sparv-sbx-sentence-sentiment-kb-sent 0.1.0

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

sparv-sbx-sentence-sentiment-kb-sent

Plugin for applying bert masking as a Sparv annotation.

Install

First, install Sparv as suggested:

pipx install sparv-pipeline

Then install install sparv-sbx-sentence-sentiment-kb-sent with

pipx inject sparv-pipeline sparv-sbx-sentence-sentiment-kb-sent

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>:sbx_sentence_sentiment_kb_sent.sentence-sentiment--kb-bert

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

export:
    annotations:
        - <token>:sbx_sentence_sentiment_kb_sent.sentence-sentiment--kb-bert
        ...

Configuration

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

Number of Decimals

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

sbx_sentence_sentiment_kb_sent:
    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/robust-swedish-sentiment-multiclass	b0ec32dca56aa6182a6955c8f12129bbcbc7fdbd
Tokenizer	KBLab/megatron-bert-large-swedish-cased-165k	90c57ab49e27b820bd85308a488409dfea25600d

Changelog

This project keeps a changelog.