https://github.com/thjbdvlt/spacy-french-parser

syntactic dependency parser for french using spacy
https://github.com/thjbdvlt/spacy-french-parser

french nlp nlp-french spacy spacy-parser syntactic-dependency-parsing universal-dependencies

Last synced: 6 days ago
JSON representation

syntactic dependency parser for french using spacy

• Host: GitHub
• URL: https://github.com/thjbdvlt/spacy-french-parser
• Owner: thjbdvlt
• License: other
• Created: 2024-08-26T07:57:18.000Z (2 months ago)
• Default Branch: sea
• Last Pushed: 2024-08-27T08:21:19.000Z (2 months ago)
• Last Synced: 2024-10-10T11:42:48.267Z (26 days ago)
• Topics: french, nlp, nlp-french, spacy, spacy-parser, syntactic-dependency-parsing, universal-dependencies
• Language: Python
• Homepage:
• Size: 1.07 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
__syntactic dependency parser__ for french with [spacy](https://spacy.io/api/).

this repository is comprised of scripts that fetch and prepare data to train a [syntactic dependencies](https://universaldependencies.org/u/dep/) parser with [spacy](https://spacy.io/api/) for the french language, along with a configuration file and script to train it. the __model__ itself is available under __releases__.

the data used for the training is an aggregation of three [UD](https://universaldependencies.org/) datasets and makes some minor changes to these datasets.

in the datasets i used, the word _du_ is splitted into its logical component _de_ and _le_. a text like _on parle du ciel_ becomes _on parle de le ciel_ in the `.conllu` files. but in the texts i have to analyze, _du_ isn't splitted at all, so i need to unsplit it. thus the following:

```conllu

11-12	du	...	_	_	_	_

11	de	...	19	case	_	_

12	le	...	11	det	_	_

```

is transformed into:

```conllu

11	du	...	19	case:det	_	_

```

upon that, some labels are replaced by others, and sentences containing certain labels (such as `dep` which indicates than the parsing failed) are removed. for a list of replaced or removed labels, refer the file [lookup_labels.txt](./lookup_labels.txt).

usage

-----

the __parser__ is not a full pipeline. you have to source it from another pipeline as a component:

```python3

import spacy

# load your main pipeline

nlp = spacy.load('fr_core_news_sm', exclude=['parser'])

# load the model containing the parser

nlp_deps = spacy.load('./model', exclude=['tokenizer'])

# put the parser in the main pipeline

nlp.add_pipe('parser', source=nlp_deps)

```