python类group()的实例源码

def _default_local_init_op():
    return control_flow_ops.group(variables.initialize_local_variables(),
                                  data_flow_ops.initialize_all_tables())

def _auc_hist_accumulate(hist_true, hist_false, nbins, collections):
  """Accumulate histograms in new variables."""
  with variable_scope.variable_scope(
      None, 'hist_accumulate', [hist_true, hist_false]):
    # Holds running total histogram of scores for records labeled True.
    hist_true_acc = variable_scope.get_variable(
        'hist_true_acc',
        initializer=init_ops.zeros_initializer(
            [nbins],
            dtype=hist_true.dtype),
        collections=collections,
        trainable=False)
    # Holds running total histogram of scores for records labeled False.
    hist_false_acc = variable_scope.get_variable(
        'hist_false_acc',
        initializer=init_ops.zeros_initializer(
            [nbins],
            dtype=hist_false.dtype),
        collections=collections,
        trainable=False)

    update_op = control_flow_ops.group(
        hist_true_acc.assign_add(hist_true),
        hist_false_acc.assign_add(hist_false),
        name='update_op')

    return hist_true_acc, hist_false_acc, update_op

def update_weights(self, train_op):
    """Updates the model weights.

    This function must be called on at least one worker after `minimize`.
    In distributed training this call can be omitted on non-chief workers to
    speed up training.

    Args:
      train_op: The operation returned by the `minimize` call.

    Returns:
      An Operation that updates the model weights.
    """
    with ops.control_dependencies([train_op]):
      update_ops = []
      # Copy over unshrinked weights to user provided variables.
      for name in ['sparse_features_weights', 'dense_features_weights']:
        for var, slot_var in zip(self._variables[name],
                                 self._slots['unshrinked_' + name]):
          update_ops.append(var.assign(slot_var))

    # Apply proximal step.
    with ops.control_dependencies(update_ops):
      update_ops = []
      for name in ['sparse_features_weights', 'dense_features_weights']:
        for var in self._variables[name]:
          with ops.device(var.device):
            # pylint: disable=protected-access
            update_ops.append(
                gen_sdca_ops._sdca_shrink_l1(
                    self._convert_n_to_tensor(
                        [var], as_ref=True),
                    l1=self._symmetric_l1_regularization(),
                    l2=self._symmetric_l2_regularization()))
      return control_flow_ops.group(*update_ops)

def _train_op(self, features, labels, train_op_fn, logits):
    """Returns op for the training step."""
    loss = self._training_loss(features, labels, logits)
    train_op = train_op_fn(loss)

    if self._enable_centered_bias:
      centered_bias_step = [_centered_bias_step(
          self.logits_dimension,
          self._centered_bias_weight_collection,
          labels,
          self._train_loss_fn)]
      train_op = control_flow_ops.group(train_op, *centered_bias_step)

    return train_op

def _train_op(self, features, labels, train_op_fn, logits):
    """Returns op for the training step."""
    loss = self._training_loss(features, labels, logits)
    train_op = train_op_fn(loss)

    if self._enable_centered_bias:
      centered_bias_step = [_centered_bias_step(
          self.logits_dimension,
          self._centered_bias_weight_collection,
          labels,
          self._train_loss_fn)]
      train_op = control_flow_ops.group(train_op, *centered_bias_step)

    return train_op

def _auc_hist_accumulate(hist_true, hist_false, nbins, collections):
  """Accumulate histograms in new variables."""
  with variable_scope.variable_scope(
      None, 'hist_accumulate', [hist_true, hist_false]):
    # Holds running total histogram of scores for records labeled True.
    hist_true_acc = variable_scope.get_variable(
        'hist_true_acc',
        initializer=init_ops.zeros_initializer(
            [nbins],
            dtype=hist_true.dtype),
        collections=collections,
        trainable=False)
    # Holds running total histogram of scores for records labeled False.
    hist_false_acc = variable_scope.get_variable(
        'hist_false_acc',
        initializer=init_ops.zeros_initializer(
            [nbins],
            dtype=hist_false.dtype),
        collections=collections,
        trainable=False)

    update_op = control_flow_ops.group(
        hist_true_acc.assign_add(hist_true),
        hist_false_acc.assign_add(hist_false),
        name='update_op')

    return hist_true_acc, hist_false_acc, update_op

def _AddShardedRestoreOps(self, filename_tensor, per_device,
                            restore_sequentially, reshape):
    """Add Ops to save variables from multiple devices.

