All Classes and Interfaces
Class
Description
Abstract base class for FeatureFormatMapper.
Mappings from Feathr FeatureTypes to their "auto-tensorized" tensor type.
The base for custom iterators over dense tensors of ranks 1+.
A FeatureType class for Feathr's BOOLEAN feature type.
A specific FeatureValue class for BOOLEAN features.
An interface for building Tensors from arrays and collections of values
This is targeted towards building DenseTensor.
Base class for dense int/float/double/long/boolean/string/byte[] tensors backed by a ByteBuffer using TensorFlow-compatible layout.
A FeatureType class for Feathr's CATEGORICAL feature type.
A specific FeatureValue class for CATEGORICAL features.
A FeatureType class for Feathr's CATEGORICAL_SET feature type.
A specific FeatureValue class for CATEGORICAL_SET features.
Coerce types in any TensorData to the one requested by the client.
Utilities to coerce untyped data into feature vectors
This is needed in order to make it easy to define features via simple expressions.
Utility function to build internal flat array representation based on the types of the input.
When an error is encountered during config processing, this exception is thrown
Marker interface for all config objects used in Feathr
Utils to read typesafe configs
Represent a time period or a time point.
Build DateTimeConfig from config
Dense tensor which can be converted from/to a tensorflow tensor object.
Dense tensor which can be converted from/to a TensorFlow tensor object.
Dense tensor which can be converted from/to a tensorflow tensor object.
Dense tensor which can be converted from/to a tensorflow tensor object.
Dense tensor which can be converted from/to a tensorflow tensor object.
Dense tensor which can be converted from/to a tensorflow tensor object.
Dense tensor which can be converted from/to a TensorFlow tensor object.
A dense subkind of tensor data.
Build a dense tensor based on input column types and shape,
All columns except the last one, must have
Primitive.INT representation.A list that is backed by a dense tensor.
A FeatureType class for Feathr's DENSE_VECTOR feature type.
A specific FeatureValue class for DENSE_VECTOR features.
Base type for all the dimension types.
Compare whether two tensor data are equal.
A
TaggedFeatureName whose String tags have been "erased", i.e.Error label that is used in exception message.
Marks the target of this annotation as an experimental feature that may evolve in future, or may be removed
entirely.
This exception is thrown when the feature definition is incorrect.
This exception is thrown when the data output cannot be written successfully.
Base exception for Feathr
This exception is thrown when the feature definition is invalid.
This exception is thrown when the input data can't be read or is invalid.
Feature aggregation types
A dependency graph for feature anchors and feature derivations.
A converter for a single feature from Spark value to a TensorData of a specific
TensorType.An error associated with a particular feature containing an error msg and the corresponding
FeatureErrorCodeError code associated with a feature request
The entry point for Py4j to access the feature experiment component in Java world.
A feature extractor that extracts
FeatureValue from source data.A mapper, or translator, that can convert between the standard FeatureValue representation and some other
external representation.
Define the feature generation specification, i.e., list of features to generate and other settings.
Feature generation config builder
Represents a fully-qualified reference to a feature.
Top level interface for Feature Types in Feathr.
Enum for the top-level feature types supported by Feathr.
This class encapsulates the type definition for a feature.
Config deserializer for FeatureType config.
An Enum that defines the supported feature types in feathr.
Describes the basic representation of a feature value in Feathr.
Represents a value of a feature in Feathr.
