TensorFlowSharp API
Classes
MonoPInvokeCallbackAttribute
Thisattribute can be applied to callback functions that will be invoked fromunmanaged code to managed code.
此属性可应用于回调函数，该函数将从非托管代码调用到托管代码
PaddingFIFOQueue
AFIFOQueue that supports batching variable-sized tensors by padding.
可以填充变长张量的先入先出队列
QueueBase
Baseclass for queue implementations.
队列的基类
TFBuffer
Holdsa block of data, suitable to pass, or retrieve from TensorFlow.
存储数据块，便于TF进行传递和检索
TFCore
ContainsTensorFlow fundamental methods and utility functions.
包含实用方法和效用函数
TFDependencies
TFGraphvariable dependencies handle.
用于处理Graph的依赖关系
TFDisposable
Baseclass for many TensorFlow data types that provides a common idiom to disposeand release resources associated with the native data types. Generally, you donot need to use this.许多TensorFlow数据类型的基类，它提供了一个通用的习惯用法来处理和释放与本机数据类型相关的资源。通常，您不需要使用这个
TFException
TensorFlowException
异常
TFFunction
Agrouping of operations with defined inputs and outputs. Once created and addedto graphs, functions can be invoked by creating an operation whose operationtype matches the function name.
一组针对已定义输入输出的操作。创建并添加到图中之后，可以通过创建操作类型与函数名匹配的操作来调用函数
TFGraph
Representsa computation graph. Graphs may be shared between sessions and are thread safe.
代表了一种计算图。图可以在会话之间共享，并且是线程安全的。
TFImportGraphDefOptions
Containsoptions that are used to control how graph importing works.
包含用于控制图形导入工作方式的选项。
TFLibrary
Representsa dynamically loaded library of TensorFlow operations, use to load and consumeTensorFlow operations from an external library.
表示动态加载的TensorFlow操作库，用于从外部库加载和使用TensorFlow操作
TFOperation
Representsa computation node in the graph. Tensorflow operations are attached to a Tensorflow.TFGraph.
表示图中的计算节点。TFOperation作用于TFGraph。
TFOperationDesc
Low-levelTensorFlow operation builder
低级TensorFlow操作
TFScope
TFGraphname scope handle
TFGraph命名空间管理
TFSession
Drivesthe execution of a graph
驱动针对Graph的操作
TFSession.PartialRunToken
Tokenreturned from using one of the Partial Run Setup methods from TFSession, and use this token subsequently forother invocations.
调用TFSession中的一个局部的Run Setup方法的返回值
TFSession.Runner
Usethe runner class to easily configure inputs, outputs and targets to be passedto the session runner.
方便地配置需要传递给SessionRunner的输入输出和目标值
TFSessionOptions
Thesession options object holds configuration options that you want to use duringyour session, like the TensorFlow target or the configuration.
sessionoptions对象包含您希望在会话期间使用的配置选项，如TensorFlow目标或配置。
TFShape
Representsthe shape of a tensor, it describes how many dimensions the tensor has in agiven axis
表示一个张量的形状，它描述了张量在给定坐标轴中有多少维。
TFStatus
Usedto track the result of TensorFlow operations.
用于跟踪TensorFlow操作的结果。
TFTensor
TFTensorholds a multi-dimensional array of elements of a single data type.
保存一个单一的多维数组类型
Variable
TheVariable class holds the TFOutput nodes that are used to initialize, read andassign a value to a variable.
变量类包含用于初始化、读取和为变量赋值的TFOutput节点。
Structs
TFAttributeMetadata
Low-level:this describes the tensorflow type information for an attribute in thelow-level attributes used by operations.
低级:描述操作使用的低级属性中的属性的tensorflow类型信息。
TFInput
Representsa specific input of an operation.
表示一组operation的输入
TFOutput
Representsa specific output of an operation on a tensor.
表示作用在张量上的operation的一组输出
Enums
TFAttributeType
Low-level:Enumeration describing the types of a metadata attribute
描述元数据属性类型的枚举
TFCode
Statuscode for invoking a tensorflow operation.
用于激活operation的编码
TFDataType
Thedata type for a specific tensor.
特定张量的数据类型
Delegates
TFBuffer.BufferReleaseFunc
Signatureof the method that is invoked to release the data.
用于释放数据的方法的签名。
TFGraph.WhileConstructor
Signatureof the method that will be invoked by the TFGraph.While method to construct awhile loop
TFGraph将调用的方法的签名。建立While循环的方法
TFTensor.Deallocator
Signaturethat methods must conform to to be used to release memory that was passed to amanually allocated TFTensor
方法必须遵循的签名用于释放传递给手动分配的tf张量的内存