Expand description
ProtocolBuffer types compiled from TensorFlow.
The types are provided by ProtocolBuffer documents from TensorFlow repository. They are used internally for {,de}serialization.
Modules
- Nested message and enum types in
ApiDef
. - Nested message and enum types in
AttrValue
. - Nested message and enum types in
CostGraphDef
. - Nested message and enum types in
Event
. - Nested message and enum types in
Feature
. - Nested message and enum types in
FeatureConfiguration
. - Nested message and enum types in
FullTypeDef
. - Nested message and enum types in
FunctionDef
. - Nested message and enum types in
GraphDebugInfo
. - Nested message and enum types in
GraphTransferInfo
. - Nested message and enum types in
KernelDef
. - Nested message and enum types in
LogMessage
. - Nested message and enum types in
NodeDef
. - Nested message and enum types in
OpDef
. - Nested message and enum types in
OptimizedFunctionGraph
. - Nested message and enum types in
ResourceHandleProto
. - Nested message and enum types in
SessionLog
. - Nested message and enum types in
Summary
. - Nested message and enum types in
SummaryMetadata
. - Nested message and enum types in
TensorShapeProto
. - Nested message and enum types in
TensorSliceProto
.
Structs
- An allocation/de-allocation operation performed by the allocator.
- 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. There will be a set of ApiDefs that are common to all client languages, and another set per client language. The per-client-language ApiDefs will inherit values from the common ApiDefs which it can either replace or modify.
- Protocol buffer representing the value for an attr used to configure an Op. Comment indicates the corresponding attr type. Only the field matching the attr type may be filled.
- LINT.IfChange Containers to hold repeated fundamental values.
- Protocol buffer representing an event that happened during the execution of a Brain model.
- Containers for non-sequential data.
- Containers for sequential data.
- Highly experimental and very likely to change. This encoding uses tags instead of dedicated messages for regularity. In particular the encoding imposes no restrictions on what the parameters of any type should be, which in particular needs to be true for type symbols.
- A function can be instantiated when the runtime can bind every attr with a value. When a GraphDef has a call to a function, it must have binding for every attr defined in the signature.
- A library is a set of named functions.
- GradientDef defines the gradient function of a function defined in a function library.
- Represents the graph of operations
- Protocol buffer representing a handle to a tensorflow resource. Handles are not valid across executions, but can be serialized back and forth from within a single run.
- Serialization format for histogram module in tsl/lib/histogram/histogram.h
- A collection of KernelDefs
- Protocol buffer used for logging messages to the events file.
- For memory tracking.
- A list of attr names and their values. The whole list is attached with a string name. E.g., MatMul[T=float].
- Time/size stats recorded for a single execution of a graph node.
- Output sizes recorded for a single execution of a graph node.
- Defines an operation. A NodeDef in a GraphDef specifies an Op by using the “op” field which should match the name of a OpDef. LINT.IfChange
- Information about version-dependent deprecation of an op
- A collection of OpDefs
- Optimized function graph after instantiation-related graph optimization passes (up till before graph partitioning). The first half of the proto is representing a GraphDef and the rest of the fields are extra information from graph optimizations.
- For serializing and restoring the state of ReaderBase, see reader_base.h for details.
- RegisteredGradient stores a gradient function that is registered in the gradients library and used in the ops of a function in the function library. Unlike GradientDef, these gradients are identified by op type, and not directly linked to any function.
- Protocol buffer representing a handle to a tensorflow resource. Handles are not valid across executions, but can be serialized back and forth from within a single run.
- Represents a serialized tf.dtypes.Dtype
- Protocol buffer used for logging session state.
- Holds the information of the source that writes the events.
- 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.
- For logging the metadata output for a single session.run() call.
- Protocol buffer representing a tensor.
- Dimensions of a tensor.
- Can only be interpreted if you know the corresponding TensorShape.
- Protocol buffer representing a Variable.
- Protocol buffer representing the serialization format of DT_VARIANT tensors.
- Version information for a piece of serialized data
Enums
- (== suppress_warning documentation-presence ==) LINT.IfChange
- LINT.IfChange Experimental. Represents the complete type information of a TensorFlow value.
- Indicates how a distributed variable will be aggregated.
- Indicates when a distributed variable will be synced.
- Current health status of a worker.
- Indicates the behavior of the worker when an internal error or shutdown signal is received.