apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-24 14:15:14 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Other changes.

Browse source
This commit is contained in:
Ricardo Andrade 2013-01-28 17:47:08 +00:00
parent fad0e07624
commit 29ec128c9d
7 changed files with 164 additions and 143 deletions
Download patch file Download diff file Expand all files Collapse all files

12
GPy/examples/ep_fix.py
View file

@ -11,11 +11,9 @@ import GPy
pb.ion()
pb.close('all')
default_seed=10000
model_type='Full'
inducing=4
seed=default_seed
"""Simple 1D classification example.
:param model_type: type of model to fit ['Full', 'FITC', 'DTC'].
:param seed : seed value for data generation (default is 4).
@ -23,21 +21,19 @@ seed=default_seed
:param inducing : number of inducing variables (only used for 'FITC' or 'DTC').
:type inducing: int
"""
data = GPy.util.datasets.toy_linear_1d_classification(seed=seed)
data = GPy.util.datasets.toy_linear_1d_classification(seed=0)
likelihood = GPy.inference.likelihoods.probit(data['Y'][:, 0:1])
m = GPy.models.GP(data['X'],likelihood=likelihood)
#m = GPy.models.GP(data['X'],Y=likelihood.Y)
#m = GPy.models.GP(data['X'],likelihood.Y)
m.constrain_positive('var')
m.constrain_positive('len')
m.tie_param('lengthscale')
m.ensure_default_constraints()
if not isinstance(m.likelihood,GPy.inference.likelihoods.gaussian):
m.approximate_likelihood()
print m.checkgrad()
# Optimize and plot
m.optimize()
#m.em(plot_all=False) # EM algorithm
m.plot()
m.plot(samples=3)
print(m)

2
GPy/examples/poisson.py
View file

@ -31,7 +31,7 @@ Y = F + E
pb.plot(X,F,'k-')
pb.plot(X,Y,'ro')
pb.figure()
likelihood = GPy.inference.likelihoods.poisson(Y,scale=4.)
likelihood = GPy.inference.likelihoods.poisson(Y,scale=6.)
m = GPy.models.GP(X,likelihood=likelihood)
#m = GPy.models.GP(data['X'],Y=likelihood.Y)

34
GPy/examples/sparse_ep_fix.py
View file

@ -31,46 +31,18 @@ noise = GPy.kern.white(1)
kernel = rbf + noise
# create simple GP model
#m1 = GPy.models.sparse_GP_regression(X, Y, kernel, M=M)
m1 = GPy.models.sparse_GP(X, kernel, M=M,likelihood= likelihood)
#m1 = GPy.models.sparse_GP(X, Y, kernel, M=M)
m1 = GPy.models.sparse_GP(X,Y=None, kernel=kernel, M=M,likelihood= likelihood)
print m1.checkgrad()
# contrain all parameters to be positive
m1.constrain_positive('(variance|lengthscale|precision)')
#m1.constrain_positive('(variance|lengthscale)')
#m1.constrain_fixed('prec',10.)
#check gradient FIXME unit test please
m1.checkgrad()
# optimize and plot
m1.optimize('tnc', messages = 1)
m1.plot()
# print(m1)
######################################
## 2 dimensional example
# # sample inputs and outputs
# X = np.random.uniform(-3.,3.,(N,2))
# Y = np.sin(X[:,0:1]) * np.sin(X[:,1:2])+np.random.randn(N,1)*0.05
# # construct kernel
# rbf = GPy.kern.rbf(2)
# noise = GPy.kern.white(2)
# kernel = rbf + noise
# # create simple GP model
# m2 = GPy.models.sparse_GP_regression(X,Y,kernel, M = 50)
# create simple GP model
# # contrain all parameters to be positive (but not inducing inputs)
# m2.constrain_positive('(variance|lengthscale|precision)')
# #check gradient FIXME unit test please
# m2.checkgrad()
# # optimize and plot
# pb.figure()
# m2.optimize('tnc', messages = 1)
# m2.plot()
# print(m2)