Protobuf type
protobuf.BooleanArrayProtobuf type
protobuf.BooleanArrayProtobuf type
protobuf.BytesArrayProtobuf type
protobuf.BytesArrayProtobuf type
protobuf.DoubleArrayProtobuf type
protobuf.DoubleArrayProtobuf type
protobuf.FeatureValueProtobuf type
protobuf.FeatureValueProtobuf type
protobuf.FloatArrayProtobuf type
protobuf.FloatArrayProtobuf type
protobuf.IntegerArrayProtobuf type
protobuf.IntegerArrayProtobuf type
protobuf.LongArrayProtobuf type
protobuf.LongArrayProtobuf type
protobuf.SparseBoolArrayProtobuf type
protobuf.SparseBoolArrayProtobuf type
protobuf.SparseDoubleArrayProtobuf type
protobuf.SparseDoubleArrayProtobuf type
protobuf.SparseFloatArrayProtobuf type
protobuf.SparseFloatArrayProtobuf type
protobuf.SparseIntegerArrayProtobuf type
protobuf.SparseIntegerArrayProtobuf type
protobuf.SparseLongArrayProtobuf type
protobuf.SparseLongArrayProtobuf type
protobuf.SparseStringArrayProtobuf type
protobuf.SparseStringArrayProtobuf type
protobuf.StringArrayProtobuf type
protobuf.StringArrayUtility functions for constructing FeatureValue instances.
A converter that knows how to convert back and forth between Feathr's
FeatureValue and some other
representation.FeatureVariableResolver takes a FeatureValue object for member variable during MVEL expression evaluation,
and then resolve the value for that variable.
This is the base tensor class, may be created from the basic properties, from a feature or operator applied
to another tensor
Generate hash for a tensor based on column types and values stored in the tensor.
An annotation indicating that the target is part of a module-private "internal API" and SHOULD NOT be used by
external modules.
Utility methods for working with FeaturizedDatasetMetadata.
Each list is of the same length and stores a single dimension (similarly to parallel arrays).
Some MVEL hackery to enable use in Feathr.
Allows easy access to the properties of GenericRecord object from MVEL.
MVEL is an open-source expression language and runtime that makes it easy to write concise statements that operate
on structured data objects (such as Avro records), among other things.
A FeatureFormatMapper that converts in and out of NTV representation, using Feathr's legacy rules for representing
types like NUMERIC, BOOLEAN, and DENSE_VECTOR as term vectors.
A FeatureType class for Feathr's NUMERIC feature type.
A specific FeatureValue class for NUMERIC features.
Operational section in feature generation config
Feature generation config contains two major sections, i.e., operational and feature list sections,
feature list specify the features to generate,
operational section contains all the related settings.
Operational section in feature generation config
This abstract class is extended by offline Operational Config.
Operation config object builder
Output processor config object builder, e.g., HDFS, Redis processor
Output processor config, e.g., write to HDFS processor or push to Redis processor
This extension of the FeatureExtractor rely on extra parameters to extract features.
This class extracts default
FeatureValue from pegasus modelsThis class maps from the pegasus models for feature types to Frame's common domain models for feature types and vice
versa.
Supported primitive types.
The most basic type of a dimension.
Type definitions for the supported primitives.
A FeatureFormatMapper that can represent any FeatureValue as a Quince TensorData.
A mapper, or translator, that provides a Quince TensorType for a given FeatureType.
A readable tuple of primitives.
A scalar with a primitive representation - either a dimension or a value.
A base for all wrappers of scalars.
Internal utilities for handling FDS metadata in Avro schema.
An implementation of
WriteableTuple which keeps all values in primitive arrays.A converter from FDS Spark objects to TensorData.
A readable tuple implementation that stores single tensor row data independent of
a tensor aka in standalone manner.
A tuple of (key tag, feature name)
Essentially this is a feature name annotated with info on how we plan to query it.
A stateful object used to build instances of tensors of some specific representation without knowing its details.
The category of the tensor type, such as dense, sparse, etc.
A typed container of structured data.
A FeatureType class for Feathr's TENSOR feature type.
A FeatureValue that contains an arbitrary tensor.
TensorIterator is responsible for a range of rows (possibly empty).
Utility methods for converting various collections to tensors.
Type definition for a TypedTensor, defines the value type, dimension types and dimension names.
Utility functions for working with
TensorTypeINTERNAL debugging utility functions.
A FeatureType class for Feathr's TERM_VECTOR feature type.
A specific FeatureValue class for TERM_VECTOR features.
TypedTensor builds on top of @link{TensorType} and @{TensorData} and can be generated from tensor
operators that change the shape.
Implements a tensor from pre-allocated arrays.
Builds tensors with columns being a requested permutation of int, long, and float.
A writable tuple of primitives.