Creates an sgnht (stochastic gradient Nose Hoover Thermostat) object which can be passed to
`sgmcmcStep`

to simulate from 1 step of SGNHT for the posterior defined by
logLik and logPrior. This allows the user to code the loop themselves, as in many standard
TensorFlow procedures (such as optimization). Which means they do not need to store
the chain at each iteration. This is useful when the full chain needs a lot of memory.

sgnhtSetup(logLik, dataset, params, stepsize, logPrior = NULL, minibatchSize = 0.01, a = 0.01, seed = NULL)

logLik | function which takes parameters and dataset (list of TensorFlow variables and placeholders respectively) as input. It should return a TensorFlow expression which defines the log likelihood of the model. |
---|---|

dataset | list of numeric R arrays which defines the datasets for the problem. The names in the list should correspond to those referred to in the logLik and logPrior functions |

params | list of numeric R arrays which define the starting point of each parameter. The names in the list should correspond to those referred to in the logLik and logPrior functions |

stepsize | list of numeric values corresponding to the SGLD stepsizes for each parameter The names in the list should correspond to those in params. Alternatively specify a single numeric value to use that stepsize for all parameters. |

logPrior | optional. Default uninformative improper prior. Function which takes parameters (list of TensorFlow variables) as input. The function should return a TensorFlow tensor which defines the log prior of the model. |

minibatchSize | optional. Default 0.01. Numeric or integer value that specifies amount of dataset to use at each iteration either as proportion of dataset size (if between 0 and 1) or actual magnitude (if an integer). |

a | optional. Default 0.01. List of numeric values corresponding to SGNHT diffusion factors (see Algorithm 2 of the original paper). One value should be given for each parameter in params, the names should correspond to those in params. Alternatively specify a single float to specify that value for all parameters. |

seed | optional. Default NULL. Numeric seed for random number generation. The default does not declare a seed for the TensorFlow session. |

The function returns an 'sgnht' object, which is used to pass the required information
about the current model to the `sgmcmcStep`

function. The function
`sgmcmcStep`

runs one step of sgnht. The sgnht object has the following attributes:

- params
list of tf$Variables with the same names as the params list passed to

`sgnhtSetup`

. This is the object passed to the logLik and logPrior functions you declared to calculate the log posterior gradient estimate.- estLogPost
a tensor that estimates the log posterior given the current placeholders and params.

- N
dataset size.

- data
dataset as passed to

`sgnhtSetup`

.- n
minibatchSize as passed to

`sgnhtSetup`

.- placeholders
list of tf$placeholder objects with the same names as dataset used to feed minibatches of data to

`sgmcmcStep`

. This object gets fed to the dataset argument of the logLik and logPrior functions you declared.- stepsize
list of stepsizes as passed to

`sgnhtSetup`

.- a
list of a tuning parameters as passed to

`sgnhtSetup`

.- dynamics
a list of TensorFlow steps that are evaluated by

`sgmcmcStep`

.

# NOT RUN { # Simulate from a Normal Distribution, unknown location and known scale with uninformative prior # Run sgmcmc step by step and calculate estimate of location on the fly to reduce storage dataset = list("x" = rnorm(1000)) params = list("theta" = 0) logLik = function(params, dataset) { distn = tf$distributions$Normal(params$theta, 1) return(tf$reduce_sum(distn$log_prob(dataset$x))) } stepsize = list("theta" = 1e-4) sgnht = sgnhtSetup(logLik, dataset, params, stepsize) nIters = 10^4L # Initialize location estimate locEstimate = 0 # Initialise TensorFlow session sess = initSess(sgnht) for ( i in 1:nIters ) { sgmcmcStep(sgnht, sess) locEstimate = locEstimate + 1 / nIters * getParams(sgnht, sess)$theta } # For more examples see vignettes # }