TensorBoardLoggerCallback
Log scalar metrics + eval scalars to TensorBoard, one event dir per arm.
Opens a SummaryWriter in on_run_start (lazy import so the callback
registers without torch installed) and closes it in on_run_end.
Attributes
attributenamestr= 'tensorboard_logger'attributeConfigtype= TensorBoardLoggerConfigFunctions
func__init__(self, *, log_dir=None, flush_secs=30) -> Noneparamselfparamlog_dirstr | None= Noneparamflush_secsint= 30Returns
Nonefuncon_run_start(self, ctx) -> NoneparamselfparamctxLogContextReturns
Nonefuncon_step_end(self, state, step_idx, batch, metrics) -> NoneparamselfparamstateLoopStateparamstep_idxparambatchparammetricsReturns
Nonefuncon_eval(self, state, step_idx, eval_name, metrics) -> NoneparamselfparamstateLoopStateparamstep_idxparameval_nameparammetricsReturns
Nonefuncon_benchmark_eval(self, ctx, eval_result, predictions, *, step=None) -> Noneparamselfparamctxparameval_resultparampredictionsparamstep= NoneReturns
Nonefuncon_run_end(self, ctx, run_result, arm) -> Noneparamselfparamctxparamrun_resultparamarmReturns
None