Mbert Base Albanian Cased Ner

🚀 Albanian Named Entity Recognition (NER) Model

This model is a fine - tuned version of "bert - base - multilingual - cased", which can effectively perform named entity recognition tasks in Albanian, leveraging the well - known WikiANN dataset.

🚀 Quick Start

This model is the fine - tuned model of "bert - base - multilingual - cased" using the famous WikiANN dataset presented in the "Cross - lingual Name Tagging and Linking for 282 Languages" paper.

✨ Features

• It is based on the "bert - base - multilingual - cased" model, which has strong cross - language generalization ability.
• Fine - tuned on the WikiANN dataset, it can accurately identify named entities in Albanian text.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

model = AutoModelForTokenClassification.from_pretrained("akdeniz27/mbert-base-albanian-cased-ner")
tokenizer = AutoTokenizer.from_pretrained("akdeniz27/mbert-base-albanian-cased-ner")
ner = pipeline('ner', model=model, tokenizer=tokenizer, aggregation_strategy="first")
ner("<your text here>")

Pls refer "https://huggingface.co/transformers/_modules/transformers/pipelines/token_classification.html" for entity grouping with aggregation_strategy parameter.

🔧 Technical Details

Fine - tuning parameters:

task = "ner"
model_checkpoint = "bert-base-multilingual-cased"
batch_size = 8 
label_list = ['O', 'B-PER', 'I-PER', 'B-ORG', 'I-ORG', 'B-LOC', 'I-LOC']
max_length = 512 
learning_rate = 2e-5 
num_train_epochs = 3 
weight_decay = 0.01 

📚 Documentation

Reference test results:

📄 License

This project is licensed under the MIT license.

⚠️ Important Note

Varianti AY.4.2 është më i lehtë për t'u transmetuar, thotë Francois Balu, drejtor i Institutit të Gjenetikës në Londër. (The original text in Albanian is kept here as no clear translation context is provided for this sentence.)