Uses of Interface
org.tribuo.interop.tensorflow.FeatureConverter
Package
Description
Provides an interface to TensorFlow, allowing the training of non-sequential models using any supported
Tribuo output type.
-
Uses of FeatureConverter in org.tribuo.interop.tensorflow
Modifier and TypeClassDescriptionclass
Converts a sparse example into a dense float vector, then wraps it in aTFloat32
.class
Image converter.Modifier and TypeFieldDescriptionprotected final FeatureConverter
TensorFlowModel.featureConverter
Modifier and TypeMethodDescriptionstatic <T extends Output<T>>
TensorFlowFrozenExternalModel<T>TensorFlowFrozenExternalModel.createTensorflowModel
(OutputFactory<T> factory, Map<String, Integer> featureMapping, Map<T, Integer> outputMapping, String outputName, FeatureConverter featureConverter, OutputConverter<T> outputConverter, String filename) Creates a TensorflowFrozenExternalModel by loading in a frozen graph.static <T extends Output<T>>
TensorFlowSavedModelExternalModel<T>TensorFlowSavedModelExternalModel.createTensorflowModel
(OutputFactory<T> factory, Map<String, Integer> featureMapping, Map<T, Integer> outputMapping, String outputName, FeatureConverter featureConverter, OutputConverter<T> outputConverter, String bundleDirectory) Creates a TensorflowSavedModelExternalModel by loading in aSavedModelBundle
.ModifierConstructorDescriptionprotected
TensorFlowModel
(String name, ModelProvenance provenance, ImmutableFeatureMap featureIDMap, ImmutableOutputInfo<T> outputIDInfo, org.tensorflow.proto.framework.GraphDef trainedGraphDef, int batchSize, String outputName, FeatureConverter featureConverter, OutputConverter<T> outputConverter) Builds a TFModel.TensorFlowTrainer
(Path graphPath, String outputName, GradientOptimiser optimiser, Map<String, Float> gradientParams, FeatureConverter featureConverter, OutputConverter<T> outputConverter, int trainBatchSize, int epochs, int testBatchSize, int loggingInterval) Constructs a Trainer for a TensorFlow graph.TensorFlowTrainer
(Path graphPath, String outputName, GradientOptimiser optimiser, Map<String, Float> gradientParams, FeatureConverter featureConverter, OutputConverter<T> outputConverter, int trainBatchSize, int epochs, int testBatchSize, int loggingInterval, Path checkpointPath) Constructs a Trainer for a TensorFlow graph.TensorFlowTrainer
(org.tensorflow.Graph graph, String outputName, GradientOptimiser optimiser, Map<String, Float> gradientParams, FeatureConverter featureConverter, OutputConverter<T> outputConverter, int trainBatchSize, int epochs, int testBatchSize, int loggingInterval) Constructs a Trainer for a TensorFlow graph.TensorFlowTrainer
(org.tensorflow.Graph graph, String outputName, GradientOptimiser optimiser, Map<String, Float> gradientParams, FeatureConverter featureConverter, OutputConverter<T> outputConverter, int trainBatchSize, int epochs, int testBatchSize, int loggingInterval, Path checkpointPath) Constructs a Trainer for a TensorFlow graph.TensorFlowTrainer
(org.tensorflow.proto.framework.GraphDef graphDef, String outputName, GradientOptimiser optimiser, Map<String, Float> gradientParams, FeatureConverter featureConverter, OutputConverter<T> outputConverter, int trainBatchSize, int epochs, int testBatchSize, int loggingInterval) Constructs a Trainer for a TensorFlow graph.TensorFlowTrainer
(org.tensorflow.proto.framework.GraphDef graphDef, String outputName, GradientOptimiser optimiser, Map<String, Float> gradientParams, FeatureConverter featureConverter, OutputConverter<T> outputConverter, int trainBatchSize, int epochs, int testBatchSize, int loggingInterval, Path checkpointPath) Constructs a Trainer for a TensorFlow graph.