def predict(self, Xtest):
probs = np.random.rand(Xtest.shape[0])
ytest = utils.threshold_probs(probs)
return ytest
def predict(self, Xtest):
probs = np.random.rand(Xtest.shape[0])
ytest = utils.threshold_probs(probs)
return ytest
def predict(self, Xtest):
probs = utils.sigmoid(Xtest.dot(self.weights))
ytest = utils.threshold_probs(probs)
return ytest
def predict(self, Xtest):
ytest = utils.sigmoid(np.dot(Xtest, self.weights))
ytest = utils.threshold_probs(ytest)
return ytest
def predict(self, Xtest):
Rbf = RBF()
Xtest = Rbf.transform(Xtest)
ytest = utils.sigmoid(np.dot(Xtest, self.weights))
ytest = utils.threshold_probs(ytest)
return ytest
def predict(self, Xtest):
xw = np.dot(Xtest, self.weights)
sqrt_one_plus_xwSquare = utils.sqrt_one_plus_xwSquare
p = 0.5 * (1 + xw/sqrt_one_plus_xwSquare(xw))
p = utils.threshold_probs(p)
return p
def predict(self, Xtest):
ytest = 0.5 * (
1 + np.divide(np.dot(Xtest, self.weights),
np.sqrt(1 + np.square(np.dot(Xtest, self.weights)))))
ytest = utils.threshold_probs(ytest)
return ytest
def predict(self, Xtest):
probs = utils.sigmoid(Xtest.dot(self.weights))
ytest = utils.threshold_probs(probs)
return ytest
def predict(self, Xtest, weights=None):
if weights is None:
weights = self.weights
p = utils.sigmoid(np.dot(Xtest, weights))
p = utils.threshold_probs(p)
return p
def predict(self, Xtest):
probabilities = utils.custom_probs(np.dot(self.weights, Xtest.T))
return utils.threshold_probs(probabilities)
def predict(self, Xtest):
ytest = []
for i in range(0, len(Xtest)):
ytest.append(self.evaluate(Xtest[i]))
ytest = np.array(ytest)
return utils.threshold_probs(ytest)
def predict(self, Xtest):
xw = np.dot(Xtest, self.weights)
sqrt_one_plus_xwSquare = utils.sqrt_one_plus_xwSquare
p = 0.5 * (1 + xw / sqrt_one_plus_xwSquare(xw))
p = utils.threshold_probs(p)
return p
def predict(self, Xtest, weights=None):
if weights is None:
weights = self.weights
p = utils.sigmoid(np.dot(Xtest, weights))
p = utils.threshold_probs(p)
return p
def predict(self, Xtest):
p = utils.sigmoid(np.dot(Xtest, self.weights))
p = utils.threshold_probs(p)
return p
