etch mathematics and machine learning operations are supported by a header-only fully-templated C++ library.
Detailed developer documentation for the C++ implementation of the maths libraries will be available in the Fetch Ledger section in due course. Developers should be comfortable with SFINAE.
A core component of the maths library is the
tensor.hpp class which handles N-dimensional array mathematics. This is crucial for the machine learning library but can also be used for any generalised matrix algebra.
The remainder of the library contains templated free functions that can be called with the following types:
- Built-in C++ types such as
- C++ tensors of built-in types such as
etchtypes such as
etchtensors types such as
The header file
fundamental_operators.hpp contains common operations
standard_functions directory contains header files for additional standard operations.
The following block diagram gives a rough indication of the library structure.
This is work in progress.
Tensor is a wrapper for a
data_ object which is, by default, a
SharedArray managed by the
vectorise library. This library manages the vectorisation/SIMD on the underlying data.
Tensor objects provide interfaces for manipulating arrays at a mathematical/algebraic level while allowing implementations to be efficient and vectorisable.
Tensors have related helper classes such as
TensorSlice that permit efficient and convenient manipulation such as accessing, transposing, and slicing.
Check out the available mathematical functions in
Working with the maths library
When working with the C++ maths library, take note of the following:
Functions should have two interfaces: one that takes a reference to the return object, and one that creates the return object internally.
Function design decisions should follow Numpy conventions where possible.