def ex_dimensionality_reduction(data_matrix,
image_ids,
model,
lsa_model,
k_features,
label=None,
viz=False):
if lsa_model == "SVD":
reduced_data = reduce_dimensions_svd(data_matrix, k_features, viz)
elif lsa_model == "PCA":
reduced_data = reduce_dimensions_pca(data_matrix, k_features, viz)
elif lsa_model == "NMF":
if model == "CM":
print("NMF is not applicable to Color Moments")
return
reduced_data = reduce_dimensions_nmf(data_matrix, k_features, viz)
elif lsa_model == "LDA":
if model == "CM":
print("LDA is modified and applicable to Color Moments")
data_matrix_cmlda = convert_data_matrix_cmlda(data_matrix)
reduced_data = reduce_dimensions_lda(data_matrix_cmlda, k_features,
viz)
else:
reduced_data = reduce_dimensions_lda(data_matrix, k_features, viz)
insert_reduced_features(model, reduced_data, image_ids, label)
if extra_credit:
return reduced_data, image_ids
def dimensionality_reduction(data_matrix,
image_ids,
args,
label=None,
viz=False):
if args.lsa_model == "SVD":
reduced_data = reduce_dimensions_svd(data_matrix, args.k_features,
image_ids, viz)
elif args.lsa_model == "PCA":
reduced_data = reduce_dimensions_pca(data_matrix, args.k_features,
image_ids, viz)
elif args.lsa_model == "NMF":
if args.model == "CM":
print("NMF is not applicable to Color Moments")
return
reduced_data = reduce_dimensions_nmf(data_matrix, args.k_features,
image_ids, viz)
elif args.lsa_model == "LDA":
if args.model == "CM":
print("LDA is applicable only to a modified Color Moments.")
data_matrix_cmlda = convert_data_matrix_cmlda(data_matrix)
reduced_data = reduce_dimensions_lda(data_matrix_cmlda,
args.k_features, image_ids,
viz)
else:
reduced_data = reduce_dimensions_lda(data_matrix, args.k_features,
image_ids, viz)
insert_reduced_features(args.model, reduced_data, image_ids, label)
return reduced_data, image_ids
def get_V_matrix(data_matrix, image_ids, args, label=None, viz=False):
if args.lsa_model == "SVD":
reduced_data, V_matrix = reduce_dimensions_svd(data_matrix,
args.k_features,
get_v=True)
elif args.lsa_model == "PCA":
reduced_data, V_matrix = reduce_dimensions_pca(data_matrix,
args.k_features,
get_v=True)
elif args.lsa_model == "NMF":
if args.model == "CM":
print("NMF is not applicable to Color Moments")
return
reduced_data, V_matrix = reduce_dimensions_nmf(data_matrix,
args.k_features,
get_v=True)
elif args.lsa_model == "LDA":
if args.model == "CM":
print("LDA is NOT applicable because of the Bag Of Word model.")
# data_matrix_cmlda = convert_data_matrix_cmlda(data_matrix)
# V_matrix = reduce_dimensions_lda(data_matrix_cmlda, args.k_features, image_ids, viz, get_v=True)
return
else:
reduced_data, V_matrix = reduce_dimensions_lda(data_matrix,
args.k_features,
get_v=True)
# insert_reduced_features(args.model, reduced_data, image_ids, label)
return reduced_data, V_matrix
def main():
parser = setup_arg_parse()
args = parser.parse_args()
populate_database(args)
model = "CM"
dorsal_data_matrix, _ = get_data_matrix(
model, convert_label_to_filterstring("dorsal"))
palmar_data_matrix, _ = get_data_matrix(
model, convert_label_to_filterstring("palmar"))
dorsal_labels = np.zeros((dorsal_data_matrix.shape[0], 1))
palmar_labels = np.ones((palmar_data_matrix.shape[0], 1))
labels = np.append(dorsal_labels, palmar_labels, axis=0)
combined_data = np.append(dorsal_data_matrix, palmar_data_matrix, axis=0)
#reduced_data = reduce_dimensions_svd(combined_data, 20)
reduced_data, v_matrix = reduce_dimensions_svd(combined_data,
20,
get_v=True)
dx, ddx, labels, d_labels = train_test_split(reduced_data,
labels,
test_size=0.1,
random_state=42)
reduced_data = np.append(dx, ddx, axis=0)
labels = np.append(labels, d_labels, axis=0)
labeled_data = np.append(reduced_data, labels, axis=1)
testing_images, test_image_ids = enumerate_files_in_dir(args.test_folder)
test_dataset = []
for test_image, image_id in zip(testing_images, test_image_ids):
#test_dataset.append(np.array(extract_hog_features(test_image)))
test_dataset.append(np.array(extract_color_moments(test_image)))
test_dataset = np.array(test_dataset)
#reduced_test_dataset = reduce_dimensions_svd(test_dataset, 20)
reduced_test_dataset = np.matmul(test_dataset, v_matrix)
mongo_client = connect_to_db()
actual_labels = get_actual_labels_from_csv(args.labels_csv, test_image_ids)
predicted = []
if args.classifier == "DT":
model = DecisionTreeClassifier()
model.fit(labeled_data)
results = model.transform(reduced_test_dataset)
for test_image_id, result in zip(test_image_ids, results):
if result == 0:
label = "dorsal"
elif result == 1:
label = "palmar"
predicted.append((test_image_id, label))
print("{0} - {1}".format(test_image_id, label))
elif args.classifier == "SVM":
clf = SupportVectorMachine(kernel=rbf_kernel, power=4, coef=1)
training_labels = labels[:]
# SVM needs labels to be 1, and -1
training_labels[training_labels == 0] = -1
clf.fit(reduced_data, training_labels)
values = clf.predict(reduced_test_dataset)
print(values)
for test_image_id, result in zip(test_image_ids, values):
if result == 1:
label = "palmar"
else:
label = "dorsal"
predicted.append((test_image_id, label))
print("{0} - {1}".format(test_image_id, label))
elif args.classifier == "PPR":
args.k = 15
function_val = "manhattan"
#process_all_images(args.train_folder, "CM")
#process_all_images(args.test_folder, "CM")
outgoing_img_graph, image_ids = create_similarity_graph(
args.k, function_val, "CM")
transition_matrix = get_transition_matrix(outgoing_img_graph, args.k)
seed_matrix_dorsal = get_seed_matrix("dorsal")
seed_matrix_palmar = get_seed_matrix("palmar")
dorsal_pagerank = compute_pagerank(transition_matrix,
seed_matrix_dorsal)
palmar_pagerank = compute_pagerank(transition_matrix,
seed_matrix_palmar)
dorsal_pagerank_dict = {
x: y
for x, y in zip(image_ids, dorsal_pagerank)
}
palmar_pagerank_dict = {
x: y
for x, y in zip(image_ids, palmar_pagerank)
}
predicted = label_images(dorsal_pagerank_dict, palmar_pagerank_dict,
test_image_ids)
print(get_accuracy(actual_labels, predicted))
mongo_client.close()