composer.callbacks.callback_hparams#

Hyperparameters for callbacks.

Hparams

These classes are used with yahp for YAML-based configuration.

`CallbackHparams`	Base class for Callback hyperparameters.
`CheckpointSaverHparams`	`CheckpointSaver` hyperparameters.
`GradMonitorHparams`	`GradMonitor` hyperparamters.
`LRMonitorHparams`	`LRMonitor` hyperparameters.
`MemoryMonitorHparams`	`MemoryMonitor` hyperparameters.
`SpeedMonitorHparams`	`SpeedMonitor` hyperparameters.

class composer.callbacks.callback_hparams.CallbackHparams[source]#

Bases: yahp.hparams.Hparams, abc.ABC

Base class for Callback hyperparameters.

abstract initialize_object()[source]#

Initialize the callback.

Returns: Callback – An instance of the callback.

class composer.callbacks.callback_hparams.CheckpointSaverHparams(save_folder='{run_name}/checkpoints', filename='ep{epoch}-ba{batch}-rank{rank}', artifact_name='{run_name}/checkpoints/ep{epoch}-ba{batch}-rank{rank}', latest_filename='latest-rank{rank}', overwrite=False, weights_only=False, save_interval='1ep', num_checkpoints_to_keep=- 1)[source]#

Bases: composer.callbacks.callback_hparams.CallbackHparams

CheckpointSaver hyperparameters.

Parameters

save_folder (str, optional) – See CheckpointSaver.
filename (str, optional) – See CheckpointSaver.
artifact_name (str, optional) – See CheckpointSaver.
latest_filename (str, optional) – See CheckpointSaver.
overwrite (bool, optional) – See CheckpointSaver. Default: False.
weights_only (bool, optional) – See CheckpointSaver. Deafult: False.
save_interval (str, optional) –
Either a time-string or a path to a function. If a time-string, checkpoints will be saved according to this interval.

If a path to a function, it should be of the format 'path.to.function:function_name'. The function should take (State, Event) and return a boolean indicating whether a checkpoint should be saved given the current state and event. The event will be either Event.BATCH_CHECKPOINT or Event.EPOCH_CHECKPOINT. Default: "1ep".
num_checkpoints_to_keep (int, optional) – See CheckpointSaver. Default: -1.

class composer.callbacks.callback_hparams.EarlyStopperHparams(monitor, dataloader_label, comp=None, min_delta=0.0, patience=1)[source]#

Bases: composer.callbacks.callback_hparams.CallbackHparams

EarlyStopper hyperparameters.

Parameters

monitor (str) – The name of the metric to monitor.
dataloader_label (str) – The label of the dataloader or evaluator associated with the tracked metric. If monitor is in an Evaluator, the dataloader_label field should be set to the Evaluator’s label. If monitor is a training metric or an ordinary evaluation metric not in an Evaluator, dataloader_label should be set to ‘train’ or ‘eval’ respectively.
comp (str, optional) – A string dictating which comparison operator to use to measure change in the monitored metric. Set comp to “less” to use the function torch.less(), and “greater” to use the function torch.greater(). The comparison operator will be called comp(current_value, prev_best). For example, for metrics where the optimal value is low (error, loss, perplexity), use a less than operator.
min_delta (float, optional) – An optional float that requires a new value to exceed the best value by at least that amount. Defaults to 0.
patience (int | str, optional) – The interval of time the monitored metric can not improve without stopping training. Defaults to 1 epoch. If patience is an integer, it is interpreted as the number of epochs.

initialize_object()[source]#

Initialize the EarlyStopper callback.

Returns: EarlyStopper – An instance of EarlyStopper.

class composer.callbacks.callback_hparams.GradMonitorHparams(log_layer_grad_norms=False)[source]#

Bases: composer.callbacks.callback_hparams.CallbackHparams

GradMonitor hyperparamters.

Parameters: log_layer_grad_norms (bool, optional) – See GradMonitor for documentation. Default: False.

initialize_object()[source]#

Initialize the GradMonitor callback.

Returns: GradMonitor – An instance of GradMonitor.

class composer.callbacks.callback_hparams.LRMonitorHparams[source]#

Bases: composer.callbacks.callback_hparams.CallbackHparams

LRMonitor hyperparameters.

