OMORFI-FACTORISE(1)

💖 OMORFI-FACTORISE(1) 💖

okay bestie, welcome to the User Commands manual page for omorfi-factorise — yes, the whole vibe is Finnish running text getting factorised for moses SMT. slay 🦋

✨ NAME ✨

omorfi-factorise — Factorise running Finnish text for moses SMT 💅

🎀 SYNOPSIS 🎀

omorfi-factorise.bash [OPTION] [FILENAME...]

🌸 DESCRIPTION 🌸

Preprocess Finnish running text for factored statistical machine translation. That's the main-character job, no cap.

Here's the option menu, bestie:

• -h, --help — Print this help dialog 🫶
• -V, --version — Print version info ⭐
• -v, --verbose — Print verbosely while processing 👀
• -X, --all-words — Use large-coverage dictionary 🔥

If no FILENAMEs are given, input is read from standard input. iykyk periodt 💖

💀 BUGS 💀

Uses 1-random disambiguation separately for each factor(!). Does not improve your BLEU score.

this is the tea: that (!) is literally right there in the original, and the BLEU warning is not a drill — do not expect this to magically boost your score. 😭

🧚 EXAMPLE 🧚

Factorise running text:

omorfi-factorise.bash test.text

👑 COPYRIGHT 👑

Copyright © 2018 Tommi A Pirinen Licence GPLv3: GNU GPL version 3 <http://gnu.org/licenses/gpl.html>

This is free software: you are free to change and redistribute it.

There is NO WARRANTY, to the extent permitted by law.

🌈 SEE ALSO 🌈

omorfi-tokenise(1) — go give it a look, bestie 💜

---

OMORFI • June 2018 — manual source/version and date, signed off with love 🎀💖✨