Downloads · 30 days
86
1% of all-time downloads
bert-base-turkish-cased-ner-quantized
akdeniz27/bert-base-turkish-cased-ner-quantized
bert-base-turkish-cased-ner-quantized is a token classification model from akdeniz27. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as mit.
This model is the dynamically quantized version of the model (https://akdeniz27/bert-base-turkish-cased-ner)
All-time downloads
10.2K
Public
Repo size
191 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.onnx186 MB · 97%
At a glance
- Task
- Token Classification
- Library
- transformers
- License
- mit
- Model type
- bert
- Access
- Public
- Created
- Nov 20, 2022
- Updated
- Dec 23, 2022
- SHA
- fb6a421c
Model card
From the Hugging Face model README
Turkish Named Entity Recognition (NER) Quantized Model
This model is the dynamically quantized version of the model (https://akdeniz27/bert-base-turkish-cased-ner)
How to use:
# First install "optimum[onnxruntime]":
!pip install "optimum[onnxruntime]"
# and import "ORTModelForTokenClassification":
from transformers import AutoTokenizer, pipeline
from optimum.onnxruntime import ORTModelForTokenClassification
model = ORTModelForTokenClassification.from_pretrained("akdeniz27/bert-base-turkish-cased-ner-quantized", file_name="model_quantized.onnx")
tokenizer = AutoTokenizer.from_pretrained("akdeniz27/bert-base-turkish-cased-ner-quantized")
ner = pipeline('ner', model=model, tokenizer=tokenizer, aggregation_strategy="first")
ner("your text here")
Pls refer (https://github.com/akdeniz27/dynamic_quantization) for details of quantization.
- Task
- Token Classification
- Library
- transformers
- Type
- bert
- License
- mit
- Languages
- tr
- Created
- Nov 20, 2022
- Updated
- Dec 23, 2022