Delay execution until global step reaches to wait_until_step. (A mirror to GlobalStepWaiterHook.)

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.

  • Args:
    • wait_until_step: an int shows until which global step should we wait.



polyaxon.estimators.hooks.general_hooks.FinalOpsHook(final_ops, final_ops_feed_dict=None)

A run hook which evaluates Tensors at the end of a session. (A mirror to GlobalStepWaiterHook.)

  • Args:
    • final_ops: A single Tensor, a list of Tensors or a dictionary of names to Tensors.
    • final_ops_feed_dict: A feed dictionary to use when running final_ops_dict.



polyaxon.estimators.hooks.general_hooks.StopAfterNEvalsHook(num_evals, log_progress=True)

Run hook used by the evaluation routines to run the eval_ops N times.



polyaxon.estimators.hooks.general_hooks.NanTensorHook(loss_tensor, fail_on_nan_loss=True)

NaN Loss monitor.

A modified version of NanTensorHook. Checks the context for no_run_hooks_op before calling the the hook.

Monitors loss and stops training if loss is NaN. Can either fail with exception or just stop training.

  • Args:
    • loss_tensor: Tensor, the loss tensor.
    • fail_on_nan_loss: bool, whether to raise exception when loss is NaN.