Callbacks are objects that can customize the behavior of the training loop in the PyTorch Trainer (this feature is not yet implemented in TensorFlow) that can inspect the training loop state (for progress reporting, logging on TensorBoard or other ML platforms…) and take decisions (like early stopping).
Callbacks are “read only” pieces of code, apart from the TrainerControl object they return, they cannot change anything in the training loop. For customizations that require changes in the training loop, you should subclass Trainer and override the methods you need (see trainer for examples).
By default, TrainingArguments.report_to
is set to "all"
, so a Trainer will use the following callbacks.
If a package is installed but you don’t wish to use the accompanying integration, you can change TrainingArguments.report_to
to a list of just those integrations you want to use (e.g. ["azure_ml", "wandb"]
).
The main class that implements callbacks is TrainerCallback. It gets the TrainingArguments used to instantiate the Trainer, can access that Trainer’s internal state via TrainerState, and can take some actions on the training loop via TrainerControl.
Here is the list of the available TrainerCallback in the library:
A TrainerCallback that sends the logs to Comet ML.
Setup the optional Comet.ml integration.
Environment:
str
, optional, defaults to ONLINE
):
Whether to create an online, offline experiment or disable Comet logging. Can be OFFLINE
, ONLINE
, or
DISABLED
.str
, optional):
Comet project name for experiments.str
, optional):
Folder to use for saving offline experiments when COMET_MODE
is OFFLINE
.str
, optional, defaults to TRUE
):
Whether or not to log training assets (tf event logs, checkpoints, etc), to Comet. Can be TRUE
, or
FALSE
.For a number of configurable items in the environment, see here.
A TrainerCallback that handles the default flow of the training loop for logs, evaluation and checkpoints.
A bare TrainerCallback that just prints the logs.
A TrainerCallback that displays the progress of training or evaluation.
( early_stopping_patience: int = 1 early_stopping_threshold: Optional = 0.0 )
Parameters
int
) —
Use with metric_for_best_model
to stop training when the specified metric worsens for
early_stopping_patience
evaluation calls. float
, optional) —
Use with TrainingArguments metric_for_best_model
and early_stopping_patience
to denote how much the
specified metric must improve to satisfy early stopping conditions. ` A TrainerCallback that handles early stopping.
This callback depends on TrainingArguments argument load_best_model_at_end functionality to set best_metric in TrainerState. Note that if the TrainingArguments argument save_steps differs from eval_steps, the early stopping will not occur until the next save step.
( tb_writer = None )
A TrainerCallback that sends the logs to TensorBoard.
A TrainerCallback that logs metrics, media, model checkpoints to Weight and Biases.
Setup the optional Weights & Biases (wandb) integration.
One can subclass and override this method to customize the setup if needed. Find more information here. You can also override the following environment variables:
Environment:
WANDB_LOG_MODEL (str
, optional, defaults to "false"
):
Whether to log model and checkpoints during training. Can be "end"
, "checkpoint"
or "false"
. If set
to "end"
, the model will be uploaded at the end of training. If set to "checkpoint"
, the checkpoint
will be uploaded every args.save_steps
. If set to "false"
, the model will not be uploaded. Use along
with load_best_model_at_end()
to upload best model.
Deprecated in 5.0
Setting WANDB_LOG_MODEL
as bool
will be deprecated in version 5 of 🤗 Transformers.
WANDB_WATCH (str
, optional defaults to "false"
):
Can be "gradients"
, "all"
, "parameters"
, or "false"
. Set to "all"
to log gradients and
parameters.
WANDB_PROJECT (str
, optional, defaults to "huggingface"
):
Set this to a custom string to store results in a different project.
WANDB_DISABLED (bool
, optional, defaults to False
):
Whether to disable wandb entirely. Set WANDB_DISABLED=true
to disable.
A TrainerCallback that sends the logs to MLflow. Can be disabled by setting
environment variable DISABLE_MLFLOW_INTEGRATION = TRUE
.
Setup the optional MLflow integration.
Environment:
str
, optional):
Whether to use MLflow .log_artifact()
facility to log artifacts. This only makes sense if logging to a
remote server, e.g. s3 or GCS. If set to True
or 1, will copy each saved checkpoint on each save in
TrainingArguments’s output_dir
to the local or remote artifact storage. Using it without a remote
storage will just copy the files to your artifact location.str
, optional, defaults to ""
):
Whether to store runs at a specific path or remote server. Default to an empty string which will store runs
at ./mlruns
locally.str
, optional, defaults to None
):
Whether to use an MLflow experiment_name under which to launch the run. Default to None
which will point
to the Default
experiment in MLflow. Otherwise, it is a case sensitive name of the experiment to be
activated. If an experiment with this name does not exist, a new experiment with this name is created.str
, optional):
A string dump of a dictionary of key/value pair to be added to the MLflow run as tags. Example:
os.environ['MLFLOW_TAGS']='{"release.candidate": "RC1", "release.version": "2.2.0"}'
.str
, optional):
Whether to use MLflow nested runs. If set to True
or 1, will create a nested run inside the current
run.str
, optional):
Allow to reattach to an existing run which can be usefull when resuming training from a checkpoint. When
MLFLOW_RUN_ID
environment variable is set, start_run
attempts to resume a run with the specified run ID
and other parameters are ignored.str
, optional, defaults to False
):
Whether to flatten the parameters dictionary before logging.A TrainerCallback that sends the logs to AzureML.
