def test_bayesian_gplvm_2d():
Q = 2 # latent dimensions
x_data_mean = pca_reduce(Data.Y, Q)
kernel = gpflow.kernels.SquaredExponential()
m = gpflow.models.BayesianGPLVM(Data.Y,
x_data_mean,
np.ones((Data.N, Q)),
kernel,
num_inducing_variables=Data.M)
log_likelihood_initial = m.log_likelihood()
opt = gpflow.optimizers.Scipy()
@tf.function(autograph=False)
def objective_closure():
return -m.log_marginal_likelihood()
opt.minimize(objective_closure,
m.trainable_variables,
options=dict(maxiter=2))
assert m.log_likelihood() > log_likelihood_initial
# test prediction
Xtest = Data.rng.randn(10, Q)
mu_f, var_f = m.predict_f(Xtest)
mu_fFull, var_fFull = m.predict_f(Xtest, full_cov=True)
np.testing.assert_allclose(mu_fFull, mu_f)
for i in range(Data.D):
np.testing.assert_allclose(var_f[:, i], np.diag(var_fFull[:, :, i]))
def test_bayesian_gplvm_2d():
Q = 2 # latent dimensions
X_data_mean = pca_reduce(Data.Y, Q)
kernel = gpflow.kernels.SquaredExponential()
m = gpflow.models.BayesianGPLVM(Data.Y,
X_data_mean,
np.ones((Data.N, Q)),
kernel,
num_inducing_variables=Data.M)
elbo_initial = m.elbo()
opt = gpflow.optimizers.Scipy()
opt.minimize(m.training_loss,
m.trainable_variables,
options=dict(maxiter=2))
assert m.elbo() > elbo_initial
# test prediction
Xtest = Data.rng.randn(10, Q)
mu_f, var_f = m.predict_f(Xtest)
mu_fFull, var_fFull = m.predict_f(Xtest, full_cov=True)
np.testing.assert_allclose(mu_fFull, mu_f)
for i in range(Data.D):
np.testing.assert_allclose(var_f[:, i], np.diag(var_fFull[:, :, i]))
num_training_points=NUM_TRAINING_POINTS, observation_noise_variance=0.01)
# plt.scatter(Y_bg[:, 0], Y_bg[:, 1])
# plt.scatter(Y_fg[:, 0], Y_fg[:, 1])
# plt.show()
Y = tf.convert_to_tensor(Y_fg, dtype=default_float())
print("Number of points: {} and Number of dimensions: {}".format(
Y.shape[0], Y.shape[1]))
latent_dim = 2 # number of latent dimensions
num_inducing = 20 # number of inducing pts
num_data = Y.shape[0] # number of data points
X_mean_init = ops.pca_reduce(Y, latent_dim)
# X_mean_init = tf.ones((num_data, latent_dim), dtype=default_float())
X_var_init = tf.ones((num_data, latent_dim), dtype=default_float())
np.random.seed(1) # for reproducibility
inducing_variable = tf.convert_to_tensor(np.random.permutation(
X_mean_init.numpy())[:num_inducing],
dtype=default_float())
# lengthscales = tf.convert_to_tensor([1.0] * latent_dim, dtype=default_float())
# kernel = gpflow.kernels.RBF(lengthscales=lengthscales)
kernel = gpflow.kernels.Cosine()
gplvm = gpflow.models.BayesianGPLVM(
Y,
X_data_mean=X_mean_init,
# %% [markdown]
# ## Model construction
#
# We start by initializing the required variables:
# %%
latent_dim = 2 # number of latent dimensions
num_inducing = 20 # number of inducing pts
num_data = Y.shape[0] # number of data points
# %% [markdown]
# Initialize via PCA:
# %%
x_mean_init = tf.convert_to_tensor(ops.pca_reduce(Y, latent_dim),
dtype=default_float())
x_var_init = tf.convert_to_tensor(np.ones((num_data, latent_dim)),
dtype=default_float())
# %% [markdown]
# Pick inducing inputs randomly from dataset initialization:
# %%
inducing_variable = tf.convert_to_tensor(np.random.permutation(
x_mean_init.numpy())[:num_inducing],
dtype=default_float())
