mlr_callback_set.checkpoint {mlr3torch} | R Documentation |
Saves the optimizer and network states during training. The final network and optimizer are always stored.
Saving the learner itself in the callback with a trained model is impossible, as the model slot is set after the last callback step is executed.
mlr3torch::CallbackSet
-> CallbackSetCheckpoint
new()
Creates a new instance of this R6 class.
CallbackSetCheckpoint$new(path, freq, freq_type = "epoch")
path
(character(1)
)
The path to a folder where the models are saved.
freq
(integer(1)
)
The frequency how often the model is saved.
Frequency is either per step or epoch, which can be configured through the freq_type
parameter.
freq_type
(character(1)
)
Can be be either "epoch"
(default) or "step"
.
on_epoch_end()
Saves the network and optimizer state dict.
Does nothing if freq_type
or freq
are not met.
CallbackSetCheckpoint$on_epoch_end()
on_batch_end()
Saves the selected objects defined in save
.
Does nothing if freq_type or freq are not met.
CallbackSetCheckpoint$on_batch_end()
on_exit()
Saves the learner.
CallbackSetCheckpoint$on_exit()
clone()
The objects of this class are cloneable with this method.
CallbackSetCheckpoint$clone(deep = FALSE)
deep
Whether to make a deep clone.
Other Callback:
TorchCallback
,
as_torch_callback()
,
as_torch_callbacks()
,
callback_set()
,
mlr3torch_callbacks
,
mlr_callback_set
,
mlr_callback_set.progress
,
mlr_context_torch
,
t_clbk()
,
torch_callback()