Callbacks allow injecting functionality at many points during the training loop.
To use them, simply pass each callback to
Learner:
learner
=
Learner
(
model
,
lossfn
,
# required arguments
callbacks
=
[
ToGPU
(
)
,
Metrics
(
accuracy
)
]
,
data
=
data
)
# pass any number of callbacks as additional arguments
Learner()
Some useful callbacks are added by default. Below, the callbacks of
learner are shown. Both the explicitly passed callbacks and the default ones are included:
learner
.
callbacks
.
cbs
FluxTraining.SafeCallback[ToDevice(Flux.gpu, Flux.gpu), Metrics(Loss(), Metric(Accuracy)), ProgressPrinter(), MetricsPrinter(), StopOnNaNLoss(), Recorder()]See callback reference for a list of all callbacks included in FluxTraining.jl and their documentation.
The order the callbacks are passed in doesn't matter. FluxTraining.jl creates a dependency graph that makes sure the callbacks are run in the correct order. Read custom callbacks to find out how to create callbacks yourself.
The following page links back here: