def test(train_data_file_name, attribute_file_name, test_data_file_name,
optimizer):
data = read_data(train_data_file_name)
attributes = Attribute.read_attribute(attribute_file_name, data)
tmp_attributes = copy.deepcopy(attributes)
root = tree_generate(data, tmp_attributes, attributes, optimizer)
right = 0
# bfs(root)
test_data = read_data(test_data_file_name)
Attribute.change_types(attributes, test_data)
for d in test_data:
category, probability = tree_test(root, d, attributes)
pmax = -1
index = -1
for i in range(len(probability)):
# print(probability[i])
if probability[i] > pmax:
pmax = probability[i]
index = i
# print(category[index], d.value[-1])
if category[index] == d.value[-1]:
right += 1
print("correct percentage: %f" % (right / len(test_data)))
