def test_sexmachine_features(self):
s = DameSexmachine()
f = s.features("David")
self.assertEqual(f['has(a)'], True)
self.assertEqual(f['count(i)'], 1)
self.assertEqual(f['count(v)'], 1)
self.assertEqual(f['last_letter'], 'd')
self.assertEqual(f['first_letter'], 'd')
def test_sexmachine_gender_list_method_returns_correct_result(self):
s = DameSexmachine()
gl = s.gender_list()
self.assertEqual(gl, [1, 1, 1, 1, 2, 1, 0, 0, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1])
self.assertEqual(len(gl), 21)
self.assertEqual(s.females, 3)
self.assertEqual(s.males, 16)
self.assertEqual(s.unknown, 2)
def test_sexmachine_tree_load(self):
s = DameSexmachine()
m = s.tree_load()
predicted = m.predict([[
0, 0, 1, 0, 21, 0, 0, 0, 0, 34, 2, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0,
0, 0, 0, 2, 0, 0, 0, 34, 1, 0, 1
]])
a = np.array([0])
self.assertEqual(predicted, a)
def test_sexmachine_sgd_load(self):
s = DameSexmachine()
m = s.sgd_load()
predicted = m.predict([[
0, 0, 1, 0, 21, 0, 0, 0, 0, 34, 2, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0,
0, 0, 0, 2, 0, 0, 0, 34, 1, 0, 1
]])
n = np.array([1])
self.assertEqual(n, predicted)
def test_sexmachine_bernoulliNB_load(self):
s = DameSexmachine()
m = s.bernoulliNB_load()
predicted = m.predict([[
0, 0, 1, 0, 21, 0, 0, 0, 0, 34, 2, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0,
0, 0, 0, 2, 0, 0, 0, 34, 1, 0, 1
]])
n = np.array([2])
self.assertTrue(np.array_equal(predicted, n))
def test_sexmachine_adaboost_load(self):
s = DameSexmachine()
m = s.adaboost_load()
predicted = m.predict([[
0, 0, 1, 0, 21, 0, 0, 0, 0, 34, 2, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0,
0, 0, 0, 2, 0, 0, 0, 34, 1, 0, 1
]])
n = np.array([1])
self.assertTrue(np.array_equal(predicted, n))
def test_dame_sexmachine_guess_list_method_returns_correct_result(self):
ds = DameSexmachine()
self.assertEqual(['male', 'male', 'male', 'male', 'female', 'male', 'female', 'female', 'male', 'male', 'male', 'male', 'male', 'male', 'female', 'male', 'male', 'male', 'female', 'male', 'male'], ds.guess_list(path="files/names/partial.csv", binary=False))
self.assertEqual([1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1], ds.guess_list(path="files/names/partial.csv",binary=True))
self.assertEqual([1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1], ds.guess_list(path="files/names/partial.csv",binary=True, ml="nltk"))
# sgd_model = ds.sgd_load()
# self.assertEqual([0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 0], ds.guess_list(path="files/names/partial.csv",binary=True, ml="sgd"))
svc_model = ds.svc_load()
self.assertEqual([1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1], ds.guess_list(path="files/names/partial.csv",binary=True, ml="svc"))
def test_sexmachine_multinomialNB_load_method_returns_correct_result(self):
s = DameSexmachine()
m = s.multinomialNB_load()
array = [[ 0, 0, 1, 0, 21, 0, 0, 0, 0, 34, 2, 0, 0, 0, 0, 0, 0,
0, 0, 0, 5, 0, 0, 0, 0, 0, 2, 0, 0, 0, 34, 1, 0, 0],
[ 0, 0, 0, 0, 21, 0, 0, 0, 0, 34, 0, 0, 0, 0, 0, 1, 0,
0, 0, 0, 5, 0, 0, 1, 0, 0, 1, 0, 0, 1, 34, 0, 0, 1]]
predicted= m.predict(array)
n = np.array([1, 1])
self.assertTrue(np.array_equal(predicted, n))
def test_sexmachine_gender_list(self):
s = DameSexmachine()
gl = s.gender_list(path="files/names/partial.csv")
self.assertEqual(
gl,
[1, 1, 1, 1, 2, 1, 0, 0, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1])
self.assertEqual(len(gl), 21)
self.assertEqual(s.females, 3)
self.assertEqual(s.males, 16)
self.assertEqual(s.unknown, 2)
def test_dame_gender_guess_list(self):
ds = DameSexmachine()
self.assertEqual([
'male', 'male', 'male', 'male', 'male', 'male', 'female', 'female',
'male', 'male', 'male', 'male', 'male', 'male', 'male', 'male',
'male', 'male', 'female', 'male', 'male'
