Deprecated API

Contents

• Deprecated Methods
• Deprecated Constructors

Deprecated Methods

Method and Description
com.linkedin.feathr.common.TaggedFeatureUtils.applyStringTags(ErasedEntityTaggedFeature, List<String>)
com.linkedin.feathr.common.FeatureValue.createCategorical(Number)
com.linkedin.feathr.common.configObj.generation.OfflineOperationalConfig.get_enableIncremental()
com.linkedin.feathr.common.configObj.generation.OfflineOperationalConfig.get_retention()
com.linkedin.feathr.common.configObj.generation.OfflineOperationalConfig.get_simulateTimeDelay()
com.linkedin.feathr.common.configObj.generation.OfflineOperationalConfig.get_timeSetting()
com.linkedin.feathr.common.FeatureValue.getAsString() use FeatureValue.getAsCategorical() instead
com.linkedin.feathr.common.FeatureValue.getNumTerms()
com.linkedin.feathr.common.FeatureDependencyGraph.getOrderedPlanForRequest(Collection<TaggedFeatureName>)
com.linkedin.feathr.common.FeatureValue.getValue() Prefer to use FeatureValue.getAsTermVector() which has a more descriptive name
com.linkedin.feathr.common.tensor.TensorIterator.getValue(int) this should not be used
com.linkedin.feathr.common.tensor.DimensionType.indexToString(long) Use DimensionType.getDimensionValue(ReadableTuple, int) instead
com.linkedin.feathr.common.FeatureDependencyGraph.isReachable(String)
com.linkedin.feathr.common.TensorUtils.parseTensorType(String) use TensorTypes.parseTensorType(java.lang.String)
com.linkedin.feathr.common.FeatureValue.put(String, Float) mutability in FeatureValue is deprecated; see class-level Javadoc
com.linkedin.feathr.common.PegasusFeatureTypeResolver.resolveEmbeddingSize(FeatureVersion)
com.linkedin.feathr.common.FeatureValue.size() Prefer to use FeatureValue.getNumTerms() which has a more descriptive name
com.linkedin.feathr.common.tensor.DimensionType.stringToIndex(String) Use DimensionType.setDimensionValue(WriteableTuple, int, Object) instead

Deprecated Constructors

Constructor and Description
com.linkedin.feathr.common.FeatureValue() Use the static factory methods such as FeatureValue.createNumeric(Number) to construct features instead
com.linkedin.feathr.common.FeatureValue(Map<String, Float>) Use the static factory methods such as FeatureValue.createStringTermVector(Map) to construct features instead
com.linkedin.feathr.common.FeatureValue(Object) Use the static factory methods such as FeatureValue.createNumeric(Number) to construct features instead N.B. Although this takes Object, there are runtime type checks to ensure the passed type can be converted to a feature value.
com.linkedin.feathr.common.FeatureValue(Object, FeatureTypes) Use the static factory methods such as FeatureValue.createNumeric(Number) to construct features instead
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int[], long[], float[], Representable[]) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int[], long[], String[], double[], Representable[]) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int[], long[], String[], float[], double[], ColumnUtils) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int[], long[], String[], float[], double[], Object[], ColumnUtils) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int[], long[], String[], float[], double[], Representable[]) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int[], long[], String[], float[], Representable[]) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int, int, int, int[], long[], float[], Representable[], boolean, int...) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int, int, int, int[], long[], float[], Representable[], int...) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int, int, int, int[], long[], float[], Representable[], int, boolean, int...) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int, int, int, int, int[], long[], String[], float[], Representable[], boolean, int...) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int, int, int, int, int[], long[], String[], float[], Representable[], int...) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int, int, int, int, int, int[], long[], String[], float[], double[], Representable[], boolean, int...) Use UniversalTensorBuilder
com.linkedin.feathr.common.tensorbuilder.UniversalTensor(int, int, int, int, int, int[], long[], String[], float[], double[], Representable[], int, boolean, int...) Use UniversalTensorBuilder