rendarray 0.2.0-alpha.2

An N-dimensional array for elements of arbitrary type. Lightweight array views and slicing. The Array is copy-on-write and similar to numpy’s ndarray.
rendarray-0.2.0-alpha.2 has been yanked.

The ndarray crate provides an n-dimensional container similar to numpy's ndarray.

  • ArrayBase The n-dimensional array type itself, parameterized by data storage.
  • Array Array where the data is reference counted and copy on write, it can act as both an owner as the data as well as a lightweight view.
  • OwnedArray Array where the data is owned uniquely.
  • ArrayView A lightweight array view.
  • ArrayViewMut A lightweight read-write array view.

Crate Summary and Status

  • Implements the numpy striding and broadcasting scheme for n-dimensional arrays
  • Array is clone on write, so it can be both a view or an owner of the data. OwnedArray is a uniquely owned array.
  • Focus is on being a generic n-dimensional container
  • Due to iterators, arithmetic operations, matrix multiplication etc are not very well optimized, this is not a serious crate for numerics or linear algebra.
  • There is no integration with linear algebra packages (at least not yet).

Crate feature flags

  • assign_ops
  • Optional, requires nightly
  • Enables the compound assignment operators
  • rustc-serialize
  • Optional, stable
  • Enables serialization support