    Args:
      filename_tensor: Tensor for the path of the file to load.
      per_device: A list of (device, _VarToSave) pairs, as
        returned by _GroupByDevices().
      restore_sequentially: True if we want to restore variables sequentially
        within a shard.
      reshape: True if we want to reshape loaded tensors to the shape of
        the corresponding variable.

    Returns:
      An Operation that restores the variables.
    """
    sharded_restores = []
    for shard, (device, vars_to_save) in enumerate(per_device):
      with ops.device(device):
        sharded_restores.append(self._AddRestoreOps(
            filename_tensor,
            vars_to_save,
            restore_sequentially,
            reshape,
            preferred_shard=shard,
            name="restore_shard"))
    return control_flow_ops.group(*sharded_restores, name="restore_all")

def _apply_sparse(self, grad, var):
        lr = (self._lr_t *
              math_ops.sqrt(1 - self._beta2_power)
              / (1 - self._beta1_power))
        # m_t = beta1 * m + (1 - beta1) * g_t
        m = self.get_slot(var, "m")
        m_scaled_g_values = grad.values * (1 - self._beta1_t)
        m_t = state_ops.assign(m, m * self._beta1_t,
                               use_locking=self._use_locking)

        m_t = state_ops.scatter_add(m_t, grad.indices, m_scaled_g_values,
                                    use_locking=self._use_locking)
        # v_t = beta2 * v + (1 - beta2) * (g_t * g_t)
        v = self.get_slot(var, "v")
        v_scaled_g_values = (grad.values * grad.values) * (1 - self._beta2_t)
        v_t = state_ops.assign(v, v * self._beta2_t, use_locking=self._use_locking)
        v_t = state_ops.scatter_add(v_t, grad.indices, v_scaled_g_values,
                                    use_locking=self._use_locking)
        v_sqrt = tf.pow(v_t, self._pow_t)
        var_update = state_ops.assign_sub(var,
                                          lr * m_t / (v_sqrt + self._epsilon_t),
                                          use_locking=self._use_locking)
        # regularization
        var_update = state_ops.assign_sub(var_update,
                                          self._sparse_regularization * var,
                                          use_locking=self._use_locking)

        return control_flow_ops.group(*[var_update, m_t, v_t])

def _finish(self, update_ops, name_scope):
        # Update the power accumulators.
        with ops.control_dependencies(update_ops):
            with ops.device(self._beta1_power.device):
                update_beta1 = self._beta1_power.assign(
                        self._beta1_power * self._beta1_t,
                        use_locking=self._use_locking)
                update_beta2 = self._beta2_power.assign(
                        self._beta2_power * self._beta2_t,
                        use_locking=self._use_locking)
        return control_flow_ops.group(*update_ops + [update_beta1, update_beta2],
                                      name=name_scope)

def _apply_dense(self, grad, var):
        lr = (self._lr_t *
              math_ops.sqrt(1 - self._beta2_power)
              / (1 - self._beta1_power))
        # m_t = beta1 * m + (1 - beta1) * g_t
        m = self.get_slot(var, "m")
        m_scaled_g_values = grad * (1 - self._beta1_t)
        m_t = m * self._beta1_t
        m_t = m_t + m_scaled_g_values
        # v_t = beta2 * v + (1 - beta2) * (g_t * g_t)
        v = self.get_slot(var, "v")
        #######################################################################################
        v_scaled_g_values = tf.pow(grad - m_t, 2) * (1 - self._beta2_t)
        #######################################################################################
        v_t = v * self._beta2_t
        v_t = v_t + v_scaled_g_values
        v_sqrt = tf.pow(v_t, self._pow_t)
        var_update = state_ops.assign_sub(var,
                                          lr * m_t / (v_sqrt + self._epsilon_t),
                                          use_locking=self._use_locking)
        # regularization
        var_update = state_ops.assign_sub(var_update,
                                          self._dense_regularization * var,
                                          use_locking=self._use_locking)

        return control_flow_ops.group(*[var_update, m_t, v_t])

def _apply_sparse(self, grad, var):
        lr = (self._lr_t *
              math_ops.sqrt(1 - self._beta2_power)
              / (1 - self._beta1_power))
        # m_t = beta1 * m + (1 - beta1) * g_t
        m = self.get_slot(var, "m")
        m_scaled_g_values = grad.values * (1 - self._beta1_t)
        m_t = state_ops.assign(m, m * self._beta1_t,
                               use_locking=self._use_locking)

