Home

Contents

• API
• Output Formatting

API

For the given text:

texts = [
  # Text example
  {"text": "It was in July, 1805, and the speaker was the well-known Anna Pávlovna"},
  # other texts
  # ...
]

This is a simple example of API usage:

from bulk_ner.api import NERAnnotator
from bulk_ner.src.service_dynamic import dynamic_init


ner_model = dynamic_init(src_dir="models",
                         class_filepath="dp_130.py",
                         class_name="DeepPavlovNER")(model="ner_ontonotes_bert")

annotator = NERAnnotator(ner_model=ner_model, chunk_limit=128)

data_it = annotator.iter_annotated_data(data_dict_it=texts, prompt="{text}", batch_size=10)
    
for data in data_it:
    # Handle your NER data here ...
    print(data["result"])
    # Output:
    # ['It was in', ['July , 1805', 'DATE', 0], ', and the speaker was the well - known', ['Anna Pávlovna', 'PERSON', 1]]

Output Formatting

We use the most accessible format and represent the output content as a list.

Entries of list types are refer to recognized Named Entities.

Such lists contain the following information:

• Value of the Named Entity
• Type of the named entity
• Index of the entity

Here is an example: