TensorBoardLogger

class braintools.trainer.TensorBoardLogger(save_dir, name='default', version=None, log_graph=False, default_hp_metric=True, prefix='', **kwargs)

TensorBoard logging backend.

Parameters:

• save_dir (str) – Directory to save TensorBoard logs.

• name (str) – Experiment name (subdirectory).

• version (str | None) – Version string. If None, auto-generates based on timestamp.

• log_graph (bool) – Whether to log the model graph.

• default_hp_metric (bool) – Whether to log hyperparameters with a default metric.

• prefix (str) – Prefix for all metric names.

Examples

>>> logger = TensorBoardLogger('logs/', name='my_experiment')
>>> trainer = Trainer(logger=logger)

finalize(status='success')

Finalize logging (called at end of training).

Parameters:

status (str) – Final status (‘success’, ‘failed’, ‘interrupted’).

property log_dir: str

Directory for this specific run’s logs.

log_graph(model, input_array=None)

Log model graph/architecture.

Parameters:

• model (Any) – The model to log.

• input_array (Any) – Sample input for tracing.

log_hyperparams(params)

Log hyperparameters.

Parameters:

params (Dict[str, Any]) – Dictionary of hyperparameter names to values.

log_image(key, images, step=None)

Log images.

Parameters:

• key (str) – Image key/tag.

• images (Any) – Images to log (numpy array or list).

• step (int | None) – Global step number.

log_metrics(metrics, step=None)

Log metrics.

Parameters:

• metrics (Dict[str, float]) – Dictionary of metric names to values.

• step (int | None) – Global step number.

log_text(key, text, step=None)

Log text.

Parameters:

• key (str) – Text key/tag.

• text (str) – Text to log.

• step (int | None) – Global step number.

property root_dir: str

Root directory for logs.

save()

Save/flush any buffered data.