], ds.guess_list(path="files/names/partial.csv", binary=False))
self.assertEqual(
[1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1],
ds.guess_list(path="files/names/partial.csv",
binary=True,
ml="nltk"))
def test_dame_gender_confusion_matrix_gender(self):
ds = DameSexmachine()
cm = ds.confusion_matrix_gender(path="files/names/min.csv")
am = [[1, 0, 0], [0, 5, 0], [0, 5, 0]]
self.assertEqual(cm, am)
cm = ds.confusion_matrix_gender(path="files/names/min.csv", ml="nltk")
am = [[1, 0, 0], [0, 5, 0], [0, 5, 0]]
self.assertEqual(cm, am)
cm = ds.confusion_matrix_gender(path="files/names/min.csv",
jsonf="files/names/min.csv.json",
ml="nltk")
am = [[1, 0, 0], [0, 5, 0], [0, 5, 0]]
self.assertEqual(cm, am)
def test_sexmachine_features_int(self):
s = DameSexmachine()
dicc = s.features_int("David")
self.assertEqual(chr(dicc['last_letter']), 'd')
self.assertEqual(chr(dicc['first_letter']), 'd')
self.assertEqual(dicc['count(a)'], 1)
self.assertEqual(dicc['count(b)'], 0)
self.assertEqual(dicc['count(c)'], 0)
self.assertEqual(dicc['count(d)'], 2)
self.assertEqual(dicc['count(e)'], 0)
self.assertEqual(dicc['count(f)'], 0)
self.assertEqual(dicc['count(h)'], 0)
self.assertEqual(dicc['count(i)'], 1)
self.assertEqual(dicc['count(v)'], 1)
self.assertTrue(dicc['count(a)'] > 0)
self.assertTrue(dicc['vocals'], 2)
self.assertTrue(dicc['consonants'], 3)
self.assertEqual(dicc['first_letter_vocal'], 0)
self.assertEqual(dicc['last_letter_vocal'], 0)
self.assertTrue(len(dicc.values()) > 30)
def test_dame_sexmachine_guess(self):
s = DameSexmachine()
self.assertEqual(s.guess("David"), 'male')
self.assertEqual(s.guess("David", binary=True, ml="svc"), 1)
self.assertEqual(s.guess("Laura", binary=True, ml="svc"), 0)
self.assertEqual(s.guess("Palabra", binary=True, ml="svc"), 1)
self.assertEqual(s.guess("Laura", binary=True, ml="sgd"), 0)
self.assertEqual(s.guess("Palabra", binary=True, ml="svc"), 1)
self.assertEqual(s.guess("David", binary=True, ml="gaussianNB"), 1)
self.assertEqual(s.guess("David", binary=True, ml="multinomialNB"), 1)
self.assertEqual(s.guess("David", binary=True, ml="bernoulliNB"), 1)
self.assertEqual(s.guess("Laura"), 'female')
# With accents:
self.assertEqual(s.guess("Inés"), 'female')
# Without accents:
self.assertEqual(s.guess("Ines"), 'female')
self.assertEqual(s.guess("Nodiccionario"), 'male')
self.assertEqual(s.guess("Nadiccionaria"), 'female')
self.assertEqual(s.guess("David", binary=True), 1)
self.assertEqual(s.guess("Laura", binary=True), 0)
self.assertEqual(s.guess("Nodiccionario", binary=True), 1)
self.assertEqual(s.guess("Nadiccionaria", binary=True), 0)
def test_dame_sexmachine_guess_method_returns_correct_result(self):
s = DameSexmachine()
self.assertEqual(s.guess("David"), 'male')
self.assertEqual(s.guess("David", binary=True, ml="svc"), 1)
self.assertEqual(s.guess("Laura", binary=True, ml="svc"), 0)
self.assertEqual(s.guess("Palabra", binary=True, ml="svc"), 1)
self.assertEqual(s.guess("Inés", binary=True, ml="svc"), 0)
self.assertEqual(s.guess("David", binary=True, ml="sgd"), 1)
self.assertEqual(s.guess("Laura", binary=True, ml="sgd"), 0)
self.assertEqual(s.guess("Palabra", binary=True, ml="svc"), 1)
self.assertEqual(s.guess("David", binary=True, ml="gaussianNB"), 1)
self.assertEqual(s.guess("David", binary=True, ml="multinomialNB"), 1)
self.assertEqual(s.guess("David", binary=True, ml="bernoulliNB"), 1)
self.assertEqual(s.guess("Laura"), 'female')
self.assertEqual(s.guess("Inés"), 'female') # Con acento
self.assertEqual(s.guess("Ines"), 'female') # Sin acento
self.assertEqual(s.guess("Nodiccionario"), 'male') # Sin estar en diccionario
self.assertEqual(s.guess("Nadiccionaria"), 'female') # En diccionario
self.assertEqual(s.guess("David", binary=True), 1)
self.assertEqual(s.guess("Laura", binary=True), 0)
