avalanche.logging.WandBLogger
- class avalanche.logging.WandBLogger(project_name: str = 'Avalanche', run_name: str = 'Test', log_artifacts: bool = False, path: Union[str, pathlib.Path] = 'Checkpoints', uri: Optional[str] = None, sync_tfboard: bool = False, save_code: bool = True, config: Optional[object] = None, dir: Optional[Union[str, pathlib.Path]] = None, params: Optional[dict] = None)[source]
The WandBLogger provides an easy integration with Weights & Biases logging. Each monitored metric is automatically logged to a dedicated Weights & Biases project dashboard.
External storage for W&B Artifacts (for instance - AWS S3 and GCS buckets) uri are supported.
The wandb log files are placed by default in “./wandb/” unless specified.
Note
TensorBoard can be synced on to the W&B dedicated dashboard.
- __init__(project_name: str = 'Avalanche', run_name: str = 'Test', log_artifacts: bool = False, path: Union[str, pathlib.Path] = 'Checkpoints', uri: Optional[str] = None, sync_tfboard: bool = False, save_code: bool = True, config: Optional[object] = None, dir: Optional[Union[str, pathlib.Path]] = None, params: Optional[dict] = None)[source]
Creates an instance of the WandBLogger. :param project_name: Name of the W&B project. :param run_name: Name of the W&B run. :param log_artifacts: Option to log model weights as W&B Artifacts. :param path: Path to locally save the model checkpoints. :param uri: URI identifier for external storage buckets (GCS, S3). :param sync_tfboard: Syncs TensorBoard to the W&B dashboard UI. :param save_code: Saves the main training script to W&B. :param config: Syncs hyper-parameters and config values used to W&B. :param dir: Path to the local log directory for W&B logs to be saved at. :param params: All arguments for wandb.init() function call.
Visit https://docs.wandb.ai/ref/python/init to learn about all wand.init() parameters.
Methods
__init__
([project_name, run_name, ...])Creates an instance of the WandBLogger. :param project_name: Name of the W&B project. :param run_name: Name of the W&B run. :param log_artifacts: Option to log model weights as W&B Artifacts. :param path: Path to locally save the model checkpoints. :param uri: URI identifier for external storage buckets (GCS, S3). :param sync_tfboard: Syncs TensorBoard to the W&B dashboard UI. :param save_code: Saves the main training script to W&B. :param config: Syncs hyper-parameters and config values used to W&B. :param dir: Path to the local log directory for W&B logs to be saved at. :param params: All arguments for wandb.init() function call. Visit https://docs.wandb.ai/ref/python/init to learn about all wand.init() parameters.
after_backward
(strategy, metric_values, **kwargs)Called after criterion.backward() by the BaseStrategy.
after_eval
(strategy, metric_values, **kwargs)Called after eval by the BaseStrategy.
after_eval_dataset_adaptation
(strategy, ...)Called after eval_dataset_adaptation by the BaseStrategy.
after_eval_exp
(strategy, metric_values, **kwargs)Called after eval_exp by the BaseStrategy.
after_eval_forward
(strategy, metric_values, ...)Called after model.forward() by the BaseStrategy.
after_eval_iteration
(strategy, ...)Called after the end of an iteration by the BaseStrategy.
after_forward
(strategy, metric_values, **kwargs)Called after model.forward() by the BaseStrategy.
after_train_dataset_adaptation
(strategy, ...)Called after train_dataset_adapatation by the BaseStrategy.
after_training
(strategy, metric_values, **kwargs)Called after train by the BaseStrategy.
after_training_epoch
(strategy, ...)Called after train_epoch by the BaseStrategy.
after_training_exp
(strategy, metric_values, ...)Called after train_exp by the BaseStrategy.
after_training_iteration
(strategy, ...)Called after the end of a training iteration by the BaseStrategy.
after_update
(strategy, metric_values, **kwargs)Called after optimizer.update() by the BaseStrategy.
args_parse
()before_backward
(strategy, metric_values, ...)Called before criterion.backward() by the BaseStrategy.
before_eval
(strategy, metric_values, **kwargs)Called before eval by the BaseStrategy.
before_eval_dataset_adaptation
(*args, **kwargs)Called before eval_dataset_adaptation by the BaseStrategy.
before_eval_exp
(strategy, metric_values, ...)Called before eval_exp by the BaseStrategy.
before_eval_forward
(strategy, metric_values, ...)Called before model.forward() by the BaseStrategy.
before_eval_iteration
(strategy, ...)Called before the start of a training iteration by the BaseStrategy.
before_forward
(strategy, metric_values, **kwargs)Called before model.forward() by the BaseStrategy.
before_run
()before_train_dataset_adaptation
(*args, **kwargs)Called before train_dataset_adapatation by the BaseStrategy.
before_training
(strategy, metric_values, ...)Called before train by the BaseStrategy.
before_training_epoch
(strategy, ...)Called before train_epoch by the BaseStrategy.
before_training_exp
(strategy, metric_values, ...)Called before train_exp by the BaseStrategy.
before_training_iteration
(strategy, ...)Called before the start of a training iteration by the BaseStrategy.
before_update
(strategy, metric_values, **kwargs)Called before optimizer.update() by the BaseStrategy.
import_wandb
()log_metric
(metric_value, callback)This method will be invoked on each callback.
log_single_metric
(name, value, x_plot)This abstract method will have to be implemented by each subclass.