def method_instance(self):
if not hasattr(self, "_method_instance"):
n, p = self.X.shape
self._method_instance = randomized_modelQ(self.feature_cov * n,
self.X,
self.Y,
self.lagrange * np.sqrt(n),
randomizer_scale=self.randomizer_scale * np.std(self.Y) * np.sqrt(n))
return self._method_instance
def method_instance(self):
if not hasattr(self, "_method_instance"):
# draw sample of X for semi-supervised method
_chol = self._chol
p = _chol.shape[0]
Q = 0
batch_size = int(self.B/10)
for _ in range(10):
X_semi = np.random.standard_normal((batch_size, p)).dot(_chol.T)
Q += X_semi.T.dot(X_semi)
Q += self.X.T.dot(self.X)
Q /= (10 * batch_size + self.X.shape[0])
n, p = self.X.shape
self._method_instance = randomized_modelQ(Q * n,
self.X,
self.Y,
self.lagrange * np.sqrt(n),
randomizer_scale=self.randomizer_scale * np.std(self.Y) * np.sqrt(n))
return self._method_instance