TAI-1.1: Turkish Language Model

TAI-1.1 (Turkish AI 1.1) is a fine-tuned BERT model for Turkish language understanding and processing tasks. This model is based on the "dbmdz/bert-base-turkish-cased" architecture and has been further trained on a diverse Turkish text dataset.

Model Details

• Model Type: BERT (Bidirectional Encoder Representations from Transformers)
• Language: Turkish
• Base Model: dbmdz/bert-base-turkish-cased
• Training Dataset: Turkish Texts Dataset 2
• Tasks: Text Classification, Sentiment Analysis, Named Entity Recognition (NER)

Usage

You can use this model directly with the Hugging Face Transformers library:

from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained("yasarefe/tai-1.1")
model = AutoModelForSequenceClassification.from_pretrained("yasarefe/tai-1.1")

# Example usage
text = "Bu bir örnek Türkçe cümledir."
inputs = tokenizer(text, return_tensors="pt")
outputs = model(**inputs)

Fine-tuning

This model can be fine-tuned for various downstream tasks such as sentiment analysis, text classification, or named entity recognition. Here's an example of how to fine-tune the model:

from transformers import Trainer, TrainingArguments

training_args = TrainingArguments(
    output_dir="./results",
    num_train_epochs=3,
    per_device_train_batch_size=16,
    per_device_eval_batch_size=64,
    warmup_steps=500,
    weight_decay=0.01,
    logging_dir="./logs",
)

trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=train_dataset,
    eval_dataset=eval_dataset,
)

trainer.train()

Limitations

While TAI-1.1 has been trained on a diverse dataset, it may still have biases or limitations:

• The model's performance may vary depending on the specific domain or task.
• It may not perform well on very specialized or technical Turkish texts.
• The model may exhibit biases present in the training data.

Citation

If you use this model in your research, please cite:

@misc{tai-1.1,
  author = {Yasar Efe},
  title = {TAI-1.1: Turkish Language Model},
  year = {2024},
  publisher = {HuggingFace},
  journal = {HuggingFace Model Hub},
  howpublished = {\url{https://huggingface.co/yasarefe/tai-1.1}},
}

Contact

For any questions or feedback, please open an issue on the model repository or contact the author directly.