A TrainerCallback that tracks the CO2 emission of training.
( api_token: Optional = None project: Optional = None name: Optional = None base_namespace: str = 'finetuning' run = None log_parameters: bool = True log_checkpoints: Optional = None **neptune_run_kwargs )
Parameters
str
, optional) — Neptune API token obtained upon registration.
You can leave this argument out if you have saved your token to the NEPTUNE_API_TOKEN
environment
variable (strongly recommended). See full setup instructions in the
docs. str
, optional) — Name of an existing Neptune project, in the form “workspace-name/project-name”.
You can find and copy the name in Neptune from the project settings -> Properties. If None (default), the
value of the NEPTUNE_PROJECT
environment variable is used. str
, optional) — Custom name for the run. str
, optional, defaults to “finetuning”) — In the Neptune run, the root namespace
that will contain all of the metadata logged by the callback. bool
, optional, defaults to True
) —
If True, logs all Trainer arguments and model parameters provided by the Trainer. str
, optional) — If “same”, uploads checkpoints whenever they are saved by the Trainer.
If “last”, uploads only the most recently saved checkpoint. If “best”, uploads the best checkpoint (among
the ones saved by the Trainer). If None
, does not upload checkpoints. Run
, optional) — Pass a Neptune run object if you want to continue logging to an existing run.
Read more about resuming runs in the docs. neptune.init_run()
function when a new run is created. TrainerCallback that sends the logs to Neptune.
For instructions and examples, see the Transformers integration guide in the Neptune documentation.
A TrainerCallback that sends the logs to ClearML.
Environment:
str
, optional, defaults to HuggingFace Transformers
):
ClearML project name.str
, optional, defaults to Trainer
):
ClearML task name.bool
, optional, defaults to False
):
Whether to log models as artifacts during training.A TrainerCallback that logs to DagsHub. Extends MLflowCallback
Setup the DagsHub’s Logging integration.
Environment:
str
, optional):
Whether to save the data and model artifacts for the experiment. Default to False
.( save_log_history: bool = True sync_checkpoints: bool = True )
Parameters
A TrainerCallback that sends the logs to Flyte. NOTE: This callback only works within a Flyte task.
( live: Optional = None log_model: Union = None **kwargs )
Parameters
dvclive.Live
, optional, defaults to None
) —
Optional Live instance. If None, a new instance will be created using **kwargs. None
) —
Whether to use dvclive.Live.log_artifact()
to log checkpoints created by Trainer. If set to True
,
the final checkpoint is logged at the end of training. If set to "all"
, the entire
TrainingArguments’s output_dir
is logged at each checkpoint. A TrainerCallback that sends the logs to DVCLive.
Use the environment variables below in setup
to configure the integration. To customize this callback beyond
those environment variables, see here.
Setup the optional DVCLive integration. To customize this callback beyond the environment variables below, see here.
Environment:
str
, optional):
Whether to use dvclive.Live.log_artifact()
to log checkpoints created by Trainer. If set to True
or
1, the final checkpoint is logged at the end of training. If set to all
, the entire
TrainingArguments’s output_dir
is logged at each checkpoint.( )
Parameters
torch.nn.Module
) —
The model being trained. torch.optim.Optimizer
) —
The optimizer used for the training steps. torch.optim.lr_scheduler.LambdaLR
) —
The scheduler used for setting the learning rate. torch.utils.data.DataLoader
, optional) —
The current dataloader used for training. torch.utils.data.DataLoader
, optional) —
The current dataloader used for training. Dict[str, float]
) —
The metrics computed by the last evaluation phase.
Those are only accessible in the event on_evaluate
.
Dict[str, float]
) —
The values to log.
Those are only accessible in the event on_log
.
A class for objects that will inspect the state of the training loop at some events and take some decisions. At each of those events the following arguments are available:
The control
object is the only one that can be changed by the callback, in which case the event that changes it
should return the modified version.
The argument args
, state
and control
are positionals for all events, all the others are grouped in kwargs
.
You can unpack the ones you need in the signature of the event using them. As an example, see the code of the
simple PrinterCallback.
