org.tensorflow.proto.framework

 An allocation/de-allocation operation performed by the allocator. 

 Description of the graph-construction-time configuration of this
 Op. 

 Used to specify and override the default API & behavior in the
 generated code for client languages, from what you would get from
 the OpDef alone. 

 If you specify any endpoint, this will replace all of the
 inherited endpoints. 

 An asset file def for a single file or a set of sharded files with the same
 name. 

 Protocol buffer representing the value for an attr used to configure an Op. 

 LINT.IfChange
 

 A protobuf to represent tf.BoundedTensorSpec. 

 Defines a subgraph in another `GraphDef` as a set of feed points and nodes
 to be fetched or executed. 

 AnyList is used for collecting Any protos. 

 CollectionDef should cover most collections. 

 BytesList is used for collecting strings and serialized protobufs. 

 FloatList is used for collecting float values. 

 Int64List is used for collecting int, int64 and long values. 

 NodeList is used for collecting nodes in graph. 

 Protocol buffer representing a CondContext object. 

 Session configuration parameters. 

 Everything inside Experimental is subject to change and is not subject
 to API stability guarantees in
 https://www.tensorflow.org/guide/version_compat. 

 Container for any kind of control flow context. 

 Total cost of this graph, typically used for balancing decisions. 

 Inputs of this node. 

 Outputs of this node. 

 Options for initializing DebuggerState in TensorFlow Debugger (tfdbg). 

 Option for watching a node in TensorFlow Debugger (tfdbg). 

 Represents a Python dict keyed by `str`. 

 Attributes for function arguments. 

 A function can be instantiated when the runtime can bind every attr
 with a value. 

 A library is a set of named functions. 

 Represents `FunctionSpec` used in `Function`. 

 Configuration for breaking down a visible GPU into multiple "virtual"
 devices. 

 GradientDef defines the gradient function of a function defined in
 a function library. 

 This represents a file/line location in the source code. 

 This represents a stack trace which is a ordered list of `FileLineCol`. 

 Represents the graph of operations
 

 Protocol buffer representing a handle to a tensorflow resource. 

 Serialization format for histogram module in
 core/lib/histogram/histogram.h
 

 A collection of KernelDefs
 

 Represents a Python list. 

 For memory tracking. 

 NOTE: This protocol buffer is evolving, and will go through revisions in the
 coming months. 

 Meta information regarding the graph to be exported. 

 A list of attr names and their values. 

 A pair of tensor name and tensor values. 

 Represents Python's namedtuple. 

 Time/size stats recorded for a single execution of a graph node. 

 Output sizes recorded for a single execution of a graph node. 

 Represents None. 

 Defines an operation. 

 For describing inputs and outputs. 

 Description of the graph-construction-time configuration of this
 Op. 

 Information about version-dependent deprecation of an op
 

 A collection of OpDefs
 

 Options passed to the graph optimizer
 

 Represents a (key, value) pair. 

 Protocol buffer representing a QueueRunner. 

 For serializing and restoring the state of ReaderBase, see
 reader_base.h for details. 

 Protocol buffer representing a handle to a tensorflow resource. 

 Protocol buffer representing a handle to a tensorflow resource. 

 Protocol buffer representing a pair of (data type, tensor shape). 

 Graph rewriting is experimental and subject to change, not covered by any
 API stability guarantees. 

 Message to describe custom graph optimizer and its parameters
 

 Metadata output (i.e., non-Tensor) for a single Run() call. 

 Options for a single Run() call. 

 Everything inside Experimental is subject to change and is not subject
 to API stability guarantees in
 https://www.tensorflow.org/guide/version_compat. 

 Options for run handler thread pool. 

 A SavedAsset points to an asset in the MetaGraph. 

 Stores low-level information about a concrete function. 

 A function with multiple signatures, possibly with non-Tensor arguments. 

 SavedModel is the high level serialization format for TensorFlow Models. 

 A SavedResource represents a TF object that holds state during its lifetime. 

 A SavedUserObject is an object (in the object-oriented language of the
 TensorFlow program) of some user- or framework-defined class other than
 those handled specifically by the other kinds of SavedObjects. 

 Represents a Variable that is initialized by loading the contents from the
 checkpoint. 

 Metadata about the session. 

 SignatureDef defines the signature of a computation supported by a TensorFlow
 graph. 

 `StructuredValue` represents a dynamically typed value representing various
 data structures that are inspired by Python data structures typically used in
 TensorFlow functions as inputs and outputs. 

 A Summary is a set of named values to be displayed by the
 visualizer. 

 Metadata associated with a series of Summary data
 

 A SummaryMetadata encapsulates information on which plugins are able to make
 use of a certain summary value. 

 Defines a connection between two tensors in a `GraphDef`. 

 Information about a Tensor necessary for feeding or retrieval. 

 Generic encoding for composite tensors. 

 For sparse tensors, The COO encoding stores a triple of values, indices,
 and shape. 

 Protocol buffer representing a tensor. 

 Dimensions of a tensor. 

 One dimension of the tensor. 

 Can only be interpreted if you know the corresponding TensorShape. 

 Extent of the slice in one dimension. 

 A protobuf to represent tf.TensorSpec. 

 Represents a Python tuple. 

 Represents a tf.TypeSpec
 

 Protocol buffer representing the values in ControlFlowContext. 

 Protocol buffer representing a Variable. 

 Protocol buffer representing the serialization format of DT_VARIANT tensors. 

 The config for graph verifiers. 

 Version information for a piece of serialized data
 There are different types of versions for each type of data
 (GraphDef, etc.), but they all have the same common shape
 described here. 

 Protocol buffer representing a WhileContext object. 

 The canonical error codes for TensorFlow APIs. 

 An enum that describes the state of the MLIR bridge rollout. 

 Whether the function should be compiled by XLA. 

 Control the use of the compiler/jit. 

 Enum for layout conversion between NCHW and NHWC on CPU. 

 Enum controlling the number of times to run optimizers. 

 TODO(pbar) Turn this into a TraceOptions proto which allows
 tracing to be controlled in a more orthogonal manner?
 

 For identifying the underlying type of a variant. 

 Indicates how a distributed variable will be aggregated.