[Idea]: add matrix format parsers and data loaders

[kgryte]

Idea

The goal of this project would be to implement various matrix and multi-dimensional format parsers and data loaders. E.g.,

• Matrix Market
• NumPy npy
• DLPack
• MATLAB mat
• others?

Implementing these parsers and loaders would facilitate array data interchange with other numerical computing ecosystems.

Expected outcomes

Users will be able to load multi-dimensional array data saved in other numerical computing environments into stdlib's ndarray data structure.

Status

No work has begun on this.

Involved software

Access to MATLAB/Octave would be useful for implementing the MAT-file parser. One would likely need to use Python and NumPy in order to save and work with npy files.

Technology

JavaScript

Other technology

None.

Difficulty

4

Difficulty justification

Some of the file format specifications can be quite involved. It is also likely that we may encounter situations in which we cannot support particular formats in full due to dtype incompatibility, etc.

Prerequisite knowledge

Familiarity with JavaScript, Python, and MATLAB would be useful. Experience writing parsers and performing IO will also be beneficial.

Project length

90/175/350. Can be scoped accordingly.

Checklist

• [x] I have read and understood the Code of Conduct.
• [x] I have read and understood the application materials found in this repository.
• [x] The issue name begins with [Idea]: and succinctly describes your idea.
• [x] I understand that, in order to apply to be a GSoC contributor, I must submit my final application to https://summerofcode.withgoogle.com/ before the submission deadline.

[kgryte]

Another format is the nearly raw raster data format (NRRD). Refs:

• https://teem.sourceforge.net/nrrd/
• https://github.com/scijs/nrrd-js