mxfusion.modules.gp_modules.svgp_regression

Members

class mxfusion.modules.gp_modules.svgp_regression.SVGPRegressionLogPdf(model, posterior, observed, jitter=0.0)

Bases: mxfusion.inference.variational.VariationalInference

The inference algorithm for computing the variational lower bound of the stochastic variational Gaussian process with Gaussian likelihood.

compute(F, variables)

The abstract method for the computation of the inference algorithm

Parameters:
  • F (Python module) – the execution context (mxnet.ndarray or mxnet.symbol)
  • variables – the set of MXNet arrays that holds the values of

variables at runtime. :type variables: {str(UUID): MXNet NDArray or MXNet Symbol} :returns: the outcome of the inference algorithm :rtype: mxnet.ndarray.ndarray.NDArray or mxnet.symbol.symbol.Symbol

class mxfusion.modules.gp_modules.svgp_regression.SVGPRegressionMeanVariancePrediction(model, posterior, observed, noise_free=True, diagonal_variance=True, jitter=0.0)

Bases: mxfusion.inference.inference_alg.SamplingAlgorithm

compute(F, variables)

The method for the computation of the sampling algorithm

Parameters:
  • F (Python module) – the execution context (mxnet.ndarray or mxnet.symbol)
  • variables – the set of MXNet arrays that holds the values of

variables at runtime. :type variables: {str(UUID): MXNet NDArray or MXNet Symbol} :returns: the outcome of the inference algorithm :rtype: mxnet.ndarray.ndarray.NDArray or mxnet.symbol.symbol.Symbol

class mxfusion.modules.gp_modules.svgp_regression.SVGPRegressionSamplingPrediction(model, posterior, observed, rand_gen=None, noise_free=True, diagonal_variance=True, jitter=0.0)

Bases: mxfusion.inference.inference_alg.SamplingAlgorithm

compute(F, variables)

The method for the computation of the sampling algorithm

Parameters:
  • F (Python module) – the execution context (mxnet.ndarray or mxnet.symbol)
  • variables – the set of MXNet arrays that holds the values of

variables at runtime. :type variables: {str(UUID): MXNet NDArray or MXNet Symbol} :returns: the outcome of the inference algorithm :rtype: mxnet.ndarray.ndarray.NDArray or mxnet.symbol.symbol.Symbol

class mxfusion.modules.gp_modules.svgp_regression.SVGPRegression(X, kernel, noise_var, inducing_inputs=None, num_inducing=10, mean=None, rand_gen=None, dtype=None, ctx=None)

Bases: mxfusion.modules.module.Module

Stochastic variational sparse Gaussian process regression module

This module contains a variational sparse Gaussian process with Gaussian likelihood.

Parameters:
  • X (Variable) – the input variables on which the random variables are conditioned.
  • kernel (Kernel) – the kernel of Gaussian process.
  • noise_var (Variable) – the variance of the Gaussian likelihood
  • inducing_inputs – the inducing inputs of the sparse GP (optional). This variable will be auto-generated

if not specified. :type inducing_inputs: Variable :param num_inducing: the number of inducing points of sparse GP (default: 10) :type num_inducing: int :param mean: the mean of Gaussian process. :type mean: Variable :param rand_gen: the random generator (default: MXNetRandomGenerator). :type rand_gen: RandomGenerator :param dtype: the data type for float point numbers. :type dtype: numpy.float32 or numpy.float64 :param ctx: the mxnet context (default: None/current context). :type ctx: None or mxnet.cpu or mxnet.gpu

static define_variable(X, kernel, noise_var, shape=None, inducing_inputs=None, num_inducing=10, mean=None, rand_gen=None, dtype=None, ctx=None)

Creates and returns a variable drawn from a Stochastic variational sparse Gaussian process regression with Gaussian likelihood.

Parameters:
  • X (Variable) – the input variables on which the random variables are conditioned.
  • kernel (Kernel) – the kernel of Gaussian process.
  • noise_var (Variable) – the variance of the Gaussian likelihood
  • shape – the shape of the random variable(s) (the default shape is

the same shape as X but the last dimension is changed to one.) :type shape: tuple or [tuple] :param inducing_inputs: the inducing inputs of the sparse GP (optional). This variable will be auto-generated if not specified. :type inducing_inputs: Variable :param num_inducing: the number of inducing points of sparse GP (default: 10) :type num_inducing: int :param mean: the mean of Gaussian process. :type mean: Variable :param rand_gen: the random generator (default: MXNetRandomGenerator). :type rand_gen: RandomGenerator :param dtype: the data type for float point numbers. :type dtype: numpy.float32 or numpy.float64 :param ctx: the mxnet context (default: None/current context). :type ctx: None or mxnet.cpu or mxnet.gpu

replicate_self(attribute_map=None)

The copy constructor for the function.