# Copyright 2015 The TensorFlow Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """Contains the `Node` class.""" import collections import copy import json import numpy as np import tensorflow.compat.v2 as tf from keras import backend from keras.engine import base_layer_utils from keras.saving.saved_model import json_utils from keras.utils import tf_utils _CONSTANT_VALUE = "_CONSTANT_VALUE" # Using dict to avoid conflict with constant string tensor. _COMPOSITE_TYPE = {"_TYPE": "COMPOSITE"} class Node: """A `Node` describes a layer `__call__()` event. A Functional model is a DAG with `Node` instances as nodes, and `KerasTensor` instances as edges. Nodes aren't `Layer` instances, because a single layer could be called multiple times, which would result in graph cycles. A `__call__()` event involves input tensors (and other input arguments), the layer that was called, and the resulting output tensors. A `Node` will include all this information. Since a single `Layer` could be called multiple times, the `Node` instances are stored on layers as a list. Each time a layer is called a node is added to `layer._inbound_nodes`. Each time the output of a layer is used by another layer, a node is added to `layer._outbound_nodes`. Every `KerasTensor` instance has a `KerasHistory` object attached, which tracks the `Node` that records the `__call__()` event that created the tensor. By recursively walking through `Node` instances via the `KerasHistory` metadata of `KerasTensor` instances, once can retrieve the entire DAG of a Functional model. Args: layer: The layer that was called in the `Layer.__call__()` event that this node represents. call_args: The positional arguments the layer was called with. call_kwargs: The keyword arguments the layer was called with. outputs: The output tensors of the `Layer.__call__()` """ def __init__(self, layer, call_args=None, call_kwargs=None, outputs=None): call_args = [] if call_args is None else call_args call_kwargs = {} if call_kwargs is None else call_kwargs outputs = [] if outputs is None else outputs self.layer = layer self.is_input = not call_args and not call_kwargs # These arguments are user-provided. Copy the structures here so that # future user modifications do not affect the node's metadata. # We copy using map_structure rather than python's shallow or deep copy, # because the args can be data structures (so shallow copy is # insufficient), but individual values might not support copy.copy # or be too expensive to deep copy. call_args = tf.nest.map_structure(lambda t: t, call_args) call_kwargs = tf.nest.map_structure(lambda t: t, call_kwargs) self.outputs = tf.nest.map_structure(lambda t: t, outputs) self.call_args = call_args self.call_kwargs = call_kwargs # Cached for performance. self._flat_arguments = tf.nest.flatten( (self.call_args, self.call_kwargs) ) # Used to avoid expensive `nest` operations in the most common case. self._single_positional_tensor_passed = ( not self.call_kwargs and len(self.call_args) == 1 and tf.is_tensor(self.call_args[0]) ) if not tf.compat.v1.executing_eagerly_outside_functions(): # Create TensorFlowOpLayers if needed (in TF1) for obj in self._flat_arguments: if isinstance( obj, tf.Tensor ) and base_layer_utils.needs_keras_history( obj, ignore_call_context=True ): base_layer_utils.create_keras_history(obj) self._keras_inputs = [] self._keras_inputs_ids_and_indices = [] for i, ele in enumerate(self._flat_arguments): if is_keras_tensor(ele): self._keras_inputs.append(ele) kt_id = str(id(ele)) kt_index = i self._keras_inputs_ids_and_indices.append((kt_id, kt_index)) # Wire up Node to Layers. self.layer._inbound_nodes.append(self) for kt in self.keras_inputs: inbound_layer = kt._keras_history.layer if inbound_layer is not None: # `None` for `Input` tensors. inbound_layer._outbound_nodes.append(self) # Set metadata on outputs. node_index = len(self.layer._inbound_nodes) - 1 for i, tensor in enumerate(tf.nest.flatten(outputs)): tensor._keras_history = KerasHistory( layer=layer, node_index=node_index, tensor_index=i ) # Cached for performance. self.flat_input_ids = [str(id(t)) for t in self._keras_inputs] self.flat_output_ids = [ str(id(t)) for t in tf.nest.flatten(self.outputs) ] @property def keras_inputs(self): """Tensors input to this node that can be traced back to a `keras.Input`.""" return self._keras_inputs @property def parent_nodes(self): """Returns all the `Node`s whose output this node immediately depends on.""" node_deps = [] for kt in self.keras_inputs: layer = kt._keras_history.layer node_index = kt._keras_history.node_index if layer is not None: # `None` for `Input` tensors. node_deps.append(layer._inbound_nodes[node_index]) return node_deps def iterate_inbound(self): """Yields tuples representing the data inbound from other nodes. Yields: tuples like: (inbound_layer, node_index, tensor_index, tensor). """ for kt in self.keras_inputs: keras_history = kt._keras_history layer = keras_history.layer node_index = keras_history.node_index tensor_index = keras_history.tensor_index yield layer, node_index, tensor_index, kt def map_arguments(self, tensor_dict): """Maps Keras Tensors to computed Tensors using `tensor_dict`.""" if self._single_positional_tensor_passed: # Performance optimization for most common case. kt_id, _ = self._keras_inputs_ids_and_indices[0] return (tensor_dict[kt_id].pop(),), {} else: