[−][src]Module tensorflow_proto::xla
Modules
| buffer_allocation_proto | Nested message and enum types in |
| buffer_assignment_proto | Nested message and enum types in |
| channel_handle | Nested message and enum types in |
| debug_options | Nested message and enum types in |
| device_assignment_proto | Nested message and enum types in |
| dynamic_parameter_binding_proto | Nested message and enum types in |
| gpu | |
| heap_simulator_trace | Nested message and enum types in |
| hlo_input_output_alias_proto | Nested message and enum types in |
| hlo_instruction_proto | Nested message and enum types in |
| hlo_profile_printer_data | Nested message and enum types in |
| hlo_reduce_precision_options | Nested message and enum types in |
| hlo_schedule_proto | Nested message and enum types in |
| logical_buffer_proto | Nested message and enum types in |
| op_sharding | Nested message and enum types in |
| padding_config | Nested message and enum types in |
| precision_config | Nested message and enum types in |
| triangular_solve_options | Nested message and enum types in |
| while_loop_backend_config | Nested message and enum types in |
Structs
| BufferAllocationProto | Serialization of BufferAllocation. |
| BufferAssignmentProto | Serialization of BufferAssignment. |
| ChannelHandle | Handle given to a user to represent a channel between two computations via a Send and Recv instruction pair. Channels are unbuffered, so Send Send instructions will be blocked until the data is transferred. |
| CholeskyOptions | |
| CompileRequest | |
| CompileResponse | |
| ComputationGraphStatsRequest | |
| ComputationStats | Statistics of a computation. |
| ComputationStatsResponse | |
| ComputeConstantGraphRequest | |
| ComputeConstantResponse | |
| ConvolutionDimensionNumbers | |
| CreateChannelHandleRequest | |
| CreateChannelHandleResponse | |
| DebugOptions | Debugging options for XLA. These options may change at any time - there are no guarantees about backward or forward compatibility for these fields. |
| DeconstructTupleRequest | |
| DeconstructTupleResponse | |
| DeviceAssignmentProto | DeviceAssignmentProto is a serialized form of DeviceAssignment class, which represents the device ids assigned to a set of replicated computations. See xla::DeviceAssignment class comment for more details. |
| DeviceHandle | Handle given to a user that represents a replicated virtual device. Each replicated device represents N physical devices for execution where N is the number of replicas. |
| DotDimensionNumbers | |
| DynamicParameterBindingProto | |
| ExecuteGraphParallelRequest | |
| ExecuteGraphRequest | TODO(b/118493728): Remove this and ExecuteGraphParallelRequest and replace the uses with calls to Compile and Execute. |
| ExecuteParallelResponse | |
| ExecuteRequest | |
| ExecuteResponse | |
| ExecutionHandle | Handle given to a user that represents an execution that the user launched asynchronously on the device. |
| ExecutionOptions | These settings control how XLA compiles and/or runs code. Not all settings will have an effect on every platform. |
| ExecutionProfile | Profile data from the execution of a computation. |
| GatherDimensionNumbers | Describes the dimension numbers for a gather operation. |
| GetDeviceHandlesRequest | |
| GetDeviceHandlesResponse | |
| GetShapeRequest | |
| GetShapeResponse | |
| GlobalDataHandle | Handle given to a user that represents a globally accessible allocation. Contrast this against a ComputationDataHandle, which is not globally accessible, since it only exists within a specific computation. |
| HeapSimulatorTrace | A trace of a HeapSimulator run. |
| HloComputationProto | Serialization of HloComputation. |
| HloInputOutputAliasProto | |
| HloInstructionProto | Serialization of HloInstruction. Next ID: 68 |
| HloModuleGroupProto | An abstraction representing a set of HLO module built to run concurrently across different devices. |
| HloModuleProto | Serialization of HloModule. |
| HloProfilePrinterData | Describes how to pretty-print a profile counter array gathered for a specific HloModule. |
| HloProto | Grouping message that contains all of the information above. |
| HloReducePrecisionOptions | Options for the HLO insert-reduce-precision-operations pass. |
| HloScheduleProto | Serialization of an HLO schedule. An HLO schedule contains a total order of instructions for each non-fusion computation in the module. |
| HloSnapshot | Encapsulates HloProto together with the arguments, result, and execution_platform. This message is used for purposes such as analysis/replay/file-storage. |
| LayoutProto | A layout describes how the array is placed in (1D) memory space. This includes the minor-to-major ordering of dimensions within a shape. |
| LiteralProto | Literals are used when the server and client need to exchange materialized data / results. Literals are also used to describe constants used in computations. |
| LoadDataRequest | |
| LoadDataResponse | |
| LogicalBufferProto | Serialization of LogicalBuffer. |
| OpMetadata | Symbolization metadata for HLO Instructions. |
| OpSharding | |
| PaddingConfig | Describes the padding configuration for Pad operation. The padding amount on both edges as well as between the elements are specified for each dimension. |
| ParameterReplication | Describes whether all data-parallelism replicas will receive the same parameter data at each buffer. |
| PrecisionConfig | Used to indicate the precision configuration. It has backend specific meaning. |
| ProgramShapeProto | Shape of the parameters and output of a computation (like a traditional function signature). |
| ReplicaGroup | Describes the replica groups in a cross replica op (e.g., all-reduce and all-to-all). |
| ResetDeviceRequest | |
| ResetDeviceResponse | |
| ScatterDimensionNumbers | Describes the dimension numbers for a scatter operation. |
| ShapeProto | A shape describes the number of dimensions in the array, the size of each dimension, and the primitive component type. |
| SourceTarget | Describes the source target pair in the collective permute op. |
| TileProto | Describes a tile used in tiling-based layout. Refer to g3doc/third_party/tensorflow/compiler/xla/g3doc/layout_with_tiling.md for details about tiling-based layout. |
| TransferFromOutfeedRequest | |
| TransferFromOutfeedResponse | |
| TransferToClientRequest | |
| TransferToClientResponse | |
| TransferToInfeedRequest | |
| TransferToInfeedResponse | |
| TransferToServerRequest | |
| TransferToServerResponse | |
| TriangularSolveOptions | |
| UnpackRequest | |
| UnpackResponse | |
| UnregisterRequest | |
| UnregisterResponse | |
| WaitForExecutionRequest | |
| WaitForExecutionResponse | |
| WhileLoopBackendConfig | A backend-config for kWhile loops that stores the loop's trip count, if it is known. |
| Window | Describes the windowing in an operation such as convolution. |
| WindowDimension |
Enums
| FftType | |
| Format | A format specifies the method used by a layout to store an array in memory. |
| PrimitiveType | Primitive types are the individual values that can be held in rectangular multidimensional arrays. A description of the rectangular multidimensional array dimensions / primitive type is given by Shape, below. |
| RandomDistribution |