# GPy.inference.optimization package¶

## GPy.inference.optimization.stochastics module¶

class SparseGPMissing(model, batchsize=1)[source]
class SparseGPStochastics(model, batchsize=1, missing_data=True)[source]

For the sparse gp we need to store the dimension we are in, and the indices corresponding to those

do_stochastics()[source]
reset()[source]
class StochasticStorage(model)[source]

Bases: object

This is a container for holding the stochastic parameters, such as subset indices or step length and so on.

self.d has to be a list of lists: [dimension indices, nan indices for those dimensions] so that the minibatches can be used as efficiently as possible.

do_stochastics()[source]

Update the internal state to the next batch of the stochastic descent algorithm.

reset()[source]

Reset the state of this stochastics generator.