mirror of
https://github.com/SheffieldML/GPy.git
synced 2026-05-24 14:15:14 +02:00
Merge branch 'devel' of github.com:SheffieldML/GPy into devel
Conflicts: GPy/core/fitc.py
This commit is contained in:
Ricardo
commit
c774432fee
56 changed files with 783 additions and 807 deletions
|
|
@ -12,7 +12,7 @@ default_seed = np.random.seed(123344)
|
|||
|
||||
def BGPLVM(seed=default_seed):
|
||||
N = 10
|
||||
M = 3
|
||||
num_inducing = 3
|
||||
Q = 2
|
||||
D = 4
|
||||
# generate GPLVM-like data
|
||||
|
|
@ -26,7 +26,7 @@ def BGPLVM(seed=default_seed):
|
|||
# k = GPy.kern.rbf(Q) + GPy.kern.bias(Q) + GPy.kern.white(Q, 0.00001)
|
||||
# k = GPy.kern.rbf(Q, ARD = False) + GPy.kern.white(Q, 0.00001)
|
||||
|
||||
m = GPy.models.BayesianGPLVM(Y, Q, kernel=k, M=M)
|
||||
m = GPy.models.BayesianGPLVM(Y, Q, kernel=k, num_inducing=num_inducing)
|
||||
m.constrain_positive('(rbf|bias|noise|white|S)')
|
||||
# m.constrain_fixed('S', 1)
|
||||
|
||||
|
|
@ -62,7 +62,7 @@ def GPLVM_oil_100(optimize=True):
|
|||
m.plot_latent(labels=m.data_labels)
|
||||
return m
|
||||
|
||||
def swiss_roll(optimize=True, N=1000, M=15, Q=4, sigma=.2, plot=False):
|
||||
def swiss_roll(optimize=True, N=1000, num_inducing=15, Q=4, sigma=.2, plot=False):
|
||||
from GPy.util.datasets import swiss_roll_generated
|
||||
from GPy.core.transformations import logexp_clipped
|
||||
|
||||
|
|
@ -100,11 +100,11 @@ def swiss_roll(optimize=True, N=1000, M=15, Q=4, sigma=.2, plot=False):
|
|||
S = (var * np.ones_like(X) + np.clip(np.random.randn(N, Q) * var ** 2,
|
||||
- (1 - var),
|
||||
(1 - var))) + .001
|
||||
Z = np.random.permutation(X)[:M]
|
||||
Z = np.random.permutation(X)[:num_inducing]
|
||||
|
||||
kernel = GPy.kern.rbf(Q, ARD=True) + GPy.kern.bias(Q, np.exp(-2)) + GPy.kern.white(Q, np.exp(-2))
|
||||
|
||||
m = BayesianGPLVM(Y, Q, X=X, X_variance=S, M=M, Z=Z, kernel=kernel)
|
||||
m = BayesianGPLVM(Y, Q, X=X, X_variance=S, num_inducing=num_inducing, Z=Z, kernel=kernel)
|
||||
m.data_colors = c
|
||||
m.data_t = t
|
||||
|
||||
|
|
@ -117,7 +117,7 @@ def swiss_roll(optimize=True, N=1000, M=15, Q=4, sigma=.2, plot=False):
|
|||
m.optimize('scg', messages=1)
|
||||
return m
|
||||
|
||||
def BGPLVM_oil(optimize=True, N=100, Q=5, M=25, max_f_eval=4e3, plot=False, **k):
|
||||
def BGPLVM_oil(optimize=True, N=100, Q=5, num_inducing=25, max_f_eval=4e3, plot=False, **k):
|
||||
np.random.seed(0)
|
||||
data = GPy.util.datasets.oil()
|
||||
from GPy.core.transformations import logexp_clipped
|
||||
|
|
@ -128,7 +128,7 @@ def BGPLVM_oil(optimize=True, N=100, Q=5, M=25, max_f_eval=4e3, plot=False, **k)
|
|||
Yn = Y - Y.mean(0)
|
||||
Yn /= Yn.std(0)
|
||||
|
||||
m = GPy.models.BayesianGPLVM(Yn, Q, kernel=kernel, M=M, **k)
|
||||
m = GPy.models.BayesianGPLVM(Yn, Q, kernel=kernel, num_inducing=num_inducing, **k)
|
||||
m.data_labels = data['Y'][:N].argmax(axis=1)
|
||||
|
||||
# m.constrain('variance|leng', logexp_clipped())
|
||||
|
|
@ -167,7 +167,7 @@ def oil_100():
