hook_global_step_waiter: Delay Execution until Global Step Reaches to...

In tfestimators: Interface to 'TensorFlow' Estimators

Description

This hook delays execution until global step reaches to wait_until_step. It is used to gradually start workers in distributed settings. One example usage would be setting wait_until_step=int(K*log(task_id+1)) assuming that task_id=0 is the chief.

Usage

hook_global_step_waiter(wait_until_step)

Arguments

wait_until_step

An integer indicating that until which global step should we wait.

See Also

Other session_run_hook wrappers: hook_checkpoint_saver(), hook_history_saver(), hook_logging_tensor(), hook_nan_tensor(), hook_progress_bar(), hook_step_counter(), hook_stop_at_step(), hook_summary_saver(), session_run_hook()

tfestimators documentation built on Aug. 10, 2021, 1:06 a.m.