You can deploy this model and start earning money today!

seleznevchamp1/
rubert-base-cased-NER-Matkin

Name: rubert-base-cased-NER-Matkin
Availability: InStock

safetensors

bert

mit

Views

No views yet

language: ru tags:

NER
baseline
ruBERT
DeepPavlov license: mit datasets:
custom: hh.ru metrics:
f1

ruBERT-base-cased (Baseline) for NER on Russian Job Postings

Baseline model reproduced from Matkin et al. (2024), fine-tuned for Skill and Knowledge extraction.

Base model: DeepPavlov/rubert-base-cased
Training: NER fine-tuning with augmentation (500 samples).
F1: 0.67

Usage

python

from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline
model = AutoModelForTokenClassification.from_pretrained("seleznevchamp1/rubert-base-cased-NER-Matkin")
tok = AutoTokenizer.from_pretrained("seleznevchamp1/rubert-base-cased-NER-Matkin")
nlp = pipeline("ner", model=model, tokenizer=tok)
print(nlp("Требуется знание Python и опыт управления командой"))