = iris.data
@@ -31,18 +32,19 @@
 ################################################################################
 # Demo path functions
 
-alphas = np.logspace(2, -4, 100)
+cs = l1_min_c(X, y, loss='log') * np.logspace(0, 3)
+
 
 print "Computing regularization path ..."
 start = datetime.now()
 clf = linear_model.LogisticRegression(C=1.0, penalty='l1', tol=1e-6)
-coefs_ = [clf.fit(X, y, C=1.0/alpha).coef_.ravel().copy() for alpha in alphas]
+coefs_ = [clf.fit(X, y, C=c).coef_.ravel().copy() for c in cs]
 print "This took ", datetime.now() - start
 
 coefs_ = np.array(coefs_)
-pl.plot(-np.log10(alphas), coefs_)
+pl.plot(np.log10(cs), coefs_)
 ymin, ymax = pl.ylim()
-pl.xlabel('-log(alpha)')
+pl.xlabel('log(C)')
 pl.ylabel('Coefficients')
 pl.title('Logistic Regression Path')
 pl.axis('tight')