# %% [markdown]
# We construct a Squared Exponential (SE) kernel operating on the two-dimensional latent space.
# The `ARD` parameter stands for Automatic Relevance Determination, which in practice means that
Y = np.array(mocap[1])
print("shape Y", shape(Y))
# Y = np.reshape(Y, (np.shape(Y)[0], -1))
Y = Y[:, 0, :]
Y = tf.convert_to_tensor(Y, dtype=default_float())
print("shape Y", np.shape(Y))
print('Number of points: {} and Number of dimensions: {}'.format(
Y.shape[0], Y.shape[1]))
latent0_dim = 2 # number of latent dimensions
latent1_dim = 8 # number of latent dimensions from hidden layer
# num_inducing = 20 # number of inducing pts
num_data = Y.shape[0] # number of data points
x_mean_latent = tf.convert_to_tensor(ops.pca_reduce(Y, latent1_dim),
dtype=default_float())
x_mean_init = tf.convert_to_tensor(ops.pca_reduce(x_mean_latent, latent0_dim),
dtype=default_float())
lengthscale0 = tf.convert_to_tensor([1.0] * latent0_dim, dtype=default_float())
lengthscale1 = tf.convert_to_tensor([1.0] * latent1_dim, dtype=default_float())
kernel0 = gpflow.kernels.RBF(lengthscale=lengthscale0)
kernel1 = gpflow.kernels.RBF(lengthscale=lengthscale1)
print("shape x_mean_latent", np.shape(x_mean_latent))
print("shape x_mean_init", np.shape(x_mean_init))
model = myGPLVM(Y,
latent_data=x_mean_latent,
x_data_mean=x_mean_init,
kernel=[kernel0, kernel1])
def __init__(
self,
data: OutputData,
kernel: Optional[Kernel] = None,
latent_dimensions: Optional[int] = 2,
num_inducing_variables: Optional[int] = None,
inducing_variable=None,
*,
mean_function=None,
q_diag: bool = False,
q_mu=None,
q_sqrt=None,
whiten: bool = False,
):
"""
- kernel, likelihood, inducing_variables, mean_function are appropriate
GPflow objects
- num_latent_gps is the number of latent processes to use, defaults to 2, as
the dimensionality reduction is at dimensions 2
- q_diag is a boolean. If True, the covariance is approximated by a
diagonal matrix.
- whiten is a boolean. If True, we use the whitened representation of
the inducing points.
- num_data is the total number of observations, defaults to X.shape[0]
(relevant when feeding in external minibatches)
"""
self.latent_dimensions = latent_dimensions
#grab data
self.data = data_input_to_tensor(data)
#define lat-space initialization
X_data_mean = pca_reduce(data, self.latent_dimensions)
num_data, num_latent_gps = data.shape
self.num_data = num_data
X_data_var = tf.ones((self.num_data, self.latent_dimensions),
dtype=default_float())
assert X_data_var.ndim == 2
#def kernel
if kernel is None:
kernel = gpflow.kernels.SquaredExponential()
#init Parameters latent
self.X_data_mean = Parameter(X_data_mean)
self.X_data_var = Parameter(X_data_var, transform=positive())
#init parameter inducing point
if (inducing_variable is None) == (num_inducing_variables is None):
raise ValueError(
"BayesianGPLVM needs exactly one of `inducing_variable` and `num_inducing_variables`"
)
if inducing_variable is None:
# By default we initialize by subset of initial latent points
# Note that tf.random.shuffle returns a copy, it does not shuffle in-place
#maybe use k-means clustering
Z = tf.random.shuffle(X_data_mean)[:num_inducing_variables]
inducing_variable = InducingPoints(Z)
self.inducing_variable = inducingpoint_wrapper(inducing_variable)
#loss placeholder for analysis purpuse
self.loss_placeholder = defaultdict(
list, {k: []
for k in ("KL_x", "ELBO", "KL_u")})
# deal with parameters for the prior mean variance of X
X_prior_mean = tf.zeros((self.num_data, self.latent_dimensions),
dtype=default_float())
X_prior_var = tf.ones((self.num_data, self.latent_dimensions),
dtype=default_float())
self.X_prior_mean = tf.convert_to_tensor(np.atleast_1d(X_prior_mean),
dtype=default_float())
self.X_prior_var = tf.convert_to_tensor(np.atleast_1d(X_prior_var),
dtype=default_float())
#sanity check
assert np.all(X_data_mean.shape == X_data_var.shape)
assert X_data_mean.shape[0] == self.data.shape[
0], "X mean and Y must be same size."
assert X_data_var.shape[0] == self.data.shape[
0], "X var and Y must be same size."
