def test_get_coefs(self, data):
model = VW()
model.fit(data.x, data.y)
weights = model.get_coefs()
print weights.data
assert np.allclose(weights.indices,
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 116060])
def test_predict(self, data):
raw_model = VW()
raw_model.fit(data.x, data.y)
model = VWRegressor()
model.fit(data.x, data.y)
assert np.allclose(raw_model.predict(data.x), model.predict(data.x))
def test_get_coefs(self, data):
model = VW()
model.fit(data.x, data.y)
weights = model.get_coefs()
print weights.data
assert np.allclose(weights.indices, [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 116060])
assert np.allclose(weights.data, [0.11553502, -0.0166647, -0.00349924, 0.06911729, 0.00252684,
-0.00826817, 0.01991862, -0.02473332, 0.00483846, -0.04616702, -0.00744559])
def test_predict(self, data):
raw_model = VW()
raw_model.fit(data.x, data.y)
model = VWRegressor()
model.fit(data.x, data.y)
assert np.allclose(raw_model.predict(data.x), model.predict(data.x))
def test_predict(self, data):
raw_model = VW()
raw_model.fit(data.x, data.y)
model = VWRegressor()
model.fit(data.x, data.y)
assert np.allclose(raw_model.predict(data.x), model.predict(data.x))
# ensure model can make multiple calls to predict
assert np.allclose(raw_model.predict(data.x), model.predict(data.x))
def test_oaa(self):
X = [
'1 | feature1:2.5', '2 | feature1:0.11 feature2:-0.0741',
'3 | feature3:2.33 feature4:0.8 feature5:-3.1',
'1 | feature2:-0.028 feature1:4.43',
'2 | feature5:1.532 feature6:-3.2'
]
model = VW(convert_to_vw=False, oaa=3)
model.fit(X)
assert np.allclose(model.predict(X), [1., 2., 3., 1., 2.])
def test_oaa(self):
X = [
"1 | feature1:2.5",
"2 | feature1:0.11 feature2:-0.0741",
"3 | feature3:2.33 feature4:0.8 feature5:-3.1",
"1 | feature2:-0.028 feature1:4.43",
"2 | feature5:1.532 feature6:-3.2",
]
model = VW(convert_to_vw=False, oaa=3)
model.fit(X)
assert np.allclose(model.predict(X), [1.0, 2.0, 3.0, 1.0, 2.0])
def test_decision_function(self, data):
classes = np.array([-1., 1.])
raw_model = VW(loss_function='logistic')
raw_model.fit(data.x, data.y)
predictions = raw_model.predict(data.x)
class_indices = (predictions > 0).astype(np.int)
class_predictions = classes[class_indices]
model = VWClassifier()
model.fit(data.x, data.y)
assert np.allclose(class_predictions, model.predict(data.x))
def test_decision_function(self, data):
classes = np.array([-1., 1.])
raw_model = VW(loss_function='logistic')
raw_model.fit(data.x, data.y)
predictions = raw_model.predict(data.x)
class_indices = (predictions > 0).astype(np.int)
class_predictions = classes[class_indices]
model = VWClassifier()
model.fit(data.x, data.y)
assert np.allclose(class_predictions, model.predict(data.x))
def test_predict(self, data):
raw_model = VW()
raw_model.fit(data.x, data.y)
model = VWRegressor()
model.fit(data.x, data.y)
assert np.allclose(raw_model.predict(data.x), model.predict(data.x))
# ensure model can make multiple calls to predict
assert np.allclose(raw_model.predict(data.x), model.predict(data.x))
def test_set_params(self):
model = VW()
assert 'l' not in model.params
model.set_params(l=0.1)
assert model.params['l'] == 0.1
# confirm model params reset with new construction
model = VW()
assert 'l' not in model.params
def test_predict(self, data):
model = VW(loss_function='logistic')
model.fit(data.x, data.y)
assert np.isclose(model.predict(data.x[:1][:1])[0], 0.406929)
def test_fit(self, data):
model = VW(loss_function='logistic')
assert not hasattr(model, 'fit_')
model.fit(data.x, data.y)
assert model.fit_
def test_predict_not_fit(self, data):
model = VW(loss_function='logistic')
with pytest.raises(NotFittedError):
model.predict(data.x[0], data.y[0])
def test_fit(self, data):
model = VW(loss_function="logistic")
assert not hasattr(model, "fit_")
model.fit(data.x, data.y)
assert model.fit_
def test_get_coefs(self, data):
model = VW()
model.fit(data.x, data.y)
weights = model.get_coefs()
print weights.data
assert np.allclose(weights.indices, [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 116060])
def test_get_intercept(self, data):
model = VW()
model.fit(data.x, data.y)
intercept = model.get_intercept()
assert np.isclose(intercept, -0.00744559)
def test_predict_no_convert(self):
model = VW(loss_function='logistic', convert_to_vw=False)
model.fit(['-1 | bad', '1 | good'])
assert np.isclose(model.predict(['| good'])[0], 0.245515)
def test_set_params(self):
model = VW()
assert 'l' not in model.params
model.set_params(l=0.1)
assert model.params['l'] == 0.1
def test_passes(self, data):
n_passes = 2
model = VW(loss_function='logistic', passes=n_passes)
assert model.passes_ == n_passes
model.fit(data.x, data.y)
weights = model.get_coefs()
model = VW(loss_function='logistic')
# first pass weights should not be the same
model.fit(data.x, data.y)
assert not np.allclose(weights.data, model.get_coefs().data)
# second pass weights should match
model.fit(data.x, data.y)
assert np.allclose(weights.data, model.get_coefs().data)
def test_fit(self, data):
model = VW(loss_function='logistic')
assert not hasattr(model, 'fit_')
model.fit(data.x, data.y)
assert model.fit_
def test_init(self):
assert isinstance(VW(), VW)
def test_get_intercept(self, data):
model = VW()
model.fit(data.x, data.y)
intercept = model.get_intercept()
assert isinstance(intercept, float)
def test_delete(self):
raw_model = VW()
del raw_model
def test_predict(self, data):
model = VW(loss_function='logistic')
model.fit(data.x, data.y)
assert np.isclose(model.predict(data.x[:1][:1])[0], 0.406929)
def test_predict_not_fit(self, data):
model = VW(loss_function='logistic')
with pytest.raises(NotFittedError):
model.predict(data.x[0])
def test_predict_no_convert(self):
model = VW(loss_function="logistic", convert_to_vw=False)
model.fit(["-1 | bad", "1 | good"])
assert np.isclose(model.predict(["| good"])[0], 0.245515)
def test_predict_no_convert(self):
model = VW(loss_function='logistic')
model.fit(['-1 | bad', '1 | good'], convert_to_vw=False)
assert np.isclose(model.predict(['| good'], convert_to_vw=False)[0], 0.245515)
