Spacy Predictor for Named Entity Recognition

This is a module for spaCy Named Entity Recognition industrial-strength prediction.

Getting started

Your trained model

To get started, you will need a pretrained spacy 3 model. Please note that spacy 2 models may or may not be compatible with this module. Please refer to spaCy's documentation on loading models if you have further questions.

Dependencies

Please ensure that your environment has spacy installed (please see requirements.txt for specific versions) and that you are running Python 3.8 or higher.

main.py

Once you have identified your model (which should be a folder), place the folder in the same directory as the main.py. Then open main.py and make appropriate changes to use the spacy model to predict on your text of choice.

Sample models

If you would like to have some models to test out this setup, feel free to work with our content on Hugging Face: https://huggingface.co/sigtica.

Example 1

For English, a fun one to try out is the following:

First in shell, run:

pip install https://huggingface.co/sigtica/en_data_dev_spacy_trf_1/resolve/main/en_data_dev_spacy_trf_1-any-py3-none-any.whl

Then in a python runtime, run:

# Using spacy.load().
import spacy
nlp = spacy.load("en_data_dev_spacy_trf_1")
doc = nlp('I live in London.')

# print out the tokens
for token in doc:
    print(token)

# print out the entities
for ent in doc.ents:
    print(ent, ent.label_)


# Importing as module.
import en_data_dev_spacy_trf_1
nlp2 = en_data_dev_spacy_trf_1.load()
doc = nlp2('I live in London.')

# print out the tokens
for token in doc:
    print(token)

# print out the entities
for ent in doc.ents:
    print(ent, ent.label_)

Example 2

For Chinese, a fun one to try out is the following:

First in shell, run:

pip install https://huggingface.co/sigtica/zh_data_dev_spacy_trf_1/resolve/main/zh_data_dev_spacy_trf_1-any-py3-none-any.whl

Then in a python runtime, run:

# Using spacy.load().
import spacy
nlp = spacy.load("zh_data_dev_spacy_trf_1")
doc = nlp('北京大兴国际机场是位于中国北京市大兴区榆垡镇、礼贤镇和河北省廊坊市广阳区之间的一座大型民用机场，于2019年9月25日正式运营。其飞行区等级为4F，是北京市的第二座国际机场，具有“新国门”之稱。
')

# print out the tokens
for token in doc:
    print(token)

# print out the entities
for ent in doc.ents:
    print(ent, ent.label_)


# Importing as module.
import zh_data_dev_spacy_trf_1
nlp2 = zh_data_dev_spacy_trf_1.load()
doc = nlp2('北京大兴国际机场是位于中国北京市大兴区榆垡镇、礼贤镇和河北省廊坊市广阳区之间的一座大型民用机场，于2019年9月25日正式运营。其飞行区等级为4F，是北京市的第二座国际机场，具有“新国门”之稱。
')

# print out the tokens
for token in doc:
    print(token)

# print out the entities
for ent in doc.ents:
    print(ent, ent.label_)

Support

Please direct any enquiries to info@sigtica.com or check out our Hugging Face page.