dsl
Attribute class stores a named value.
Attribute class stores a named value.
name of the value
stored value of the attribute
Class that holds a list of Attribute objects.
Root object for serializing Bundle.ML pipelines and graphs.
Root object for serializing Bundle.ML pipelines and graphs.
name of the bundle
serialization format of the Bundle
Bundle.ML version used for serializing
optional AttributeList to serialize with the bundle
list of root nodes in the bundle
This trait provides easy access to reading/writing attributes to objects that contain an AttributeList.
This trait provides easy access to reading/writing attributes to objects that contain an AttributeList.
class type that contains the attribute list
Class that encodes all information need to serialize or deserialize a machine learning model.
Class that encodes all information need to serialize or deserialize a machine learning model.
Models encode things like coefficients for linear regressions, labels for string indexers, sizes for one hot encoders, decision trees, and any other data needed to serialize and deserialize ML models and feature builders.
Usually you will want to pass ReadableModel or WritableModel to a user depending on the use case to constrict the allowed operations.
op name for the model
optional list of attributes for the model
Class for storing a node in the Bundle.ML graph.
Class for storing a node in the Bundle.ML graph.
Bundle.ML is composed of a set of Node objects, each with their own inputs and outputs defined by their Node#shape.
Every Node needs a unique identifier within a Bundle.ML graph. This is the Node#name.
unique identifier for the node
shape of the node
Trait for read-only operations of an AttributeList.
Trait for read-only operations of an AttributeList.
Use this trait when you only want to grant read-access to an AttributeList. This is used when deserializing Model or Bundle objects.
Trait for read-only operations on a Model.
Trait for read-only interface to a Node.
Trait for read-only interface to a Node.
Use this trait when deserializing node objects.
Trait for read-only interface to Shape objects.
Class for holding the input fields and output fields of a Node.
Class for holding the input fields and output fields of a Node. The shape also holds information for connecting the input/output fields to the underlying ML model.
A Shape contains input and output sockets. Sockets map field data to certain functionality within a Model. For instance, say we want to run a "label" field through a string indexer and have the result output to the field "label_name". We could wire up the node like so:
scala> import ml.bundle.dsl._ scala> Shape().withInput("label", "input"). // connect the "label" field to the model input withOutput("label_name", "output") // connect the model output to the "label_name" field
Or more concisely:
scala> import ml.bundle.dsl._ scala> Shape().withStandardIO("label", "label_name") // shorthand for the above code
protobuf shape object containing the shape information
This class is used to wrap Scala objects for later serialization into Bundle.ML
This class is used to wrap Scala objects for later serialization into Bundle.ML
data type of the value being stored
Scala object that will be serialized later
Trait for write operations on an AttributeList.
Trait for write operations on an AttributeList.
Use this trait when you need to provide a writable interface to the AttributeList. This is mainly used when serializing Model or Bundle objects.
Trait for a writable interface to a Model.
Trait for writable interface to a Node.
Trait for writable interface to a Shape.
Companion object for Attribute class.
Companion class for construction and conversion of AttributeList objects.
Companion class for constants and constructors of Bundle objects.
Companion class for constants and constructors of Bundle objects.
Contains file names for bundle JSON files and model JSON files.
Companion object for holding constant values.
Provides a set of helper methods for easily creating ml.combust.bundle.dsl.Value objects.
Provides a set of helper methods for easily creating ml.combust.bundle.dsl.Value objects.
Easily create ml.combust.bundle.dsl.Value objects of any type using the helper methods provided here. The helper methods will wrap a Scala value for later serializing into Bundle.ML.
Also provides several helper methods for converting from Bundle.ML protobuf objects back into wrapped Scala objects.
Domain specific language for generating Bundle.ML models.
DSL provides objects for generating various Bundle.ML objects.
ml.combust.bundle.dsl.Bundle - Stores the name, version, serialization format, list of attributes, and list of root nodes in the bundle graph.
ml.combust.bundle.dsl.Node - A node for transforming data in the model graph. Nodes have a unique name within a Bundle.ML graph and are responsible for piping inputs to a model and outputs from a model to the rest of the graph via sockets. Sockets map a field name in the graph to a specific port in a model.
A model can have input ports and output ports. An example would be a StringIndexer node named "my_string_indexer" that indexes the input field "my_string_label" to the output field "my_double_label". There are two sockets for this node, one that maps the "my_string_label" field to the "input" port of the model and another socket that maps the "output" port of the model to the "my_double_label" field.
ml.combust.bundle.dsl.Model - A model operation that transforms data within the graph. Models have an op name that define the type of model. For instance, the op name could be "string_indexer", "linear_regression", or "random_forest_regression". In addition to the op name of the model, all of the required attributes, such as coefficients and intercepts ,for executing the model are stored here.
ml.combust.bundle.dsl.Attribute - Attributes are used to store values for ml.combust.bundle.dsl.Models and ml.combust.bundle.dsl.Bundles. They contain a name, a ml.bundle.DataType.DataType and a ml.combust.bundle.dsl.Value.
ml.combust.bundle.dsl.Value - Values store concrete data for an ml.combust.bundle.dsl.Attribute. You can store basic data like strings, doubles, longs and booleans, tensors of any dimension, custom data types, and lists of any depth and data type.
ml.combust.bundle.dsl.AttributeList - Stores a set of ml.combust.bundle.dsl.Attributes.
ml.combust.bundle.dsl.Shape - Stores the map of graph input/output fields to model inputs and outputs.