Hugging Face's logo

CouchCat
/
ma_ner_v7_distil

Token Classification
Transformers
PyTorch
English
distilbert
ner
Inference Endpoints
Model card Files Community
1
Edit model card

Description

A Named Entity Recognition model trained on a customer feedback data using DistilBert. Possible labels are in BIO-notation. Performance of the PERS tag could be better because of low data samples:

  • PROD: for certain products
  • BRND: for brands
  • PERS: people names

The following tags are simply in place to help better categorize the previous tags

  • MATR: relating to materials, e.g. cloth, leather, seam, etc.
  • TIME: time related entities
  • MISC: any other entity that might skew the results

Usage 

from transformers import AutoTokenizer, AutoModelForTokenClassification

  tokenizer = AutoTokenizer.from_pretrained("CouchCat/ma_ner_v7_distil")

  model = AutoModelForTokenClassification.from_pretrained("CouchCat/ma_ner_v7_distil")
Downloads last month
23
Inference Examples
Token Classification
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.