|
|||
|
||||
|
||||
|
||||
def _simulate_sincos(D1, D2, D3, N, M, Q, plot_sim=False):
|
||||
def _simulate_sincos(D1, D2, D3, N, num_inducing, Q, plot_sim=False):
|
||||
x = np.linspace(0, 4 * np.pi, N)[:, None]
|
||||
s1 = np.vectorize(lambda x: np.sin(x))
|
||||
s2 = np.vectorize(lambda x: np.cos(x))
|
||||
|
|
@ -227,13 +227,13 @@ def bgplvm_simulation_matlab_compare():
|
|||
Y = sim_data['Y']
|
||||
S = sim_data['S']
|
||||
mu = sim_data['mu']
|
||||
M, [_, Q] = 3, mu.shape
|
||||
num_inducing, [_, Q] = 3, mu.shape
|
||||
|
||||
from GPy.models import mrd
|
||||
from GPy import kern
|
||||
reload(mrd); reload(kern)
|
||||
k = kern.linear(Q, ARD=True) + kern.bias(Q, np.exp(-2)) + kern.white(Q, np.exp(-2))
|
||||
m = BayesianGPLVM(Y, Q, init="PCA", M=M, kernel=k,
|
||||
m = BayesianGPLVM(Y, Q, init="PCA", num_inducing=num_inducing, kernel=k,
|
||||
# X=mu,
|
||||
# X_variance=S,
|
||||
_debug=False)
|
||||
|
|
@ -247,8 +247,8 @@ def bgplvm_simulation(optimize='scg',
|
|||
plot=True,
|
||||
max_f_eval=2e4):
|
||||
# from GPy.core.transformations import logexp_clipped
|
||||
D1, D2, D3, N, M, Q = 15, 8, 8, 100, 3, 5
|
||||
slist, Slist, Ylist = _simulate_sincos(D1, D2, D3, N, M, Q, plot)
|
||||
D1, D2, D3, N, num_inducing, Q = 15, 8, 8, 100, 3, 5
|
||||
slist, Slist, Ylist = _simulate_sincos(D1, D2, D3, N, num_inducing, Q, plot)
|
||||
|
||||
from GPy.models import mrd
|
||||
from GPy import kern
|
||||
|
|
@ -258,7 +258,7 @@ def bgplvm_simulation(optimize='scg',
|
|||
Y = Ylist[0]
|
||||
|
||||
k = kern.linear(Q, ARD=True) + kern.bias(Q, np.exp(-2)) + kern.white(Q, np.exp(-2)) # + kern.bias(Q)
|
||||
m = BayesianGPLVM(Y, Q, init="PCA", M=M, kernel=k, _debug=True)
|
||||
m = BayesianGPLVM(Y, Q, init="PCA", num_inducing=num_inducing, kernel=k, _debug=True)
|
||||
# m.constrain('variance|noise', logexp_clipped())
|
||||
m.ensure_default_constraints()
|
||||
m['noise'] = Y.var() / 100.
|
||||
|
|
@ -275,8 +275,8 @@ def bgplvm_simulation(optimize='scg',
|
|||
return m
|
||||
|
||||
def mrd_simulation(optimize=True, plot=True, plot_sim=True, **kw):
|
||||
D1, D2, D3, N, M, Q = 150, 200, 400, 500, 3, 7
|
||||
slist, Slist, Ylist = _simulate_sincos(D1, D2, D3, N, M, Q, plot_sim)
|
||||
D1, D2, D3, N, num_inducing, Q = 150, 200, 400, 500, 3, 7
|
||||
slist, Slist, Ylist = _simulate_sincos(D1, D2, D3, N, num_inducing, Q, plot_sim)
|
||||
|
||||
from GPy.models import mrd
|
||||
from GPy import kern
|
||||
|
|
@ -284,7 +284,7 @@ def mrd_simulation(optimize=True, plot=True, plot_sim=True, **kw):
|
|||
reload(mrd); reload(kern)
|
||||
|
||||
k = kern.linear(Q, [.05] * Q, ARD=True) + kern.bias(Q, np.exp(-2)) + kern.white(Q, np.exp(-2))
|
||||
m = mrd.MRD(Ylist, input_dim=Q, M=M, kernels=k, initx="", initz='permute', **kw)
|
||||
m = mrd.MRD(Ylist, input_dim=Q, num_inducing=num_inducing, kernels=k, initx="", initz='permute', **kw)
|
||||
|
||||
for i, Y in enumerate(Ylist):
|
||||
m['{}_noise'.format(i + 1)] = Y.var() / 100.
