classes_label = 'ABCDE'
# numbers = '1245'
letter_to_digit = Task_E.letter_2_digit_convert(classes_label)
# for i in numbers:
# letter_to_digit.append(i)
data_frame = Task_E.pickDataClass(train_data_file_name,
letter_to_digit)
train_data_set_without_labels, train_y, test_data_set_without_labels, test_y, train_data_with_labels, test_data_with_labels = Task_E.splitData2TestTrain(
data_frame, 39, 9)
centroid_data_frame_train = deepcopy(train_data_with_labels)
centroid_data_frame_test = deepcopy(test_data_with_labels)
# make_file_and_save_data_train = Task_E.store(train_data_set_without_labels.T, train_y, 'jenil_train.csv')
# make_file_and_save_data_test = Task_E.store(test_data_set_without_labels.T, test_y, 'jenil_test.csv')
k = 5
knn_object = Knn(k)
data_with_euclidean_distance = knn_object.calculate_distance(
train_data_with_labels.values, test_data_with_labels.values)
accuracy = knn_object.get_accuracy([
(k['Test Label'], k['Classification'])
for k in data_with_euclidean_distance
])
print('Accuracy of Knn is:', accuracy)
# Linear Regression
linear_regression_object = LinearRegression.LinearRegression()
N_train, L_train, Xtrain = len(
train_y), train_y, train_data_set_without_labels.T
N_test, Ytest, Xtest = len(
test_y), test_y, test_data_set_without_labels.T
Ytrain = linear_regression_object.indicator_matrix(L_train)
linear_regression_object.accuracy(N_train, N_test, Xtrain, Xtest,