self.assertEqual(s.guess("Nodiccionario", binary=True), 1)
self.assertEqual(s.guess("Nadiccionaria", binary=True), 0)
def test_sexmachine_classifier_load(self):
s = DameSexmachine()
m = s.classifier_load()
n = s.features("David")
guess = m.classify(n)
self.assertTrue(1, n)
def test_sexmachine_confusion_matrix_method_returns_correct_result(self):
s = DameSexmachine()
cm = s.confusion_matrix()
am = np.array([[3, 0, 0],[1, 15, 0],[1, 1, 0]])
self.assertTrue(np.array_equal(cm,am))
def test_dame_sexmachine_features_int(self):
s = DameSexmachine()
f = s.features_int("David")
self.assertTrue(len(f) > 0)
def test_sexmachine_string2gender_method_returns_correct_result(self):
s = DameSexmachine()
gender1 = s.string2gender("Arroyo Menéndez, David")
gender2 = s.string2gender("xxxxx Laura")
self.assertTrue(gender1, 'male')
self.assertTrue(gender2, 'female')
def test_dame_sexmachine_features_int_method_returns_correct_result(self):
s = DameSexmachine()
f = s.features_int("David")
self.assertTrue(f['syllables'] > 0)
self.assertTrue(len(f) > 0)
def test_sexmachine_bernoulliNB_load_method_returns_correct_result(self):
s = DameSexmachine()
m = s.bernoulliNB_load()
predicted = m.predict([[ 0, 0, 1, 0, 21, 0, 0, 0, 0, 34, 2, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0, 0, 0, 0, 2, 0, 0, 0, 34, 1, 0, 1, 0]])
n = np.array([2])
self.assertTrue(np.array_equal(predicted, n))
def test_sexmachine_sgd_load_method_returns_correct_result(self):
s = DameSexmachine()
m = s.sgd_load()
predicted = m.predict([[ 0, 0, 1, 0, 21, 0, 0, 0, 0, 34, 2, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0, 0, 0, 0, 2, 0, 0, 0, 34, 1, 0, 1, 0]])
n = np.array([2])
self.assertEqual(n, predicted)
def test_sexmachine_accuracy(self):
s = DameSexmachine()
self.assertTrue(s.accuracy(path="files/names/partial.csv") > 0.5)
def test_string2array_method_returns_correct_result(self):
array = "muchos espacios en blanco"
s = DameSexmachine()
arr = s.string2array(array)
self.assertEqual(["muchos", "espacios", "en", "blanco"], arr)
def test_dame_sexmachine_guess_surname_method_returns_correct_result(self):
s = DameSexmachine()
self.assertTrue(s.guess_surname("Smith"))
def test_sexmachine_classifier(self):
s = DameSexmachine()
classifier = s.classifier(locale="us")
n = s.features("David")
guess = classifier.classify(n)
self.assertTrue(1, n)
def test_sexmachine_features_list(self):
s = DameSexmachine()
fl = s.features_list()
self.assertTrue(len(fl) > 20)
def test_sexmachine_accuracy_method_returns_correct_result(self):
s = DameSexmachine()
self.assertTrue(s.accuracy(path="files/names/partial.csv") > 0.5)
def test_sexmachine_features_list_all(self):
s = DameSexmachine()
fl = s.features_list(path="files/names/all.csv")
self.assertTrue(len(fl) > 1000)
from app.dame_sexmachine import DameSexmachine
from app.dame_utils import DameUtils
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('ml',
choices=[
'nltk', 'svc', 'sgd', 'gaussianNB', 'multinomialNB',
'bernoulliNB', 'forest', 'tree', 'mlp'
])
parser.add_argument('--noshow', dest='noshow', action='store_true')
parser.add_argument('--verbose', default=False, action="store_true")
args = parser.parse_args()
ds = DameSexmachine()
X = np.array(ds.features_list(path="files/names/allnoundefined.csv"))
y = ds.gender_list(path="files/names/allnoundefined.csv")
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
if (args.verbose):
print(X)
print(y)
if (args.ml == "svc"):
svc = SVC(random_state=42)
svc.fit(X_train, y_train)
svc_disp = plot_roc_curve(svc, X_test, y_test)
elif (args.ml == "forest"):
rfc = RandomForestClassifier(n_estimators=10, random_state=42)
def test_sexmachine_gender_list_all(self):
s = DameSexmachine()
gl = s.gender_list(path="files/names/all.csv")
self.assertTrue(len(gl) > 1000)