ML Example constructs an example from an input component and an optional target component.

ML Example Decompose does the opposite; it takes an example and decomposes it into its input component and its optional target component. ML Example Decompose is useful for doing pre-processing passes on sets of examples. It also provides a way to inspect the input and target components of examples for troubleshooting and visualization purposes.

ML Example Decompose isn’t limited to processing a single example at a time. If both Keep Input Component Packed and Keep Target Component Packed are on, then it can decompose in its order a set of examples into a set of input components and a set of target components. However, if either Keep Input Component Packed and Keep Target Component Packed is disabled, ML Example Decompose expects the Example input to consist of only a single example and each output will have only a single component.

See Machine Learning documentation for more general information.