Fix Reduced PCA and face Reconstruction

1 files changed, 13 insertions, 7 deletions

diff --git a/train.py b/train.py
index 0927943..fd03220 100755
--- a/train.py
+++ b/train.py
@@ -69,7 +69,6 @@ parser.add_argument("-q", "--pca_r", help="Use Reduced PCA", action='store_true'
 
 args = parser.parse_args()
 
-
 M = args.eigen
 
 raw_faces = genfromtxt(args.data, delimiter=',')
@@ -83,6 +82,7 @@ faces_train, faces_test, target_train, target_test = test_split(n_faces, raw_fac
 sc = StandardScaler()
 #faces_train = sc.fit_transform(faces_train)
 #faces_test = sc.transform(faces_test)
+raw_faces_train = faces_train
 
 explained_variances = ()
 
@@ -93,9 +93,12 @@ if args.pca or args.pca_r:
     faces_train = normalise_faces(average_face, faces_train)
     faces_test = normalise_faces(average_face, faces_test)
     if (args.pca_r):
-        e_vals, e_vecs = LA.eigh(np.cov(faces_train))
+        print('Reduced PCA')
+        e_vals, e_vecs = LA.eigh(np.dot(faces_train, faces_train.T))
         e_vecs = np.dot(faces_train.T, e_vecs) 
+        e_vecs = e_vecs/LA.norm(e_vecs, axis = 0)
     else:
+        print('Standard PCA')
         e_vals, e_vecs = LA.eigh(np.cov(faces_train.T))
         # e_vecs = normalise_faces(np.mean(e_vecs,axis=0), e_vecs)
         # e_vecs = sc.fit_transform(e_vecs)
@@ -108,12 +111,15 @@ if args.pca or args.pca_r:
     faces_test = np.dot(faces_test, e_vecs.T)
 
     if (args.reconstruct):
-        for face in range(args.reconstruct):
-            rec_vec = np.add(average_face, np.dot(faces_train[face], e_vecs) * deviations)
-            ar = plt.subplot(2, args.reconstruct/2, face + 1)
-            ar.imshow(rec_vec.reshape([46,56]).T, cmap = 'gist_gray')
+        rec_vec = np.add(average_face, np.dot(faces_train[args.reconstruct], e_vecs) * deviations)
+        rec_error = LA.norm(np.subtract(raw_faces_train[args.reconstruct], rec_vec))
+        print(rec_error)
+        ar = plt.subplot(2, 1, 1)
+        ar.imshow(rec_vec.reshape([46,56]).T, cmap = 'gist_gray')
+        ar = plt.subplot(2, 1, 2)
+        ar.imshow(raw_faces_train[args.reconstruct].reshape([46,56]).T, cmap = 'gist_gray')
         plt.show()
-    
+
 if args.lda or (args.pca and args.lda):
     lda = LinearDiscriminantAnalysis(n_components=M)
     faces_train = lda.fit_transform(faces_train, target_train)