|  View source on GitHub | 
A stateful process for finalization of a round of training.
Inherits From: MeasuredProcess, IterativeProcess
tff.learning.templates.FinalizerProcess(
    initialize_fn: tff.Computation,
    next_fn: tff.Computation,
    *,
    get_hparams_fn: Optional[tff.Computation] = None,
    set_hparams_fn: Optional[tff.Computation] = None
)
A FinalizerProcess is a tff.templates.MeasuredProcess that formalizes the
type signature of initialize_fn and next_fn for the work performed by
server in a learning process after aggregating model updates from clients.
| Args | |
|---|---|
| initialize_fn | A tff.Computationmatching the criteria above. | 
| next_fn | A tff.Computationmatching the criteria above. | 
| get_hparams_fn | An optional tff.Computationmatching the criteria above.
If not provided, this defaults to a computation that returns an empty
ordred dictionary, regardless of the contents of the state. | 
| set_hparams_fn | An optional tff.Computationmatching the criteria above.
If not provided, this defaults to a pass-through computation, that
returns the input state regardless of the hparams passed in. | 
| Attributes | |
|---|---|
| get_hparams | |
| initialize | A no-arg tff.Computationthat returns the initial state. | 
| next | A tff.Computationthat produces the next state.Its first argument should always be the current state (originally produced
by  | 
| set_hparams | |
| state_type | The tff.Typeof the state of the process. |