Example:
class PrinterCallback(TrainerCallback):
def on_log(self, args, state, control, logs=None, **kwargs):
_ = logs.pop("total_flos", None)
if state.is_local_process_zero:
print(logs)
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called at the beginning of an epoch.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called at the end of an epoch.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called after an evaluation phase.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called at the end of the initialization of the Trainer.
Event called after logging the last logs.
( args: TrainingArguments state: TrainerState control: TrainerControl metrics **kwargs )
Event called after a successful prediction.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called after a prediction step.
Event called after a checkpoint save.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called at the beginning of a training step. If using gradient accumulation, one training step might take several inputs.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called at the end of a training step. If using gradient accumulation, one training step might take several inputs.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called at the end of an substep during gradient accumulation.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called at the beginning of training.
( args: TrainingArguments state: TrainerState control: TrainerControl **kwargs )
Event called at the end of training.
Here is an example of how to register a custom callback with the PyTorch Trainer:
class MyCallback(TrainerCallback):
"A callback that prints a message at the beginning of training"
def on_train_begin(self, args, state, control, **kwargs):
print("Starting training")
trainer = Trainer(
model,
args,
train_dataset=train_dataset,
eval_dataset=eval_dataset,
callbacks=[MyCallback], # We can either pass the callback class this way or an instance of it (MyCallback())
)
Another way to register a callback is to call trainer.add_callback()
as follows:
trainer = Trainer(...)
trainer.add_callback(MyCallback)
# Alternatively, we can pass an instance of the callback class
trainer.add_callback(MyCallback())
( epoch: Optional = None global_step: int = 0 max_steps: int = 0 logging_steps: int = 500 eval_steps: int = 500 save_steps: int = 500 train_batch_size: int = None num_train_epochs: int = 0 num_input_tokens_seen: int = 0 total_flos: float = 0 log_history: List = None best_metric: Optional = None best_model_checkpoint: Optional = None is_local_process_zero: bool = True is_world_process_zero: bool = True is_hyper_param_search: bool = False trial_name: str = None trial_params: Dict = None )
Parameters
float
, optional) —
Only set during training, will represent the epoch the training is at (the decimal part being the
percentage of the current epoch completed). int
, optional, defaults to 0) —
During training, represents the number of update steps completed. int
, optional, defaults to 0) —
The number of update steps to do during the current training. int
, optional, defaults to 500) —
Log every X updates steps int
, optional) —
Run an evaluation every X steps. int
, optional, defaults to 500) —
Save checkpoint every X updates steps. int
, optional) —
The batch size for the training dataloader. Only needed when
auto_find_batch_size
has been used. int
, optional, defaults to 0) —
The number of tokens seen during training (number of input tokens, not the number of prediction tokens). float
, optional, defaults to 0) —
The total number of floating operations done by the model since the beginning of training (stored as floats
to avoid overflow). List[Dict[str, float]]
, optional) —
The list of logs done since the beginning of training. float
, optional) —
When tracking the best model, the value of the best metric encountered so far. str
, optional) —
When tracking the best model, the value of the name of the checkpoint for the best model encountered so
far. bool
, optional, defaults to True
) —
Whether or not this process is the local (e.g., on one machine if training in a distributed fashion on
several machines) main process. bool
, optional, defaults to True
) —
Whether or not this process is the global main process (when training in a distributed fashion on several
machines, this is only going to be True
for one process). bool
, optional, defaults to False
) —
Whether we are in the process of a hyper parameter search using Trainer.hyperparameter_search. This will
impact the way data will be logged in TensorBoard. A class containing the Trainer inner state that will be saved along the model and optimizer when checkpointing and passed to the TrainerCallback.
In all this class, one step is to be understood as one update step. When using gradient accumulation, one update
step may require several forward and backward passes: if you use gradient_accumulation_steps=n
, then one update
step requires going through n batches.
Create an instance from the content of json_path
.
Save the content of this instance in JSON format inside json_path
.
( should_training_stop: bool = False should_epoch_stop: bool = False should_save: bool = False should_evaluate: bool = False should_log: bool = False )
Parameters
bool
, optional, defaults to False
) —
Whether or not the training should be interrupted.
If True
, this variable will not be set back to False
. The training will just stop.
bool
, optional, defaults to False
) —
Whether or not the current epoch should be interrupted.
If True
, this variable will be set back to False
at the beginning of the next epoch.
bool
, optional, defaults to False
) —
Whether or not the model should be saved at this step.
If True
, this variable will be set back to False
at the beginning of the next step.
bool
, optional, defaults to False
) —
Whether or not the model should be evaluated at this step.
If True
, this variable will be set back to False
at the beginning of the next step.
bool
, optional, defaults to False
) —
Whether or not the logs should be reported at this step.
If True
, this variable will be set back to False
at the beginning of the next step.
A class that handles the Trainer control flow. This class is used by the TrainerCallback to activate some switches in the training loop.