There are no parameters as LRMonitor does not take any parameters.

initialize_object()[source]#

Initialize the LRMonitor callback.

Returns: LRMonitor – An instance of LRMonitor.

class composer.callbacks.callback_hparams.MLPerfCallbackHparams(root_folder, index, benchmark='resnet', target=0.759, division='open', metric_name='Accuracy', metric_label='eval', submitter='MosaicML', system_name=None, status='onprem', cache_clear_cmd=None, host_processors_per_node=None)[source]#

Bases: composer.callbacks.callback_hparams.CallbackHparams

MLPerfCallback hyperparameters.

Parameters

root_folder (str) – The root submission folder
index (int) – The repetition index of this run. The filename created will be result_[index].txt.
benchmark (str, optional) – Benchmark name. Currently only resnet supported. Default: resnet.
target (float, optional) – The target metric before the mllogger marks the stop of the timing run. Default: 0.759 (resnet benchmark).
division (str, optional) – Division of submission. Currently only open division supported. Default: "open".
metric_name (str, optional) – name of the metric to compare against the target. Default: "Accuracy".
metric_label (str, optional) – label name. The metric will be accessed via state.current_metrics[metric_label][metric_name]. Default: "eval".
submitter (str, optional) – Submitting organization. Default: "MosaicML".
system_name (str, optional) – Name of the system (e.g. 8xA100_composer). If None, system name will default to [world_size]x[device_name]_composer, e.g. 8xNVIDIA_A100_80GB_composer. Default: None.
status (str, optional) – Submission status. One of (onprem, cloud, or preview). Default: "onprem".
cache_clear_cmd (str, optional) – Command to invoke during the cache clear. This callback will call subprocess(cache_clear_cmd). Default is disabled (None)
host_processors_per_node (int, optional) – Total number of host processors per node. Default: None.

initialize_object()[source]#

Initialize the MLPerf Callback.

Returns: MLPerfCallback – An instance of MLPerfCallback

class composer.callbacks.callback_hparams.MemoryMonitorHparams[source]#

Bases: composer.callbacks.callback_hparams.CallbackHparams

MemoryMonitor hyperparameters.

There are no parameters as MemoryMonitor does not take any parameters.

initialize_object()[source]#

Initialize the MemoryMonitor callback.

Returns: MemoryMonitor – An instance of MemoryMonitor.

class composer.callbacks.callback_hparams.SpeedMonitorHparams(window_size=100)[source]#

Bases: composer.callbacks.callback_hparams.CallbackHparams

SpeedMonitor hyperparameters.

Parameters: window_size (int, optional) – See SpeedMonitor for documentation.

initialize_object()[source]#

Initialize the SpeedMonitor callback.

Returns: SpeedMonitor – An instance of SpeedMonitor.

class composer.callbacks.callback_hparams.ThresholdStopperHparams(monitor, dataloader_label, threshold, comp=None, stop_on_batch=False)[source]#

Bases: composer.callbacks.callback_hparams.CallbackHparams

ThresholdStopper hyperparameters.

Parameters

monitor (str) – The name of the metric to monitor.
dataloader_label (str) – The label of the dataloader or evaluator associated with the tracked metric. If monitor is in an Evaluator, the dataloader_label field should be set to the Evaluator’s label. If monitor is a training metric or an ordinary evaluation metric not in an Evaluator, dataloader_label should be set to ‘train’ or ‘eval’ respectively.
threshold (float) – The threshold that dictates when to halt training. Whether training stops if the metric exceeds or falls below the threshold depends on the comparison operator.
comp (str, optional) – A string dictating which comparison operator to use to measure change in the monitored metric. Set comp to “less” to use the function torch.less(), and “greater” to use the function torch.greater(). The comparison operator will be called comp(current_value, prev_best). For example, for metrics where the optimal value is low (error, loss, perplexity), use the less than operator.
stop_on_batch (bool, optional) – A bool that indicates whether to stop training in the middle of an epoch if the training metrics satisfy the threshold comparison. Defaults to False.

initialize_object()[source]#

Initialize the ThresholdStopper callback.

Returns: ThresholdStopper – An instance of ThresholdStopper.