Python tensorflow.python.training.evaluation._StopAfterNEvalsHook() Examples

def _convert_eval_steps_to_hooks(self, steps):
    with self._ctx.with_mode(model_fn_lib.ModeKeys.EVAL) as ctx:
      if ctx.is_running_on_cpu():
        return super(TPUEstimator, self)._convert_eval_steps_to_hooks(steps)

    if steps is None:
      raise ValueError('Evaluate `steps` must be set on TPU. Cannot be `None`.')

    util_lib.check_positive_integer(steps, 'Eval steps')

    return [
        evaluation._StopAfterNEvalsHook(  # pylint: disable=protected-access
            num_evals=steps),
        _SetEvalIterationsHook(steps)
    ] 

def _convert_eval_steps_to_hooks(self, steps):
    with self._ctx.with_mode(model_fn_lib.ModeKeys.EVAL) as ctx:
      if ctx.is_running_on_cpu():
        return super(TPUEstimator, self)._convert_eval_steps_to_hooks(steps)

    if steps is None:
      raise ValueError('Evaluate `steps` must be set on TPU. Cannot be `None`.')

    util_lib.check_positive_integer(steps, 'Eval steps')

    return [
        evaluation._StopAfterNEvalsHook(  # pylint: disable=protected-access
            num_evals=steps),
        _SetEvalIterationsHook(steps)
    ] 

def _convert_eval_steps_to_hooks(self, steps):
    with self._ctx.with_mode(model_fn_lib.ModeKeys.EVAL) as ctx:
      if ctx.is_running_on_cpu():
        return super(TPUEstimator, self)._convert_eval_steps_to_hooks(steps)

    if steps is None:
      raise ValueError('Evaluate `steps` must be set on TPU. Cannot be `None`.')

    util_lib.check_positive_integer(steps, 'Eval steps')

    return [
        evaluation._StopAfterNEvalsHook(  # pylint: disable=protected-access
            num_evals=steps),
        _SetEvalIterationsHook(steps)
    ] 

def _convert_eval_steps_to_hooks(self, steps):
    if steps is None:
      return []

    if steps <= 0:
      raise ValueError('Must specify steps > 0, given: {}'.format(steps))
    return [evaluation._StopAfterNEvalsHook(num_evals=steps)]  # pylint: disable=protected-access 

def _convert_eval_steps_to_hooks(self, steps):
    with self._ctx.with_mode(model_fn_lib.ModeKeys.EVAL) as ctx:
      if ctx.is_running_on_cpu():
        return super(TPUEstimator, self)._convert_eval_steps_to_hooks(steps)

    if steps is None:
      raise ValueError('Evaluate `steps` must be set on TPU. Cannot be `None`.')

    util_lib.check_positive_integer(steps, 'Eval steps')

    return [
        evaluation._StopAfterNEvalsHook(  # pylint: disable=protected-access
            num_evals=steps),
        _SetEvalIterationsHook(steps)
    ] 

def _convert_eval_steps_to_hooks(self, steps):
    with self._ctx.with_mode(model_fn_lib.ModeKeys.EVAL) as ctx:
      if ctx.is_running_on_cpu():
        return super(TPUEstimator, self)._convert_eval_steps_to_hooks(steps)

    if steps is None:
      raise ValueError('Evaluate `steps` must be set on TPU. Cannot be `None`.')

    util_lib.check_positive_integer(steps, 'Eval steps')

    return [
        evaluation._StopAfterNEvalsHook(  # pylint: disable=protected-access
            num_evals=steps),
        _SetEvalIterationsHook(steps)
    ] 

def _convert_eval_steps_to_hooks(self, steps):
    """Create hooks to run correct number of steps in evaluation.

    Args:
      steps: number of steps to run during evaluation.

    Raises:
      ValueError: if steps is less than or equal to zero.

    Returns:
      List of hooks to be passed to the estimator.
    """
    if steps is None:
      return []

    if steps <= 0:
      raise ValueError('Must specify steps > 0, given: {}'.format(steps))

    # The hooks are declared as private in evaluation.py discourage the use
    # by other libraries or open source users. This should be the only usage
    # of the estimator evaluation hooks.
    if self._eval_distribution:
      steps_per_run = getattr(self._eval_distribution.extended, 'steps_per_run',
                              1)
      if steps_per_run > 1:
        return [
            evaluation._MultiStepStopAfterNEvalsHook(  # pylint: disable=protected-access
                num_evals=steps,
                steps_per_run=steps_per_run)
        ]
    return [evaluation._StopAfterNEvalsHook(num_evals=steps)]  # pylint: disable=protected-access 

def _convert_eval_steps_to_hooks(self, steps):
    if steps is None:
      return []

    if steps <= 0:
      raise ValueError('Must specify steps > 0, given: {}'.format(steps))
    return [evaluation._StopAfterNEvalsHook(num_evals=steps)]  # pylint: disable=protected-access 

def evaluate(self, input_fn, steps=None, hooks=None, checkpoint_path=None,
               name=None):
    """Evaluates the model given evaluation data input_fn.

    For each step, calls `input_fn`, which returns one batch of data.
    Evaluates until:
    - `steps` batches are processed, or
    - `input_fn` raises an end-of-input exception (`OutOfRangeError` or
    `StopIteration`).

    Args:
      input_fn: Input function returning a tuple of:
          features - Dictionary of string feature name to `Tensor` or
            `SparseTensor`.
          labels - `Tensor` or dictionary of `Tensor` with labels.
      steps: Number of steps for which to evaluate model. If `None`, evaluates
        until `input_fn` raises an end-of-input exception.
      hooks: List of `SessionRunHook` subclass instances. Used for callbacks
        inside the evaluation call.
      checkpoint_path: Path of a specific checkpoint to evaluate. If `None`, the
        latest checkpoint in `model_dir` is used.
      name: Name of the evaluation if user needs to run multiple evaluations on
        different data sets, such as on training data vs test data. Metrics for
        different evaluations are saved in separate folders, and appear
        separately in tensorboard.

    Returns:
      A dict containing the evaluation metrics specified in `model_fn` keyed by
      name, as well as an entry `global_step` which contains the value of the
      global step for which this evaluation was performed.

    Raises:
      ValueError: If `steps <= 0`.
      ValueError: If no model has been trained, namely `model_dir`, or the
        given `checkpoint_path` is empty.
    """
    hooks = _check_hooks_type(hooks)
    if steps is not None:
      if steps <= 0:
        raise ValueError('Must specify steps > 0, given: {}'.format(steps))
      hooks.append(evaluation._StopAfterNEvalsHook(  # pylint: disable=protected-access
          num_evals=steps))

    return self._evaluate_model(
        input_fn=input_fn,
        hooks=hooks,
        checkpoint_path=checkpoint_path,
        name=name)