2020-01-26 13:19:43 +00:00
|
|
|
---
|
|
|
|
|
title: Spacy
|
2020-10-01 14:05:57 +00:00
|
|
|
published: true
|
|
|
|
|
taxonomy:
|
|
|
|
|
category: [project]
|
|
|
|
|
tag: [spacy,nn,pos,ner,annotation,nlp]
|
|
|
|
|
author: Daniel Hladek
|
2020-01-26 13:19:43 +00:00
|
|
|
---
|
2019-12-07 11:58:50 +00:00
|
|
|
# Podpora slovenčiny v knižnici Spacy
|
2020-04-14 14:48:36 +00:00
|
|
|
[Trénovanie modelov Spacy](https://spacy.io/usage/training)
|
2019-12-07 11:58:50 +00:00
|
|
|
|
|
|
|
|
## Úlohy
|
|
|
|
|
|
|
|
|
|
- Vytvoriť modul v jazyku Python pre podporu slovenčiny
|
2020-01-25 16:26:46 +00:00
|
|
|
- Pripraviť trénovacie dáta pomocu frameworku [Prodigy](/topics/prodigy).
|
2019-12-07 11:58:50 +00:00
|
|
|
- Natrénovať štatistické modely:
|
|
|
|
|
- pre morfologickú analýzu
|
2019-12-07 12:06:01 +00:00
|
|
|
- rozpoznávanie pomenovaných entít
|
|
|
|
|
|
2019-12-10 10:04:54 +00:00
|
|
|
## Repozitár
|
|
|
|
|
|
2020-04-14 14:46:05 +00:00
|
|
|
- https://git.kemt.fei.tuke.sk/jo673fv/sk-spacy (slovenské stop words a pravidlá)
|
2020-04-14 14:48:36 +00:00
|
|
|
- https://git.kemt.fei.tuke.sk/dano/spacy-skmodel (slovenské NER a POS, vychádza s oficiálneho modelu, obsahuje niektoré dátové zdroje)
|
2020-04-14 14:46:05 +00:00
|
|
|
- https://files.kemt.fei.tuke.sk/minio/models/spacy/ Hotové spacy modely
|
2019-12-10 10:04:54 +00:00
|
|
|
|
2019-12-07 12:06:01 +00:00
|
|
|
## Ľudia
|
|
|
|
|
|
2019-12-12 09:16:04 +00:00
|
|
|
- [Maroš Harahus](../../../../students/2016/maros_harahus) (štatistické značkovanie)
|
2020-04-14 14:46:05 +00:00
|
|
|
- Jozef Olekšák
|
2020-10-01 14:05:57 +00:00
|
|
|
- [Lukáš Pokrývka](/students/2016/lukas_pokryvka) (Word Embedding Modely)
|
