org.tribuo.interop.tensorflow (Tribuo 4.1.1 API)

Interface Summary
Interface	Description
FeatureConverter	Transforms an `Example` or `SGDVector`, extracting the features from it as a `TensorMap`.
OutputConverter<T extends Output<T>>	Converts the `Output` into a `Tensor` and vice versa.

Class Summary
Class	Description
DenseFeatureConverter	Converts a sparse example into a dense float vector, then wraps it in a `TFloat32`.
ImageConverter	Image converter.
LabelConverter	Can convert a `Label` into a `Tensor` containing one hot encoding of the label and can convert a `TFloat16` or `TFloat32` into a `Prediction` or a `Label`.
MultiLabelConverter	Can convert a `MultiLabel` into a `Tensor` containing a binary encoding of the label vector and can convert a `TFloat16` or `TFloat32` into a `Prediction` or a `MultiLabel`.
RegressorConverter	Can convert a `Regressor` to a `TFloat32` vector and a `TFloat32` into a `Prediction` or Regressor.
TensorFlowCheckpointModel<T extends Output<T>>	This model encapsulates a simple model with an input feed dict, and produces a single output tensor.
TensorFlowFrozenExternalModel<T extends Output<T>>	A Tribuo wrapper around a TensorFlow frozen model.
TensorFlowModel<T extends Output<T>>	Base class for a TensorFlow model that operates on `Example`s.
TensorFlowNativeModel<T extends Output<T>>	This model encapsulates a TensorFlow model running in graph mode with a single tensor output.
TensorFlowSavedModelExternalModel<T extends Output<T>>	A Tribuo wrapper around a TensorFlow saved model bundle.
TensorFlowTrainer<T extends Output<T>>	Trainer for TensorFlow.
TensorFlowTrainer.TensorFlowTrainerProvenance
TensorFlowUtil	Helper functions for working with TensorFlow.
TensorFlowUtil.TensorTuple	A serializable tuple containing the tensor class name, the shape and the data.
TensorMap	A map of names and tensors to feed into a session.
TrainTest	Build and run a Tensorflow multi-class classifier for a standard dataset.
TrainTest.TensorflowOptions	Options for training a model in TensorFlow.

Enum Summary
Enum	Description
GradientOptimiser	An enum for the gradient optimisers exposed by TensorFlow-Java.
TensorFlowTrainer.TFModelFormat	The model format to emit.
TrainTest.InputType	Type of feature extractor.

Package org.tribuo.interop.tensorflow Description

Provides an interface to TensorFlow, allowing the training of non-sequential models using any supported Tribuo output type. Sequential model support is found in org.tribuo.interop.tensorflow.sequence.

Tribuo's TensorFlow support operates in Graph mode, as in v0.3.1 that is the only way to access gradients. The set of supported gradients is determined by TensorFlow, and not all gradients are available in TensorFlow Java in v0.3.1. Unsupported gradients will trigger an exception when the train method is called.

Models can store their trained parameters in two ways, either inside the Tribuo serialized model file (using TensorFlowTrainer.TFModelFormat.TRIBUO_NATIVE) or as a TensorFlow checkpoint folder on disk (using TensorFlowTrainer.TFModelFormat.CHECKPOINT). The choice is made at training time, as they result in slightly different TF graph structures.

Similarly there are two supported kinds of ExternalModel for TensorFlow, TensorFlowSavedModelExternalModel which loads a SavedModelBundle and always reads from that path, and TensorFlowFrozenExternalModel which loads a TensorFlow v1 frozen graph and stores the graph inside the Tribuo serialized object.

There are two main interfaces for interacting with TensorFlow Tensors, FeatureConverter and OutputConverter, with provide conversions to and from Tribuo's features and outputs respectively. There are implementations of a dense feature transformation and one for images as 3d arrays, along with output converters for Label Regressor, and MultiLabel. The loss function and output transformation used is controlled by the OutputConverter, if a different one is desired then users are recommended to implement that interface separately.

N.B. TensorFlow support is experimental and may change without a major version bump.