assert X_data_mean.shape[1] == self.latent_dimensions
assert self.X_prior_mean.shape[0] == self.num_data
assert self.X_prior_mean.shape[1] == self.latent_dimensions
assert self.X_prior_var.shape[0] == self.num_data
assert self.X_prior_var.shape[1] == self.latent_dimensions
# init the super class, accept args
super().__init__(kernel, likelihoods.Gaussian(variance=0.1),
mean_function, num_latent_gps)
self.q_diag = q_diag
self.whiten = whiten
#self.inducing_variable = inducingpoint_wrapper(inducing_variable)
# init variational parameters
num_inducing = self.inducing_variable.num_inducing
self._init_variational_parameters(num_inducing, q_mu, q_sqrt, q_diag)
def __init__(
self,
data: OutputData,
X_data_mean: Optional[tf.Tensor] = None,
X_data_var: Optional[tf.Tensor] = None,
kernel: Optional[Kernel] = None,
num_inducing_variables: Optional[int] = None,
inducing_variable=None,
X_prior_mean=None,
X_prior_var=None,
):
"""
Initialise Bayesian GPLVM object. This method only works with a Gaussian likelihood.
:param data: data matrix, size N (number of points) x D (dimensions)
:param X_data_mean: initial latent positions, size N (number of points) x Q (latent dimensions).
:param X_data_var: variance of latent positions ([N, Q]), for the initialisation of the latent space.
:param kernel: kernel specification, by default Squared Exponential
:param num_inducing_variables: number of inducing points, M
:param inducing_variable: matrix of inducing points, size M (inducing points) x Q (latent dimensions). By default
random permutation of X_data_mean.
:param X_prior_mean: prior mean used in KL term of bound. By default 0. Same size as X_data_mean.
:param X_prior_var: prior variance used in KL term of bound. By default 1.
"""
self.latent_dimensions = 2
#grab data
self.data = data_input_to_tensor(data)
#define lat-space initialization
if X_data_mean is None:
X_data_mean = pca_reduce(data, self.latent_dimensions)
num_data, num_latent_gps = X_data_mean.shape
self.num_data = num_data
if X_data_var is None:
X_data_var = tf.ones((self.num_data, self.latent_dimensions),
dtype=default_float())
assert X_data_var.ndim == 2
self.output_dim = self.data.shape[-1] #num_latent maybe
#def kernel
if kernel is None:
kernel = gpflow.kernels.SquaredExponential()
#init GPMODEL
super().__init__(kernel,
likelihoods.Gaussian(variance=0.1),
num_latent_gps=num_latent_gps)
#init Parameters latent
self.X_data_mean = Parameter(X_data_mean)
self.X_data_var = Parameter(X_data_var, transform=positive())
#init parameter inducing point
if (inducing_variable is None) == (num_inducing_variables is None):
raise ValueError(
"BayesianGPLVM needs exactly one of `inducing_variable` and `num_inducing_variables`"
)
if inducing_variable is None:
# By default we initialize by subset of initial latent points
# Note that tf.random.shuffle returns a copy, it does not shuffle in-place
#maybe use k-means clustering
Z = tf.random.shuffle(X_data_mean)[:num_inducing_variables]
inducing_variable = InducingPoints(Z)
self.inducing_variable = inducingpoint_wrapper(inducing_variable)
#loss placeholder for analysis purpuse
self.loss_placeholder = defaultdict(list,
{k: []
for k in ("KL_x", "ELBO")})
# deal with parameters for the prior mean variance of X
if X_prior_mean is None:
X_prior_mean = tf.zeros((self.num_data, self.latent_dimensions),
dtype=default_float())
if X_prior_var is None:
X_prior_var = tf.ones((self.num_data, self.latent_dimensions),
dtype=default_float())
self.X_prior_mean = tf.convert_to_tensor(np.atleast_1d(X_prior_mean),
dtype=default_float())
self.X_prior_var = tf.convert_to_tensor(np.atleast_1d(X_prior_var),
dtype=default_float())
#sanity check
assert np.all(X_data_mean.shape == X_data_var.shape)
assert X_data_mean.shape[0] == self.data.shape[
0], "X mean and Y must be same size."
assert X_data_var.shape[0] == self.data.shape[
0], "X var and Y must be same size."
assert X_data_mean.shape[1] == self.latent_dimensions
assert self.X_prior_mean.shape[0] == self.num_data
assert self.X_prior_mean.shape[1] == self.latent_dimensions
assert self.X_prior_var.shape[0] == self.num_data
assert self.X_prior_var.shape[1] == self.latent_dimensions
