precision = tp / (tp + fp) f_score = (2 * (precision * recall)) / (precision + recall) print("The recall is: {0}".format(recall)) print("The precision is: {0}".format(precision)) print("The F-score is: {0}".format(f_score)) if __name__ == '__main__': #location containing the training set location = os.path.dirname(os.path.abspath(__file__)) training_set_location = os.path.join(location, 'extras/files/training_set1.csv') #read training data training_set = read_csv(training_set_location) #remove the data lables labels = training_set.pop(0) neuron = NeuralNetwork(training_set, 1, 0.1, 0.2) print(neuron.weights) neuron.output_weights() print(neuron.normalized) neuron.iterate() #neuron.test() #backprop # neuron = NeuralNetwork(training_set, 2, 0.1, 0.8, bias = 0, hidden_layer = 2) # print(neuron.weights)
to_consider = results[:neigbours] #since its an ordered list to_consider = [x[-1] for x in to_consider] voting_results = {data: to_consider.count(data) for data in to_consider} max_key = max(voting_results) return max_key if __name__ == '__main__': file_loation = os.path.dirname(os.path.abspath(__file__)) location = os.path.join(file_loation, 'extras/files/training_set1.csv') #get the training set training_set = read_csv(location) #remove the data lables labels = training_set.pop(0) #get the users input query_set, neigbours = get_input(labels) #calculate euclidean distances response = get_euclidean_distances(training_set, query_set) #voting winning_label = vote(response, neigbours) print("The query set belongs to the label '{0}'.".format(winning_label))