apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-06-08 15:05:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

added centering and fixed serious bug

Browse source
This commit is contained in:
Nicolo Fusi 2013-02-07 15:50:45 +00:00
parent baa06181e8
commit 7b1e1eb18b
1 changed files with 20 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

30
GPy/inference/SGD.py
View file

@ -18,7 +18,7 @@ class opt_SGD(Optimizer):
"""
def __init__(self, start, iterations = 10, learning_rate = 1e-4, momentum = 0.9, model = None, messages = False, batch_size = 1, self_paced = False, **kwargs):
def __init__(self, start, iterations = 10, learning_rate = 1e-4, momentum = 0.9, model = None, messages = False, batch_size = 1, self_paced = False, center = True, **kwargs):
self.opt_name = "Stochastic Gradient Descent"
self.model = model
@ -30,6 +30,7 @@ class opt_SGD(Optimizer):
self.messages = messages
self.batch_size = batch_size
self.self_paced = self_paced
self.center = center
num_params = len(self.model._get_params())
if isinstance(self.learning_rate, float):
@ -48,7 +49,7 @@ class opt_SGD(Optimizer):
return status
def non_null_samples(self, data):
return np.isnan(data).sum(axis=1) == 0
return (np.isnan(data).sum(axis=1) == 0)
def check_for_missing(self, data):
return np.isnan(data).sum() > 0
@ -138,8 +139,9 @@ class opt_SGD(Optimizer):
j = self.subset_parameter_vector(self.x_opt, samples, shapes)
self.model.X = X[samples]
# self.model.Y -= self.model.Y.mean() # <----------------- WARNING!!!!
# self.model.Y /= self.model.Y.std()
if self.center:
self.model.Y -= self.model.Y.mean()
self.model.Y /= self.model.Y.std()
model_name = self.model.__class__.__name__
if model_name == 'Bayesian_GPLVM':
@ -161,7 +163,7 @@ class opt_SGD(Optimizer):
return f, step, self.model.N
def opt(self, f_fp=None, f=None, fp=None):
self.x_opt = self.model._get_params()
self.x_opt = self.model._get_params_transformed()
X, Y = self.model.X.copy(), self.model.Y.copy()
N, Q = self.model.X.shape
D = self.model.Y.shape[1]
@ -172,9 +174,14 @@ class opt_SGD(Optimizer):
missing_data = self.check_for_missing(self.model.Y)
self.model.Youter = None # this is probably not very efficient
self.model.YYT = None
step = np.zeros_like(self.model._get_params())
num_params = self.model._get_params()
if self.center:
print "WARNING: centering the data"
for it in range(self.iterations):
step = np.zeros_like(num_params)
if it == 0 or self.self_paced is False:
features = np.random.permutation(Y.shape[1])
else:
@ -189,12 +196,9 @@ class opt_SGD(Optimizer):
for j in features:
count += 1
self.model.D = len(j)
self.model.Y = Y[:, j:j+1]
self.model.Y = Y[:, j]
# self.model.trYYT = np.sum(np.square(self.model.Y))
if missing_data or sparse_matrix:
# if self.model.Y.std() == 0.0 or self.model.Y.shape[0] == 0: <--- not sure about this
# continue
shapes = self.get_param_shapes(N, Q)
f, step, Nj = self.step_with_missing_data(f_fp, X, step, shapes, sparse_matrix)
else:
@ -206,6 +210,12 @@ class opt_SGD(Optimizer):
if self.messages == 2:
status = "evaluating {feature: 5d}/{tot: 5d} \t f: {f: 2.3f} \t non-missing: {nm: 4d}\r".format(feature = count, tot = len(features), f = f, nm = Nj)
# TODO: remove this, it's only for debugging
if self.model.__class__.__name__ == 'Bayesian_GPLVM':
beta = np.exp(self.x_opt)[-7]
status = "evaluating {feature: 5d}/{tot: 5d} \t f: {f: 2.3f} \t non-missing: {nm: 4d} \t inv_bbeta: {beta: 1.5f}\r".format(feature = count, tot = len(features), f = f, nm = Nj, beta = 1./beta)
sys.stdout.write(status)
sys.stdout.flush()
last_printed_count = count