        m_t = state_ops.scatter_add(m_t, grad.indices, m_scaled_g_values,
                                    use_locking=self._use_locking)
        # v_t = beta2 * v + (1 - beta2) * (g_t * g_t)
        v = self.get_slot(var, "v")
        v_scaled_g_values = (grad.values * grad.values) * (1 - self._beta2_t)
        v_t = state_ops.assign(v, v * self._beta2_t, use_locking=self._use_locking)
        v_t = state_ops.scatter_add(v_t, grad.indices, v_scaled_g_values,
                                    use_locking=self._use_locking)
        v_sqrt = tf.pow(v_t, self._pow_t)
        var_update = state_ops.assign_sub(var,
                                          lr * m_t / (v_sqrt + self._epsilon_t),
                                          use_locking=self._use_locking)
        # regularization
        var_update = state_ops.assign_sub(var_update,
                                          self._sparse_regularization * var,
                                          use_locking=self._use_locking)

        return control_flow_ops.group(*[var_update, m_t, v_t])

def export(self, last_checkpoint, output_dir):
    """Builds a prediction graph and xports the model.

    Args:
      last_checkpoint: Path to the latest checkpoint file from training.
      output_dir: Path to the folder to be used to output the model.
    """
    logging.info('Exporting prediction graph to %s', output_dir)
    with tf.Session(graph=tf.Graph()) as sess:
      # Build and save prediction meta graph and trained variable values.
      inputs, outputs = self.build_prediction_graph()
      signature_def_map = {
        'serving_default': signature_def_utils.predict_signature_def(inputs, outputs)
      }
      init_op = tf.global_variables_initializer()
      sess.run(init_op)
      self.restore_from_checkpoint(sess, self.inception_checkpoint_file,
                                   last_checkpoint)
      init_op_serving = control_flow_ops.group(
          variables.local_variables_initializer(),
          tf.tables_initializer())

      builder = saved_model_builder.SavedModelBuilder(output_dir)
      builder.add_meta_graph_and_variables(
          sess, [tag_constants.SERVING],
          signature_def_map=signature_def_map,
          legacy_init_op=init_op_serving)
      builder.save(False)

def _extract_metric_update_ops(eval_dict):
        """Separate update operations from metric value operations."""
        update_ops = []
        value_ops = {}
        # Sort metrics lexicographically so graph is identical every time.
        for name, metric_ops in sorted(six.iteritems(eval_dict)):
            value_ops[name] = metric_ops[0]
            update_ops.append(metric_ops[1])

        if update_ops:
            update_op = control_flow_ops.group(*update_ops)
        else:
            update_op = None

        return update_op, value_ops

def _finish(self, update_ops, steps_and_params, name_scope):
    """"""

    return control_flow_ops.group(*update_ops, name=name_scope)

  #=============================================================

def _finish(self, update_ops, steps_and_params, name_scope):
    """"""

    return control_flow_ops.group(*update_ops, name=name_scope)

  #=============================================================

def update_weights(self, train_op):
    """Updates the model weights.

    This function must be called on at least one worker after `minimize`.
    In distributed training this call can be omitted on non-chief workers to
    speed up training.

    Args:
      train_op: The operation returned by the `minimize` call.

    Returns:
      An Operation that updates the model weights.
    """
    with ops.control_dependencies([train_op]):
      update_ops = []
      # Copy over unshrinked weights to user provided variables.
      for name in ['sparse_features_weights', 'dense_features_weights']:
        for var, slot_var in zip(self._variables[name],
                                 self._slots['unshrinked_' + name]):
          update_ops.append(var.assign(slot_var))

