https://github.com/NeonGeckoCom/neon_utterance_zeroshot_ner_plugin

https://github.com/NeonGeckoCom/neon_utterance_zeroshot_ner_plugin

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/NeonGeckoCom/neon_utterance_zeroshot_ner_plugin
• Owner: NeonGeckoCom
• Created: 2021-11-23T11:27:31.000Z (about 3 years ago)
• Default Branch: master
• Last Pushed: 2022-02-15T02:12:38.000Z (almost 3 years ago)
• Last Synced: 2024-02-15T15:32:53.386Z (10 months ago)
• Language: Python
• Size: 4.88 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 2
• Metadata Files:
  • Readme: readme.md

Awesome Lists containing this project

• awesome-ovos-plugins - neon_utterance_zeroshot_ner_plugin - inject context (NER) (OPM plugins / Utterance Transformers)

README

        
```python

tars = TarsZeroShotNER()

# this usually comes from mycroft.conf and is set by transformers service

# labels can be anything, no training needed (zero shot learning)

# the labels in this example are the default values if not set

tars.config = { "labels": ["Organization", "Vehicle", "Location", "City", "Country", "Person"]}

utts = ["The Humboldt University of Berlin is situated near the Spree in Berlin, Germany",

        "The president of Portugal is Marcelo Rebelo de Sousa"]

_, context = tars.transform(utts)

print(context)

# {'zeroshot_ner': {'City': [{'entity': 'City',

#                             'source_text': 'The Humboldt University of Berlin '

#                                            'is situated near the Spree in '

#                                            'Berlin, Germany',

#                             'span': (64, 70),

#                             'value': 'Berlin'}],

#                   'Country': [{'entity': 'Country',

#                                'source_text': 'The Humboldt University of '

#                                               'Berlin is situated near the '

#                                               'Spree in Berlin, Germany',

#                                'span': (72, 79),

#                                'value': 'Germany'}],

#                   'Location': [{'entity': 'Location',

#                                 'source_text': 'The Humboldt University of '

#                                                'Berlin is situated near the '

#                                                'Spree in Berlin, Germany',

#                                 'span': (55, 60),

#                                 'value': 'Spree'},

#                                {'entity': 'Location',

#                                 'source_text': 'The president of Portugal is '

#                                                'Marcelo Rebelo de Sousa',

#                                 'span': (17, 25),

#                                 'value': 'Portugal'}],

#                   'Organization': [{'entity': 'Organization',

#                                     'source_text': 'The Humboldt University of '

#                                                    'Berlin is situated near '

#                                                    'the Spree in Berlin, '

#                                                    'Germany',

#                                     'span': (4, 33),

#                                     'value': 'Humboldt University of Berlin'}],

#                   'Person': [{'entity': 'Person',

#                               'source_text': 'The president of Portugal is '

#                                              'Marcelo Rebelo de Sousa',

#                               'span': (29, 52),

#                               'value': 'Marcelo Rebelo de Sousa'}]}}

```