Glossary

All of these new words can get confusing! We’ve created a glossary to help nail down some Ivy terms that you might find tricky.

Applied Libraries: Suite of various machine learning libraries that have been built using the Ivy framework.
Automatic Code Conversions: Allows code to be converted from framework to another whilst retaining its funtional assets.
Compositional Functions: Compositional functiona are functions that are implemented as a composition of other Ivy functions,
Convenience Functions: Convenience functions can be used to organize and improve the code for other functions.
Framework: Frameworks are interfaces that allow scientists and developers to build and deploy machine learning models faster and easier. E.g. Tensorflow and PyTorch.
Framework Handler
Backend Handler: Used to control which framework Ivy is converting code to.
Graph Compiler: Graph compilers map the high-level computational graph coming from frameworks to operations that are executable on a specific device.
Ivy Array: The ivy.Array is a simple wrapper class, which wraps around the ivy.NativeArray.
Ivy Backends: Ivy Backends are supported frameworks that Ivy can convert code to. The default is NumPy.
Ivy Builder: Helpful classes and functions for creating training workflows.
Ivy Compiler: A wrapper function around native compiler functions, which uses lower level compilers such as XLA to compile to lower level languages such as C++, CUDA, TorchScript etc.
Ivy Container: An Ivy class which inherits from dict allows for storing nested data.
Ivy Frontends: Ivy Frontends are supported frameworks that Ivy can convert code from.
Ivy Functional API: Is used for defining complex models, the Ivy functional API does not implement its own backend but wraps around other frameworks functional APIs and brings them into alignment.
Ivy Graph Compiler: Ivy’s Graph Compiler, which traces the graph as a composition of functions in the functional API in Python.
Ivy Transpiler: The transpiler allows framework to framework code conversions for supported frameworks.
Mixed Functions: Mixed functions are functions that have some backend-specific implementations but not for all backends.
Native Array: The ivy.NativeArray is simply a placeholder class for a backend-specific array class, such as np.ndarray, tf.Tensor or torch.Tensor.
Nestable Functions: Nestable functions are functions which can accept ivy.Container instances in place of any of the arguments.
Pipeline: A pipeline is a means of automating the machine learning workflow by enabling data to be transformed and correlated into a model that can then be analyzed to achieve outputs.
Primary Functions: Primary functions are the lowest level building blocks in Ivy and are generally implemented as light wrapping around an existing function in the backend framework, which serves a near-identical purpose.
Standalone Functions: Standalone functions are functions which do not reference any other primary, compositional or mixed functions whatsoever. These are mainly convenience functions.
Submodule Helper Functions: These are standalone/convenience functions that are specific to a submodule.