apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-15 06:52:39 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

made SGD work with new get/set param

Browse source
This commit is contained in:
Nicolo Fusi 2013-01-31 11:13:25 +00:00
parent 31b6c06a72
commit 7fdfc244f6
1 changed files with 4 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

8
GPy/inference/SGD.py
View file

@ -29,7 +29,7 @@ class opt_SGD(Optimizer):
self.batch_size = batch_size self.batch_size = batch_size
self.self_paced = self_paced self.self_paced = self_paced
num_params = len(self.model.get_param()) num_params = len(self.model._get_params())
if isinstance(self.learning_rate, float): if isinstance(self.learning_rate, float):
self.learning_rate = np.ones((num_params,)) * self.learning_rate self.learning_rate = np.ones((num_params,)) * self.learning_rate
@ -135,7 +135,7 @@ class opt_SGD(Optimizer):
import pdb; pdb.set_trace() import pdb; pdb.set_trace()
if model_name == 'Bayesian_GPLVM': if model_name == 'Bayesian_GPLVM':
self.model.trYYT = np.sum(np.square(self.model.Y)) self.model.trYYT = np.sum(np.square(self.model.Y))
if self.model.N == 0: if self.model.N == 0:
return 0, step, self.model.N return 0, step, self.model.N
@ -152,7 +152,7 @@ class opt_SGD(Optimizer):
return f, step, self.model.N return f, step, self.model.N
def opt(self, f_fp=None, f=None, fp=None): def opt(self, f_fp=None, f=None, fp=None):
self.x_opt = self.model.get_param() self.x_opt = self.model._get_params()
X, Y = self.model.X.copy(), self.model.Y.copy() X, Y = self.model.X.copy(), self.model.Y.copy()
N, Q = self.model.X.shape N, Q = self.model.X.shape
D = self.model.Y.shape[1] D = self.model.Y.shape[1]
@ -168,7 +168,7 @@ class opt_SGD(Optimizer):
b = len(features)/self.batch_size b = len(features)/self.batch_size
features = [features[i::b] for i in range(b)] features = [features[i::b] for i in range(b)]
step = np.zeros_like(self.model.get_param()) step = np.zeros_like(self.model._get_params())
LL = [] LL = []
count = 0 count = 0