similarity_val2 = kmeans.get_similarity_val(
labelled_dataset_features=palmar_features,
unlabelled_dataset_features=unlabelled_features)
result = {}
for image_id in list(unlabelled_features.keys()):
if similarity_val1[image_id] <= similarity_val2[image_id]:
result[image_id] = 'dorsal'
else:
result[image_id] = 'palmar'
print(result)
#ACCURACY
metadata = Metadata(metadatapath='Data/HandInfo.csv')
images_dop_dict = metadata.getimagesdop_dict()
print('Accuracy:', misc.getAccuracy(result, images_dop_dict))
elif task == '3':
folder_path = input("Enter folder path: ")
start_images = list(map(str, input("Enter 3 imageids: ").split()))
k = int(input("Enter number of outgoing edges: "))
m = int(input("Enter number of dominant images to show: "))
pagerank = PageRankUtil(folder_path, k, m, start_images)
pagerank.page_rank_util()
pagerank.plot_k_similar()
elif task == '4':
classifier = input("1.SVM\n2.DT\n3.PPR\nSelect Classifier: ")
labelled_dataset_path = input('Enter labelled dataset path: ')
unlabelled_dataset_path = input('Enter unlabelled dataset path: ')
