Package org.tribuo.interop.tensorflow

Class RegressorConverter

java.lang.Object
org.tribuo.interop.tensorflow.RegressorConverter
All Implemented Interfaces:
com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance>, Serializable, OutputConverter<Regressor>, ProtoSerializable<org.tribuo.interop.tensorflow.protos.OutputConverterProto>
public class RegressorConverter extends Object implements OutputConverter<Regressor>
Can convert a Regressor to a TFloat32 vector and a TFloat32 into a Prediction or Regressor.
See Also:

  • Field Details

    • CURRENT_VERSION

      public static final int CURRENT_VERSION
      Protobuf serialization version.
      See Also:

  • Constructor Details

    • RegressorConverter

      public RegressorConverter()
      Constructs a RegressorConverter.

  • Method Details

    • deserializeFromProto

      public static RegressorConverter deserializeFromProto(int version, String className, com.google.protobuf.Any message)
      Deserialization factory.
      Parameters:
      version - The serialized object version.
      className - The class name.
      message - The serialized data.
      Returns:
      The deserialized object.

    • serialize

      public org.tribuo.interop.tensorflow.protos.OutputConverterProto serialize()
      Description copied from interface: ProtoSerializable
      Serializes this object to a protobuf.
      Specified by:
      serialize in interface ProtoSerializable<org.tribuo.interop.tensorflow.protos.OutputConverterProto>
      Returns:
      The protobuf.

    • loss

      public BiFunction<org.tensorflow.op.Ops,com.oracle.labs.mlrg.olcut.util.Pair<org.tensorflow.op.core.Placeholder<? extends org.tensorflow.types.family.TNumber>,org.tensorflow.Operand<org.tensorflow.types.family.TNumber>>,org.tensorflow.Operand<org.tensorflow.types.family.TNumber>> loss()
      Returns a mean squared error loss.
      Specified by:
      loss in interface OutputConverter<Regressor>
      Returns:
      The mse loss.

    • outputTransformFunction

      public <U extends org.tensorflow.types.family.TNumber> BiFunction<org.tensorflow.op.Ops,org.tensorflow.Operand<U>,org.tensorflow.op.Op> outputTransformFunction()
      Applies the identity function
      Specified by:
      outputTransformFunction in interface OutputConverter<Regressor>
      Type Parameters:
      U - The input type (should be TFloat32).
      Returns:
      A function which applies the identity function.

    • convertToPrediction

      public Prediction<Regressor> convertToPrediction(org.tensorflow.Tensor tensor, ImmutableOutputInfo<Regressor> outputIDInfo, int numValidFeatures, Example<Regressor> example)
      Description copied from interface: OutputConverter
      Converts a Tensor into a Prediction.
      Specified by:
      convertToPrediction in interface OutputConverter<Regressor>
      Parameters:
      tensor - The tensor to convert.
      outputIDInfo - The output info to use to identify the outputs.
      numValidFeatures - The number of valid features used by the prediction.
      example - The example to insert into the prediction.
      Returns:
      A prediction object.

    • convertToOutput

      public Regressor convertToOutput(org.tensorflow.Tensor tensor, ImmutableOutputInfo<Regressor> outputIDInfo)
      Description copied from interface: OutputConverter
      Converts a Tensor into the specified output type.
      Specified by:
      convertToOutput in interface OutputConverter<Regressor>
      Parameters:
      tensor - The tensor to convert.
      outputIDInfo - The output info to use to identify the outputs.
      Returns:
      A output.

    • convertToBatchPrediction

      public List<Prediction<Regressor>> convertToBatchPrediction(org.tensorflow.Tensor tensor, ImmutableOutputInfo<Regressor> outputIDInfo, int[] numValidFeatures, List<Example<Regressor>> examples)
      Description copied from interface: OutputConverter
      Converts a Tensor containing multiple outputs into a list of Predictions.
      Specified by:
      convertToBatchPrediction in interface OutputConverter<Regressor>
      Parameters:
      tensor - The tensor to convert.
      outputIDInfo - The output info to use to identify the outputs.
      numValidFeatures - The number of valid features used by the prediction.
      examples - The example to insert into the prediction.
      Returns:
      A list of predictions.

    • convertToBatchOutput

      public List<Regressor> convertToBatchOutput(org.tensorflow.Tensor tensor, ImmutableOutputInfo<Regressor> outputIDInfo)
      Description copied from interface: OutputConverter
      Converts a Tensor containing multiple outputs into a list of Outputs.
      Specified by:
      convertToBatchOutput in interface OutputConverter<Regressor>
      Parameters:
      tensor - The tensor to convert.
      outputIDInfo - The output info to use to identify the outputs.
      Returns:
      A list of outputs.

    • convertToTensor

      public org.tensorflow.Tensor convertToTensor(Regressor example, ImmutableOutputInfo<Regressor> outputIDInfo)
      Description copied from interface: OutputConverter
      Converts an Output into a Tensor representing it's output.
      Specified by:
      convertToTensor in interface OutputConverter<Regressor>
      Parameters:
      example - The output to convert.
      outputIDInfo - The output info to use to identify the outputs.
      Returns:
      A Tensor representing this output.

    • convertToTensor

      public org.tensorflow.Tensor convertToTensor(List<Example<Regressor>> examples, ImmutableOutputInfo<Regressor> outputIDInfo)
      Description copied from interface: OutputConverter
      Converts a list of Example into a Tensor representing all the outputs in the list. It accepts a list of Example rather than a list of Output for efficiency reasons.
      Specified by:
      convertToTensor in interface OutputConverter<Regressor>
      Parameters:
      examples - The examples to convert.
      outputIDInfo - The output info to use to identify the outputs.
      Returns:
      A Tensor representing all the supplied Outputs.

    • generatesProbabilities

      public boolean generatesProbabilities()
      Description copied from interface: OutputConverter
      Does this OutputConverter generate probabilities.
      Specified by:
      generatesProbabilities in interface OutputConverter<Regressor>
      Returns:
      True if it produces a probability distribution in the Prediction.

    • toString

      public String toString()
      Overrides:
      toString in class Object

    • getProvenance

      public com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance getProvenance()
      Specified by:
      getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance>

    • getTypeWitness

      public Class<Regressor> getTypeWitness()
      Description copied from interface: OutputConverter
      The type witness used when deserializing the TensorFlow model from a protobuf.

      The default implementation throws UnsupportedOperationException for compatibility with implementations which don't use protobuf serialization. This implementation will be removed in the next major version of Tribuo.

      Specified by:
      getTypeWitness in interface OutputConverter<Regressor>
      Returns:
      The output class this object produces.