|
||||
|
|
@ -312,7 +312,7 @@ def brendan_faces():
|
|||
Yn /= Yn.std()
|
||||
|
||||
m = GPy.models.GPLVM(Yn, Q)
|
||||
# m = GPy.models.BayesianGPLVM(Yn, Q, M=100)
|
||||
# m = GPy.models.BayesianGPLVM(Yn, Q, num_inducing=100)
|
||||
|
||||
# optimize
|
||||
m.constrain('rbf|noise|white', GPy.core.transformations.logexp_clipped())
|
||||
|
|
@ -376,16 +376,16 @@ def cmu_mocap(subject='35', motion=['01'], in_place=True):
|
|||
# X /= X.std(axis=0)
|
||||
#
|
||||
# Q = 10
|
||||
# M = 30
|
||||
# num_inducing = 30
|
||||
#
|
||||
# kernel = GPy.kern.rbf(Q, ARD=True) + GPy.kern.bias(Q) + GPy.kern.white(Q)
|
||||
# m = GPy.models.BayesianGPLVM(X, Q, kernel=kernel, M=M)
|
||||
# m = GPy.models.BayesianGPLVM(X, Q, kernel=kernel, num_inducing=num_inducing)
|
||||
# # m.scale_factor = 100.0
|
||||
# m.constrain_positive('(white|noise|bias|X_variance|rbf_variance|rbf_length)')
|
||||
# from sklearn import cluster
|
||||
# km = cluster.KMeans(M, verbose=10)
|
||||
# km = cluster.KMeans(num_inducing, verbose=10)
|
||||
# Z = km.fit(m.X).cluster_centers_
|
||||
# # Z = GPy.util.misc.kmm_init(m.X, M)
|
||||
# # Z = GPy.util.misc.kmm_init(m.X, num_inducing)
|
||||
# m.set('iip', Z)
|
||||
# m.set('bias', 1e-4)
|
||||
# # optimize
|
||||
|
|
|
|||
|
|
@ -10,16 +10,16 @@ import numpy as np
|
|||
import GPy
|
||||
|
||||
|
||||
def toy_rbf_1d(max_nb_eval_optim=100):
|
||||
def toy_rbf_1d(optimizer='tnc', max_nb_eval_optim=100):
|
||||
"""Run a simple demonstration of a standard Gaussian process fitting it to data sampled from an RBF covariance."""
|
||||
data = GPy.util.datasets.toy_rbf_1d()
|
||||
|
||||
# create simple GP model
|
||||
# create simple GP Model
|
||||
m = GPy.models.GPRegression(data['X'],data['Y'])
|
||||
|
||||
# optimize
|
||||
m.ensure_default_constraints()
|
||||
m.optimize(max_f_eval=max_nb_eval_optim)
|
||||
m.optimize(optimizer, max_f_eval=max_nb_eval_optim)
|
||||
# plot
|
||||
m.plot()
|
||||
print(m)
|
||||
|
|
@ -29,7 +29,7 @@ def rogers_girolami_olympics(optim_iters=100):
|
|||
"""Run a standard Gaussian process regression on the Rogers and Girolami olympics data."""
|
||||
data = GPy.util.datasets.rogers_girolami_olympics()
|
||||
|
||||
# create simple GP model
|
||||
# create simple GP Model
|
||||
m = GPy.models.GPRegression(data['X'],data['Y'])
|
||||
|
||||
#set the lengthscale to be something sensible (defaults to 1)
|
||||
|
|
@ -48,7 +48,7 @@ def toy_rbf_1d_50(optim_iters=100):
|
|||
"""Run a simple demonstration of a standard Gaussian process fitting it to data sampled from an RBF covariance."""
|
||||
data = GPy.util.datasets.toy_rbf_1d_50()
|
||||
|
||||
# create simple GP model
|
||||
# create simple GP Model
|
||||
m = GPy.models.GPRegression(data['X'],data['Y'])
|
||||
|
||||
# optimize
|
||||
|
|
@ -64,7 +64,7 @@ def silhouette(optim_iters=100):
|
|||
"""Predict the pose of a figure given a silhouette. This is a task from Agarwal and Triggs 2004 ICML paper."""
|
||||
data = GPy.util.datasets.silhouette()
|
||||
|
||||
# create simple GP model
|
||||
# create simple GP Model
|
||||
m = GPy.models.GPRegression(data['X'],data['Y'])
|
||||
|
||||
# optimize
|
||||
|
|
@ -151,8 +151,8 @@ def coregionalisation_sparse(optim_iters=100):
|
|||
Y2 = -np.sin(X2) + np.random.randn(*X2.shape)*0.05
|
||||
Y = np.vstack((Y1,Y2))
|
||||
|
||||
M = 40
|
||||
Z = np.hstack((np.random.rand(M,1)*8,np.random.randint(0,2,M)[:,None]))
|
||||
num_inducing = 40
|
||||
Z = np.hstack((np.random.rand(num_inducing,1)*8,np.random.randint(0,2,num_inducing)[:,None]))
|
||||
|
||||
k1 = GPy.kern.rbf(1)
|
||||
k2 = GPy.kern.Coregionalise(2,2)
|
||||
|
|
@ -244,24 +244,24 @@ def _contour_data(data, length_scales, log_SNRs, kernel_call=GPy.kern.rbf):
|
|||
lls = []
|
||||
total_var = np.var(data['Y'])
|
||||
kernel = kernel_call(1, variance=1., lengthscale=1.)
