luz_callback_resume_from_checkpoint: Allow resume model training from a specific checkpoint

In mlverse/torchlight: Higher Level 'API' for 'torch'

Allow resume model training from a specific checkpoint

Description

Allow resume model training from a specific checkpoint

Usage

luz_callback_resume_from_checkpoint(
  path,
  ...,
  restore_model_state = TRUE,
  restore_records = FALSE,
  restore_optimizer_state = FALSE,
  restore_callbacks_state = FALSE
)

Arguments

path	Path to the checkpoint that you want to resume.
...	currently unused.
restore_model_state	Wether to restore the model state from the checkpoint.
restore_records	Wether to restore records from the checkpoint.
restore_optimizer_state	Wether to restore the optimizer state from the checkpoint.
restore_callbacks_state	Wether to restore the callbacks state from the checkpoint.

Note

Read the checkpointing article in the pkgdown website for more information.

See Also

luz_callback_model_checkpoint()

Other luz_callbacks: luz_callback(), luz_callback_auto_resume(), luz_callback_csv_logger(), luz_callback_early_stopping(), luz_callback_interrupt(), luz_callback_keep_best_model(), luz_callback_lr_scheduler(), luz_callback_metrics(), luz_callback_mixed_precision(), luz_callback_mixup(), luz_callback_model_checkpoint(), luz_callback_profile(), luz_callback_progress(), luz_callback_train_valid()

mlverse/torchlight documentation built on Sept. 19, 2024, 11:22 p.m.