flat_arguments = copy.copy(self._flat_arguments) for kt_id, kt_index in self._keras_inputs_ids_and_indices: flat_arguments[kt_index] = tensor_dict[kt_id].pop() args, kwargs = tf.nest.pack_sequence_as( (self.call_args, self.call_kwargs), flat_arguments ) return args, kwargs def serialize(self, make_node_key, node_conversion_map): """Serializes `Node` for Functional API's `get_config`.""" # Serialization still special-cases first argument. args, kwargs = self.call_args, self.call_kwargs inputs, args, kwargs = self.layer._call_spec.split_out_first_arg( args, kwargs ) # Treat everything other than first argument as a kwarg. arguments = dict(zip(self.layer._call_spec.arg_names[1:], args)) arguments.update(kwargs) kwargs = arguments def _serialize_keras_tensor(t): """Serializes a single Tensor passed to `call`.""" if hasattr(t, "_keras_history"): kh = t._keras_history node_index = kh.node_index node_key = make_node_key(kh.layer.name, node_index) new_node_index = node_conversion_map.get(node_key, 0) return [kh.layer.name, new_node_index, kh.tensor_index] if isinstance(t, np.ndarray): return t.tolist() if isinstance(t, tf.Tensor): return backend.get_value(t).tolist() # Not using json_utils to serialize both constant Tensor and # constant CompositeTensor for saving format backward compatibility. if isinstance(t, tf.__internal__.CompositeTensor): return (_COMPOSITE_TYPE, json_utils.Encoder().encode(t)) return t kwargs = tf.nest.map_structure(_serialize_keras_tensor, kwargs) try: json.dumps(kwargs, default=json_utils.get_json_type) except TypeError: kwarg_types = tf.nest.map_structure(type, kwargs) raise TypeError( "Layer " + self.layer.name + " was passed non-JSON-serializable arguments. " + "Arguments had types: " + str(kwarg_types) + ". They cannot be serialized out " "when saving the model." ) # `kwargs` is added to each Tensor in the first arg. This should be # changed in a future version of the serialization format. def serialize_first_arg_tensor(t): if is_keras_tensor(t): kh = t._keras_history node_index = kh.node_index node_key = make_node_key(kh.layer.name, node_index) new_node_index = node_conversion_map.get(node_key, 0) data = [kh.layer.name, new_node_index, kh.tensor_index, kwargs] else: # If an element in the first call argument did not originate as # a keras tensor and is a constant value, we save it using the # format ['_CONSTANT_VALUE', -1, # serialized_tensor_or_python_constant] (potentially including # serialized kwargs in an optional 4th argument). data = [_CONSTANT_VALUE, -1, _serialize_keras_tensor(t), kwargs] return tf_utils.ListWrapper(data) data = tf.nest.map_structure(serialize_first_arg_tensor, inputs) if ( not tf.nest.is_nested(data) and not self.layer._preserve_input_structure_in_config ): data = [data] data = tf_utils.convert_inner_node_data(data) return data ############################################################# # Properties for Backwards compatibility. # These only check the first input argument # As nodes are internal, they may be removed in the future. ############################################################# @property def input_tensors(self): if self.is_input: return [self.outputs] # Used in `Layer.input`. return self.call_args[0] @property def output_tensors(self): if self.is_input: return [self.outputs] # Used in `Layer.input`. return self.outputs @property def input_shapes(self): input_shapes = tf.nest.map_structure( backend.int_shape, self.input_tensors ) if len(input_shapes) == 1 and not self.is_input: return input_shapes[0] return input_shapes @property def output_shapes(self): return tf.nest.map_structure(backend.int_shape, self.output_tensors) @property def outbound_layer(self): return self.layer @property def inbound_layers(self): """Return all layers that feed into the current node.""" if self.is_input: return [] tensor_call_args = [ x for x in self._flat_arguments if tf.is_tensor(x) and hasattr(x, "_keras_history") ] inbound_layers = tf.nest.map_structure( lambda t: t._keras_history.layer, tensor_call_args ) if len(inbound_layers) == 1: return inbound_layers[0] return inbound_layers class KerasHistory( collections.namedtuple( "KerasHistory", ["layer", "node_index", "tensor_index"] ) ): """Tracks the Layer call that created a Tensor, for Keras Graph Networks. During construction of Keras Graph Networks, this metadata is added to each Tensor produced as the output of a Layer, starting with an `InputLayer`. This allows Keras to track how each Tensor was produced, and this information is later retraced by the `keras.engine.Network` class to reconstruct the Keras Graph Network. Attributes: layer: The Layer that produced the Tensor. node_index: The specific call to the Layer that produced this Tensor. Layers can be called multiple times in order to share weights. A new node is created every time a Layer is called. The corresponding node that represents the call event that produced the Tensor can be found at `layer._inbound_nodes[node_index]`. tensor_index: The output index for this Tensor. Always zero if the Layer that produced this Tensor only has one output. Nested structures of Tensors are deterministically assigned an index via `nest.flatten`. """ # Added to maintain memory and performance characteristics of `namedtuple` # while subclassing. __slots__ = () def is_keras_tensor(obj): return hasattr(obj, "_keras_history")