sparv-word-prediction-kb-bert-plugin 0.5.1

A sparv plugin for computing word neighbours using a BERT model.

sparv-word-prediction--kb-bert-plugin

Plugin for applying bert masking as a Sparv annotation.

Install

First, install Sparv, as suggested:

pipx install sparv-pipeline

Then install install sparv-word-prediction-kb-bert-plugin with

pipx inject sparv-pipeline sparv-word-prediction-kb-bert-plugin

Usage

Depending on how many explicit exports of annotations you have you can decide to use this annotation exclusively by adding it as the only annotation to export under xml_export:

xml_export:
    annotations:
        - <token>:word_prediction_kb_bert.word-prediction--kb-bert

To use it together with other annotations you might add it under export:

export:
    annotations:
        - <token>:word_prediction_kb_bert.word-prediction--kb-bert
        ...

Configuration

You can configure this plugin by the number of neighbours to generate.

Number of Neighbours

The number of neighbours defaults to 5 but can be configured in config.yaml:

word_prediction_kb_bert:
    num_neighbours: 5

Number of Decimals

The number of decimals defaults to 3 but can be configured in config.yaml:

word_prediction_kb_bert:
    num_decimals: 3

[!NOTE] This also controls the cut-off, so all values where the score round to 0.000 (or the number of decimals) is discarded.

Metadata

Model

Type	HuggingFace Model	Revision
Model	KBLab/bert-base-swedish-cased	c710fb8dff81abb11d704cd46a8a1e010b2b022c
Tokenizer	same as Model	same as Model

Changelog

This project keeps a changelog.