license: mit
language:
- tr
metrics:
- accuracy
- f1
pipeline_tag: token-classification
tags:
- ner
- turkish-ner
- turkish
- nlp
Bu model "https://github.com/stefan-it/turkish-bert" base alınarak geliştirilmiş bir NER(Varlık ismi tanıma) modelidir.
Eğitim ve validasyon verisi
Fine-tune işlemi için TDD-NER-202112-CC-002 veri seti kullanılmıştır.
@inproceedings{pan-etal-2017-cross, title = "Cross-lingual Name Tagging and Linking for 282 Languages", author = "Pan, Xiaoman and Zhang, Boliang and May, Jonathan and Nothman, Joel and Knight, Kevin and Ji, Heng", booktitle = "Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", month = jul, year = "2017", address = "Vancouver, Canada", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/P17-1178", doi = "10.18653/v1/P17-1178", pages = "1946--1958" }
Hiperparametreler
custom_labels = ["O","B-LOC","I-LOC","B-ORG","I-ORG","B-PER","I-PER"]
model_args = { "train_batch_size": 32, "eval_batch_size": 32, "num_train_epochs": 3, "seed":1, "save_steps": 625, "overwrite_output_dir": True, "output_dir": "/content/Model" }
Eğitim Metrikleri
| Epochs | Running Loss |
|---|---|
| 1 | 0.1152 |
| 2 | 0.1091 |
| 3 | 0.0586 |
Nasıl Kullanılacağı
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("token-classification", model="Gorengoz/bert-based-Turkish-NER-wikiann")
pipe("Entity X'in müşteri hizmetleri hızlı ve etkili, Entity Y'nin ürün kalitesi çok kötü.",aggregation_strategy = "simple"")