Lots of new improvements

1 files changed, 36 insertions, 14 deletions

diff --git a/train.py b/train.py
index 26567e8..b3624e8 100755
--- a/train.py
+++ b/train.py
@@ -4,7 +4,9 @@
 # EE4 Pattern Recognition coursework
 
 import matplotlib.pyplot as plt
-import sys;
+import sys
+
+from random import randint
 
 from sklearn.neighbors import KNeighborsClassifier
 from sklearn.decomposition import PCA
@@ -29,8 +31,12 @@ def normalise_faces(average_face, raw_faces):
 parser = argparse.ArgumentParser()
 parser.add_argument("-i", "--data", help="Input CSV file", required=True)
 parser.add_argument("-m", "--eigen", help="Number of eigenvalues in model", type=int, default = 8 )
-parser.add_argument("-n", "--neighbors", help="How many neighbors to use", type=int, default = 3)
-parser.add_argument("-g", "--graph", help="Should we show graphs", action='store_true')
+parser.add_argument("-n", "--neighbors", help="How many neighbors to use", type=int, default = 6)
+parser.add_argument("-f", "--faces", help="Show faces", type=int, default = 0)
+parser.add_argument("-c", "--principal", help="Show principal components", action='store_true')
+parser.add_argument("-s", "--seed", help="Seed to use", type=int, default=0)
+parser.add_argument("-t", "--split", help="Fractoin of data to use for testing", type=float, default=0.25)
+parser.add_argument("-2", "--grapheigen", help="Swow 2D graph of targets versus principal components", action='store_true')
 parser.add_argument("-p", "--pca", help="Use PCA", action='store_true')
 parser.add_argument("-l", "--lda", help="Use LDA", action='store_true')
 args = parser.parse_args()
@@ -43,7 +49,9 @@ M = args.eigen
 raw_faces = genfromtxt(args.data, delimiter=',')
 targets = np.repeat(np.arange(10),52)
 
-faces_train, faces_test, target_train, target_test = train_test_split(raw_faces, targets, test_size=0.2, random_state=0)
+split = 100
+
+faces_train, faces_test, target_train, target_test = train_test_split(raw_faces, targets, test_size=args.split, random_state=args.seed)
 
 
 # This remove the mean and scales to unit variance
@@ -52,25 +60,39 @@ faces_train = sc.fit_transform(faces_train)
 faces_test = sc.transform(faces_test)
 
 explained_variances = ()
-if args.pca:
-    # faces_pca containcts the principial components or the M most variant eigenvectors
-    pca = PCA(n_components=M)
-    faces_train = pca.fit_transform(faces_train)
-    faces_test = pca.transform(faces_test)
-    explained_variances = pca.explained_variance_ratio_
-else:
+if args.lda:
     lda = LinearDiscriminantAnalysis(n_components=M)
     faces_train = lda.fit_transform(faces_train, target_train)
     faces_test = lda.transform(faces_test)
     explained_variances = lda.explained_variance_ratio_
-
+else:
+    # faces_pca containcts the principial components or the M most variant eigenvectors
+    pca = PCA(svd_solver='full', n_components=M)
+    faces_train = pca.fit_transform(faces_train)
+    faces_test = pca.transform(faces_test)
+    explained_variances = pca.explained_variance_ratio_
 # Plot the variances (eigenvalues) from the pca object
-if args.graph:
+if args.faces:
+    if args.lda:
+        sys.exit("Can not plot eigenfaces when using LDA")
+    for i in range(args.faces):
+        ax = plt.subplot(2, args.faces/2, i + 1)
+        ax.imshow(pca.components_[i].reshape([46, 56]))
+    plt.show()
+
+if args.principal:
     plt.bar(np.arange(explained_variances.size), explained_variances)
     plt.ylabel('Varaiance ratio');plt.xlabel('Face Number')
     plt.show()
 
-classifier = KNeighborsClassifier(n_neighbors=3)
+if args.grapheigen:
+    # Colors for distinct individuals
+    cols = ['#{:06x}'.format(randint(0, 0xffffff)) for i in range(10)]
+    pltCol = [cols[int(k)] for k in target_train]
+    plt.scatter(faces_train[:, 0], faces_train[:, 1], color=pltCol)
+    plt.show()
+
+classifier = KNeighborsClassifier(n_neighbors=args.neighbors)
 classifier.fit(faces_train, target_train)
 target_pred = classifier.predict(faces_test)