9 releases (5 breaking)
0.6.2 | Nov 8, 2019 |
---|---|
0.6.1 | Jun 21, 2019 |
0.5.0 | Apr 25, 2019 |
0.4.1 | Apr 12, 2019 |
0.1.0 | Sep 10, 2018 |
#5 in #finalfusion
32 downloads per month
215KB
5.5K
SLoC
finalfrontier
Introduction
finalfrontier is a Rust program for training word embeddings. finalfrontier currently has the following features:
- Models:
- skip-gram (Mikolov et al., 2013)
- structured skip-gram (Ling et al., 2015)
- directional skip-gram (Song et al., 2018)
- dependency (Levy and Goldberg, 2014)
- Output formats:
- finalfusion
- fastText
- word2vec binary
- word2vec text
- GloVe text
- Noise contrastive estimation (Gutmann and Hyvärinen, 2012)
- Subword representations (Bojanowski et al., 2016)
- Hogwild SGD (Recht et al., 2011)
- Quantized embeddings through the
finalfusion quantize
command.
The trained embeddings can be stored in the versatile finalfusion
format, which can be read and used with the
finalfusion crate
and the
finalfusion
Python module.
The minimum required Rust version is currently 1.70.
Where to go from here
- Installation
- Quickstart
- Manual pages:
- finalfrontier-skipgram(1) — train word embeddings with the (structured) skip-gram model
- finalfrontier-deps(1) — train word embeddings with dependency contexts
- finalfusion crate
- Python module
Dependencies
~9–19MB
~242K SLoC