    # Apply proximal step.
    with ops.control_dependencies(update_ops):
      update_ops = []
      for name in ['sparse_features_weights', 'dense_features_weights']:
        for var in self._variables[name]:
          with ops.device(var.device):
            # pylint: disable=protected-access
            update_ops.append(
                gen_sdca_ops._sdca_shrink_l1(
                    self._convert_n_to_tensor(
                        [var], as_ref=True),
                    l1=self._symmetric_l1_regularization(),
                    l2=self._symmetric_l2_regularization()))
      return control_flow_ops.group(*update_ops)

def _get_local_init_op():
  local_init_op = _get_first_op_from_collection(
      ops.GraphKeys.LOCAL_INIT_OP)
  if local_init_op is None:
    op_list = [variables.local_variables_initializer(),
               data_flow_ops.tables_initializer()]
    if op_list:
      local_init_op = control_flow_ops.group(*op_list)
      ops.add_to_collection(ops.GraphKeys.LOCAL_INIT_OP, local_init_op)
  return local_init_op

def restore(self, restored_tensors, restored_shapes):
    weights = restored_tensors[:len(restored_tensors) // 2]
    biases = restored_tensors[len(restored_tensors) // 2:]
    params = self._canonical_to_params(weights, biases)
    if not isinstance(params, tuple):
      params = (params,)
    assign_ops = [
        state_ops.assign(
            variable, param, validate_shape=False)
        for variable, param in zip(self._variables, params)
    ]
    return control_flow_ops.group(*assign_ops)

def benchmarkTfRNNLSTMTraining(self):
    test_configs = self._GetTestConfig()
    for config_name, config in test_configs.items():
      num_layers = config["num_layers"]
      num_units = config["num_units"]
      batch_size = config["batch_size"]
      seq_length = config["seq_length"]

      with ops.Graph().as_default(), ops.device("/gpu:0"):
        inputs = seq_length * [
            array_ops.zeros([batch_size, num_units], dtypes.float32)
        ]
        initializer = init_ops.random_uniform_initializer(-0.01, 0.01, seed=127)

        cell = core_rnn_cell_impl.LSTMCell(
            num_units=num_units, initializer=initializer, state_is_tuple=True)
        multi_cell = core_rnn_cell_impl.MultiRNNCell(
            [cell() for _ in range(num_layers)])
        outputs, final_state = core_rnn.static_rnn(
            multi_cell, inputs, dtype=dtypes.float32)
        trainable_variables = ops.get_collection(
            ops.GraphKeys.TRAINABLE_VARIABLES)
        gradients = gradients_impl.gradients([outputs, final_state],
                                             trainable_variables)
        training_op = control_flow_ops.group(*gradients)
        self._BenchmarkOp(training_op, "tf_rnn_lstm %s %s" %
                          (config_name, self._GetConfigDesc(config)))

def training_graph(self, input_data, input_labels, data_spec=None,
                     epoch=None, **tree_kwargs):
    """Constructs a TF graph for training a random forest.

    Args:
      input_data: A tensor or SparseTensor or placeholder for input data.
      input_labels: A tensor or placeholder for labels associated with
        input_data.
      data_spec: A list of tf.dtype values specifying the original types of
        each column.
      epoch: A tensor or placeholder for the epoch the training data comes from.
      **tree_kwargs: Keyword arguments passed to each tree's training_graph.

    Returns:
      The last op in the random forest training graph.
    """
    data_spec = [constants.DATA_FLOAT] if data_spec is None else data_spec
    tree_graphs = []
    for i in range(self.params.num_trees):
      with ops.device(self.device_assigner.get_device(i)):
        seed = self.params.base_random_seed
        if seed != 0:
          seed += i
        # If using bagging, randomly select some of the input.
        tree_data = input_data
        tree_labels = input_labels
        if self.params.bagging_fraction < 1.0:
          # TODO(thomaswc): This does sampling without replacment.  Consider
          # also allowing sampling with replacement as an option.
          batch_size = array_ops.slice(array_ops.shape(input_data), [0], [1])
          r = random_ops.random_uniform(batch_size, seed=seed)
          mask = math_ops.less(
              r, array_ops.ones_like(r) * self.params.bagging_fraction)
          gather_indices = array_ops.squeeze(
              array_ops.where(mask), squeeze_dims=[1])
          # TODO(thomaswc): Calculate out-of-bag data and labels, and store
          # them for use in calculating statistics later.
          tree_data = array_ops.gather(input_data, gather_indices)
          tree_labels = array_ops.gather(input_labels, gather_indices)
        if self.params.bagged_features:
          tree_data = self._bag_features(i, tree_data)

        initialization = self.trees[i].tree_initialization()

        with ops.control_dependencies([initialization]):
          tree_graphs.append(
              self.trees[i].training_graph(
                  tree_data, tree_labels, seed, data_spec=data_spec,
                  epoch=([0] if epoch is None else epoch),
                  **tree_kwargs))

    return control_flow_ops.group(*tree_graphs, name='train')