def test_naive_bayes(hypothises):
fp = FileProcessor(test_filepath, ' ')
parsed_lines = fp.get_lines_as_array()
results = []
for row in parsed_lines:
exclude_label = row[1:]
max_sum = -float('Inf')
max_label = -1
for label in label_mappings:
label_instance = hypothises[label]
log_prior = log(label_instance.get_prior(), 2)
densities = label_instance.get_densities()
log_sum = 0
for word in exclude_label:
log_sum += log(densities[word], 2)
cur_sum = log_sum + log_prior
if cur_sum > max_sum:
max_sum = cur_sum
max_label = label
results.append(label_mappings[max_label])
fp.generate_output(output_filepath, results)
def process_data(training_file, testing_file, training_output, testing_output, attribute_descriptors):
delimiter = ','
training_file_processor = FileProcessor(training_file, delimiter)
testing_file_processor = FileProcessor(testing_file, delimiter)
training_lines = training_file_processor.get_lines_as_array()
testing_lines = testing_file_processor.get_lines_as_array()
all_lines = training_lines + testing_lines
knn_processor = KNNProcess(all_lines, attribute_descriptors)
imputed_lines = map(lambda line_no: knn_processor.replace_missing_line(line_no), range(0, len(all_lines)))
normalized_lines = map(lambda line: knn_processor.normalize_line(line), imputed_lines)
for line_no, line in enumerate(normalized_lines[:len(training_lines)]):
training_file_processor.set_line(line_no, line)
for line_no, line in enumerate(normalized_lines[len(training_lines):]):
testing_file_processor.set_line(line_no, line)
if training_file_processor.generate_output(training_output) and testing_file_processor.generate_output(testing_output):
print 'Success!'
def generate_labels():
fp = FileProcessor(testing_data_filepath, ' ')
rows = fp.parse_input_file()
expected = []
for row in rows:
expected.append(row[0])
if fp.generate_output(labels_output_filepath, expected):
return True
def test_logistic_regression(w):
if not generate_labels():
return
fp = FileProcessor(testing_data_filepath, ' ')
rows = fp.parse_input_file()
output = []
expected = []
labels = get_labels()
for row in rows:
expected.append(row[0])
row = row[1:]
sum_val = w[0]
for feature in row:
feature_id = int(feature.split(':')[0])
sum_val += w[feature_id]
if sigmoid(sum_val) >= 0.5:
output.append(labels[0])
else:
output.append(labels[1])
if fp.generate_output(output_filepath, output):
print 'Successfully generated predictions.lr'
