1 files changed, 14 insertions, 4 deletions

diff --git a/examples/minimize.hs b/examples/minimize.hs
index 5beba3b..8962006 100644
--- a/examples/minimize.hs
+++ b/examples/minimize.hs
@@ -13,6 +13,9 @@ df [x,y] = [20*(x-1), 40*(y-2)]
 -- the conjugate gradient method
 minimizeCG = minimizeConjugateGradient 1E-2 1E-4 1E-3 30
 
+-- the BFGS2 method
+minimizeBFGS2 = minimizeVectorBFGS2 1E-2 1E-2 1E-3 30
+
 -- a minimization algorithm which does not require the gradient
 minimizeS f xi = minimizeNMSimplex f xi (replicate (length xi) 1) 1E-2 100
 
@@ -24,20 +27,27 @@ partialDerivative n f v = fst (derivCentral 0.01 g (v!!n)) where
     (a,_:b) = splitAt n v
 
 main = do
-    -- conjugate gradient with true gradient
-    let (s,p) = minimizeCG f df [5,7]
+    putStrLn "BFGS2 with true gradient"
+    let (s,p) = minimizeBFGS2 f df [5,7]
     print s -- solution
     disp p -- evolution of the algorithm
     let [x,y] = drop 2 (toColumns p)
     mplot [x,y]  -- path from the starting point to the solution
 
-    -- conjugate gradient with estimated gradient
+    putStrLn "conjugate gradient with true gradient"
+    let (s,p) = minimizeCG f df [5,7]
+    print s
+    disp p
+    let [x,y] = drop 2 (toColumns p)
+    mplot [x,y]
+
+    putStrLn "conjugate gradient with estimated gradient"
     let (s,p) = minimizeCG f (gradient f) [5,7]
     print s
     disp p
     mplot $ drop 2 (toColumns p)
 
-    -- without gradient, using the NM Simplex method
+    putStrLn "without gradient, using the NM Simplex method"
     let (s,p) = minimizeS f [5,7]
     print s
     disp p