def train_and_predict(self, n_components=1, threshold=0.001):
trainer = Trainer()
trainer.train(self.train_data, n_components=n_components)
predictor = Predictor(trainer=trainer.model)
y_pred = list()
for X in self.X_test:
result = predictor.predict(X, threshold=threshold)
y_pred.append(1 if result['is_anomaly'] else 0)
predictor.init_result()
return y_pred
def test_closs_validation(self):
trainer = Trainer()
kf = KFold(n_splits=3)
self.features, self.labels = fetch_kddcup99(subset="http",
return_X_y=True)
self.labels = list(
map(lambda label: 0 if label == b"normal." else 1, self.labels))
self.labels = np.array(self.labels)
for train_index, test_index in kf.split(self.features, self.labels):
train_data = self.features[train_index]
test_data = self.features[test_index]
train_label = self.labels[train_index]
test_label = self.labels[test_index]
trainer.train(train_data)
result = trainer.model.predict(test_data)
accuracy = accuracy_score(test_label, result)
print("正解率＝", accuracy)
assert accuracy > 0.8
def setUp(self):
self.model_filename = "my_model_file.bin"
with open("data/train.pickle", mode="rb") as f:
self.train_data = pickle.load(f)["features"]
trainer = Trainer()
trainer.train(self.train_data)
trainer.save(self.model_filename)
def setUp(self):
self.model_filename = "my_model_file.bin"
with open("data/train.pickle",
mode="rb") as train_file, open("data/test.pickle",
mode="rb") as test_file:
self.train_data = pickle.load(train_file)["features"]
self.test_data = pickle.load(test_file)
trainer = Trainer()
trainer.train(self.train_data, n_components=1)
trainer.save(self.model_filename)
def test_train_with_small_data(self):
trainer = Trainer()
trainer.train([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
trainer.save("tmp.model")
def test_train_with_insufficient_data(self):
trainer = Trainer()
with self.assertRaises(InsufficientTrainingDataError):
trainer.train([])
def test_train_and_save(self):
trainer = Trainer()
trainer.train(self.train_data["features"])
trainer.save("tmp.model")
def test_train_with_small_data(self):
trainer = Trainer()
with self.assertRaises(SmallTrainingDataError):
trainer.train([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]], n_components=3)
def test_train_with_current_size_data(self):
trainer = Trainer()
test_data = []
for i in range(20):
test_data.append([i, i + 1, i + 2])
trainer.train(test_data)
def test_train_with_small_data(self):
trainer = Trainer()
with self.assertRaises(TooMuchComponentError):
trainer.train([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])