tff.templates.MeasuredProcess

A stateful process that produces metrics.

Inherits From: IterativeProcess

tff.templates.MeasuredProcess(
    initialize_fn: tff.Computation,
    next_fn: tff.Computation,
    next_is_multi_arg: Optional[bool] = None
)

This class inherits the constraints documented by tff.templates.IterativeProcess.

A tff.templates.MeasuredProcess is a tff.templates.IterativeProcess whose next computation returns a tff.templates.MeasuredProcessOutput.

Unlike tff.templates.IterativeProcess, the more generic but less-defined template, arbitrary tff.templates.MeasuredProcesses can be composed together as follows:

Guidance for Composition Two MeasuredProcesses F(x) and G(y) can be composed into a new MeasuredProcess called C with the following properties:

  • C.state is the concatenation <F.state, G.state>.
  • C.next(C.state, x).result == G.next(G.state, F.next(F.state, x).result).result
  • C.measurements is the concatenation <F.measurements, G.measurements>.

The resulting composition C would have the following type signatures: initialize: ( -> <F.initialize, G.initialize>) next: (<<F.state, G.state>, F.input> -> <state=<F.state, G.State>, result=G.result, measurements=<F.measurements, G.measurements>)

Note that the guidance for composition is not strict and details are allowed to differ.

initialize_fn A no-arg tff.Computation that returns the initial state of the measured process. Let the type of this state be called S.
next_fn A tff.Computation that represents the iterated function. The first or only argument must match the state type S. The return value must be a MeasuredProcessOutput whose state member matches the state type S.
next_is_multi_arg An optional boolean indicating that next_fn will receive more than just the state argument (if True) or only the state argument (if False). This parameter is primarily used to provide better error messages.

TypeError If initialize_fn and next_fn are not instances of tff.Computation.
TemplateInitFnParamNotEmptyError If initialize_fn has any input arguments.
TemplateStateNotAssignableError If the state returned by either initialize_fn or next_fn is not assignable to the first input argument of next_fn.
TemplateNotMeasuredProcessOutputError If next_fn does not return a MeasuredProcessOutput.

initialize A no-arg tff.Computation that returns the initial state.
next A tff.Computation that runs one iteration of the process.

Its first argument should always be the current state (originally produced by tff.templates.MeasuredProcess.initialize), and the return type must be a tff.templates.MeasuredProcessOutput.

state_type The tff.Type of the state of the process.