|
--- |
|
language: |
|
- en |
|
library_name: flair |
|
pipeline_tag: token-classification |
|
tags: |
|
- statistics |
|
datasets: |
|
- conll2003 |
|
--- |
|
|
|
|
|
## Overview |
|
This model is used to identify statistical named entities in large text. Statistical Named Entities are entities that indicate the presence of a statistical claim (such as a hypothesis of an experiment) along with the type of test and the confidence value. |
|
|
|
Use this model in your repo to categorize a text document to find claims, test statistics and probability scores. The model uses Flair NLP from ground-up to develop a Stats NER for researchers. |
|
|
|
## Usage |
|
|
|
from flair.models import SequenceTagger |
|
|
|
tagger = SequenceTagger.load("VinayNR/stats-ner") |
|
|
|
sentence = Sentence(<your_string>, use_tokenizer=True) |
|
|
|
tagger.predict(sentence) |