|
||||
model = GPy.models.GPRegression(data['X'], data['Y'], kernel=kernel)
|
||||
Model = GPy.models.GPRegression(data['X'], data['Y'], kernel=kernel)
|
||||
for log_SNR in log_SNRs:
|
||||
SNR = 10.**log_SNR
|
||||
noise_var = total_var/(1.+SNR)
|
||||
signal_var = total_var - noise_var
|
||||
model.kern['.*variance'] = signal_var
|
||||
model['noise_variance'] = noise_var
|
||||
Model.kern['.*variance'] = signal_var
|
||||
Model['noise_variance'] = noise_var
|
||||
length_scale_lls = []
|
||||
|
||||
for length_scale in length_scales:
|
||||
model['.*lengthscale'] = length_scale
|
||||
length_scale_lls.append(model.log_likelihood())
|
||||
Model['.*lengthscale'] = length_scale
|
||||
length_scale_lls.append(Model.log_likelihood())
|
||||
|
||||
lls.append(length_scale_lls)
|
||||
|
||||
return np.array(lls)
|
||||
|
||||
def sparse_GP_regression_1D(N = 400, M = 5, optim_iters=100):
|
||||
def sparse_GP_regression_1D(N = 400, num_inducing = 5, optim_iters=100):
|
||||
"""Run a 1D example of a sparse GP regression."""
|
||||
# sample inputs and outputs
|
||||
X = np.random.uniform(-3.,3.,(N,1))
|
||||
|
|
@ -270,8 +270,8 @@ def sparse_GP_regression_1D(N = 400, M = 5, optim_iters=100):
|
|||
rbf = GPy.kern.rbf(1)
|
||||
noise = GPy.kern.white(1)
|
||||
kernel = rbf + noise
|
||||
# create simple GP model
|
||||
m = GPy.models.SparseGPRegression(X, Y, kernel, M=M)
|
||||
# create simple GP Model
|
||||
m = GPy.models.SparseGPRegression(X, Y, kernel, num_inducing=num_inducing)
|
||||
|
||||
m.ensure_default_constraints()
|
||||
|
||||
|
|
@ -280,7 +280,7 @@ def sparse_GP_regression_1D(N = 400, M = 5, optim_iters=100):
|
|||
m.plot()
|
||||
return m
|
||||
|
||||
def sparse_GP_regression_2D(N = 400, M = 50, optim_iters=100):
|
||||
def sparse_GP_regression_2D(N = 400, num_inducing = 50, optim_iters=100):
|
||||
"""Run a 2D example of a sparse GP regression."""
|
||||
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
|
||||
|
|
@ -290,8 +290,8 @@ def sparse_GP_regression_2D(N = 400, M = 50, optim_iters=100):
|
|||
noise = GPy.kern.white(2)
|
||||
kernel = rbf + noise
|
||||
|
||||
# create simple GP model
|
||||
m = GPy.models.SparseGPRegression(X,Y,kernel, M = M)
|
||||
# create simple GP Model
|
||||
m = GPy.models.SparseGPRegression(X,Y,kernel, num_inducing = num_inducing)
|
||||
|
||||
# contrain all parameters to be positive (but not inducing inputs)
|
||||
m.ensure_default_constraints()
|
||||
|
|
@ -318,7 +318,7 @@ def uncertain_inputs_sparse_regression(optim_iters=100):
|
|||
|
||||
k = GPy.kern.rbf(1) + GPy.kern.white(1)
|
||||
|
||||
# create simple GP model - no input uncertainty on this one
|
||||
# create simple GP Model - no input uncertainty on this one
|
||||
m = GPy.models.SparseGPRegression(X, Y, kernel=k, Z=Z)
|
||||
m.ensure_default_constraints()
|
||||
m.optimize('scg', messages=1, max_f_eval=optim_iters)
|
||||
|
|
@ -326,7 +326,7 @@ def uncertain_inputs_sparse_regression(optim_iters=100):
|
|||
axes[0].set_title('no input uncertainty')
|
||||
|
||||
|
||||
#the same model with uncertainty
|
||||
#the same Model with uncertainty
|
||||
m = GPy.models.SparseGPRegression(X, Y, kernel=k, Z=Z, X_variance=S)
|
||||
m.ensure_default_constraints()
|
||||
m.optimize('scg', messages=1, max_f_eval=optim_iters)
|
||||
|
|
|
|||
|
|
@ -33,7 +33,7 @@ def tuto_GP_regression():
|
|||
|
||||
m.optimize()
|
||||
|
||||
m.optimize_restarts(Nrestarts = 10)
|
||||
m.optimize_restarts(num_restarts = 10)
|
||||
|
||||
###########################
|
||||
# 2-dimensional example #
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue