Fastrank 0.8 (M1 support, Python 3.7)

We revamped how CI scripts work and upgraded Maturin in order to get "Apple Silicon" support.

What's Changed

• Update zstd requirement from 0.4 to 0.9 by @dependabot-preview in #45
• Upgrade to GitHub-native Dependabot by @dependabot-preview in #42
• Update zstd requirement from 0.9 to 0.10 by @dependabot in #48
• Update ordered-float requirement from 2.0 to 3.0 by @dependabot in #50
• Update zstd requirement from 0.10 to 0.11 by @dependabot in #49
• Update zstd requirement from 0.11 to 0.12 by @dependabot in #51
• Try CI.yml modified from cramjam project by @jjfiv in #54

Full Changelog: 0.7.0...0.8.0

The 'predict_scores' release

• CModel now has predict_scores that returns a sparse representation of Dict[int, float] where the position in the arrays you've loaded correspond to the score.
• We also have predict_dense_scores which returns a List[float] with the same semantics. If you have subsampled queries, this may make less sense than the aforementioned method.
• We have some better testing covering these features.
• Note: trying again because of a glitch in automatic releases ('fix' for #32 insufficient)

0.6.1: move to manylinux2010

Rather than require the arbitrary libc etc. from the gh-actions publish, align with manylinux2010. Fixed some documentation nits along the way.

0.6.0: Windows, fast-float and python3.6

Updates in 0.6.0 are:

• Minimum python version now 3.6 -- 3.5 started failing on CI, so it's gone now.
• support for faster float parsing -- on my machine the msn30k dataset took 90s to load, and now only takes 60s. Thanks rust libraries!
• Windows supported - now that I have regular access to a windows machine, I will make sure PyPI has windows builds.

First public alpha: 0.4

FastRank

FastRank is an implementation of CoordinateAscent[1] from Ranklib that you can pip install; written in Rust and uses threads for efficiency; it will scale much better than the Java version to large datasets and many features.

It also has RandomForests, and someday LambdaMART (others depending on interest). I've been thinking a lot about what the limits of coordinate ascent are (linearity), and will probably play with that in future versions.

This is ready for production use in the sense that I used it for my TREC submission this year. The python API could use some thoughts and experiences (post issues on Github).

pip install fastrank

• Blog Post
• Jupyter Notebook Demo

[1] Metzler, D., & Croft, W. B. (2007). Linear feature-based models for information retrieval. Information Retrieval, 10(3), 257-274.