NER Recipe Ingredient Recognition with BERT (BIO)

Model Details

The model recognizes entity types using BIO tagging scheme

Model Description

This project implements a Named Entity Recognition (NER) system for recipe ingredient text using a BERT model (BertForTokenClassification и TokenizerFast). The model can identify various components within recipe ingredient lists, such as quantities, units, ingredient names, and comments.

Developed by: [Daniil Botvenko]

Downloads last month: 1

Safetensors

Model size

89.3M params

Tensor type

F32