# components = dl.components().reshape((n_components, n_features))
components = dl.dictionary
# Visualizing the components as images
plot_gallery("Visualizing top components", components.T, w, h, n_row=n_components / 10, n_col=10)
plt.show()
###############################################################################
# Sparse Encoding
print("\nSparse Encoding")
train_X_sc = dl.code.T
np.set_printoptions(precision=1, suppress=False, linewidth=800)
test_X_sc = np.zeros((len(test_X), n_components))
test_X_sc = dl.sparse_encode(test_X, components).T
print "train_X_sc.shape", train_X_sc.shape
###############################################################################
# Visualize reconstructed images
reconstructed_X = np.zeros((20, n_features))
reconstructed_X_idx = range(10)
reconstructed_X[reconstructed_X_idx] = test_X[reconstructed_X_idx]
reconstructed_X[reconstructed_X_idx] = np.dot(components, test_X_sc[:, reconstructed_X_idx])
print "reconstructed_X.shape", reconstructed_X.shape
print(reconstructed_X)
plot_gallery("Reconstructed images", reconstructed_X, w, h, n_row=2, n_col=10)
plt.show()
print("Extracting the top %d eigenfaces from %d faces"
% (n_components, X_train.shape[0]))
t0 = time()
# pca = RandomizedPCA(n_components=n_components, whiten=True).fit(X_train)
dl = KSVDSparseCoding(n_components, preserve_dc=True, approx=False, max_iter=15, verbose=1)
dl.fit(X_train)
print("done in %0.3fs" % (time() - t0))
eigenfaces = dl.dictionary
print("Projecting the input data on the eigenfaces orthonormal basis")
t0 = time()
X_train_pca = dl.code.T
X_test_pca = dl.sparse_encode(X_test, eigenfaces).T
print("done in %0.3fs" % (time() - t0))
###############################################################################
# Train a SVM classification model
print("Fitting the classifier to the training set")
t0 = time()
param_grid = {'C': [1e3, 5e3, 1e4, 5e4, 1e5],
'gamma': [0.0001, 0.0005, 0.001, 0.005, 0.01, 0.1],}
clf = GridSearchCV(SVC(kernel='rbf', class_weight='balanced'), param_grid)
clf = clf.fit(X_train_pca, y_train)
print("done in %0.3fs" % (time() - t0))
print("Best estimator found by grid search:")
print(clf.best_estimator_)
