Namespace TensorFlow
.NET Bindings for TensorFlow.
This is the API documentation for the TensorFlowSharp project.
Classes
Adagrad
Adaptive stochastic gradient descent optimizer.
AdaptiveOptimizer
The base class for all the adaptive optimizers.
DeviceAttributes
Describes the device attributes
MonoPInvokeCallbackAttribute
This attribute can be applied to callback functions that will be invoked from unmanaged code to managed code.
Optimizer
Base class for all the optimizers.
PaddingFIFOQueue
A FIFOQueue that supports batching variable-sized tensors by padding. Port of Python implementation https://github.com/tensorflow/tensorflow/blob/b46340f40fe5e2ec9bfcd385b07cfb914055fb51/tensorflow/python/ops/data_flow_ops.py#L697
QueueBase
Base class for queue implementations. Port of Python implementation https://github.com/tensorflow/tensorflow/blob/r1.3/tensorflow/python/ops/data_flow_ops.py
RMSProp
RMSProp: Adaptive stochastic gradient descent optimizer.
SGD
Stochastic gradient descent optimizer. Includes support for momentum, learning rate decay, and Nesterov momentum
TFBuffer
Holds a block of data, suitable to pass, or retrieve from TensorFlow.
TFCore
Contains TensorFlow fundamental methods and utility functions.
TFDependencies
TFGraph variable dependencies handle.
TFDevice
Class to unset device name in the graph within using block.
TFDisposable
Base class for many TensorFlow data types that provides a common idiom to dispose and release resources associated with the native data types. Generally, you do not need to use this.
TFDisposableThreadSafe
ase class for many TensorFlow data types that provides a common idiom to dispose and release resources associated with the native data types and whose unmanaged resource disposing can be called from a background thread (the finalizer). Users do not need to deal with this class.
TFException
TensorFlow Exception
TFFunction
A grouping of operations with defined inputs and outputs. Once created and added to graphs, functions can be invoked by creating an operation whose operation type matches the function name.
TFGraph
Represents a computation graph. Graphs may be shared between sessions and are thread safe.
TFImportGraphDefOptions
Contains options that are used to control how graph importing works.
TFLibrary
Represents a dynamically loaded library of TensorFlow operations, use to load and consume TensorFlow operations from an external library.
TFOperation
Represents a computation node in the graph. Tensorflow operations are attached to a Tensorflow.TFGraph.
TFOperationDesc
Low-level TensorFlow operation builder
TFScope
TFGraph name scope handle
TFSession
Drives the execution of a graph
TFSession.PartialRunToken
Token returned from using one of the Partial Run Setup methods from TFSession, and use this token subsequently for other invocations.
TFSession.Runner
Use the runner class to easily configure inputs, outputs and targets to be passed to the session runner.
TFSessionOptions
The session options object holds configuration options that you want to use during your session, like the TensorFlow target or the configuration.
TFShape
Represents the shape of a tensor, it describes how many dimensions the tensor has in a given axis
TFStatus
Used to track the result of TensorFlow operations.
TFTensor
TFTensor holds a multi-dimensional array of elements of a single data type.
Variable
The Variable class holds the TFOutput nodes that are used to initialize, read and assign a value to a variable.
Structs
TFAttributeMetadata
Low-level: this describes the tensorflow type information for an attribute in the low-level attributes used by operations.
TFInput
Represents a specific input of an operation.
TFOutput
Represents a specific output of an operation on a tensor.
Enums
DeviceType
Device type
TFAttributeType
Low-level: Enumeration describing the types of a metadata attribute
TFCode
Status code for invoking a tensorflow operation.
TFDataType
The data type for a specific tensor.
Delegates
TFBuffer.BufferReleaseFunc
Signature of the method that is invoked to release the data.
TFGraph.WhileConstructor
Signature of the method that will be invoked by the TFGraph.While method to construct a while loop
TFTensor.Deallocator
Signature that methods must conform to to be used to release memory that was passed to a manually allocated TFTensor