electra-tr-product-reviews-daft

Electra-TR Product Reviews (DAFT)

This is a Domain-Adaptive Fine-Tuned (DAFT) version of a Turkish ELECTRA model, adapted for Turkish product reviews sentiment analysis. The model was fine-tuned on the Turkish Product Reviews dataset.

📌 Model Details

• Base model: incidelen/electra-small-turkish-sentiment-analysis-cased
• Language: Turkish (tr)
• Task: Sentiment Analysis (binary classification)
• Domain: Product reviews (e-commerce)
• Method: Domain-Adaptive Fine-Tuning (DAFT)

🚀 Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

model_name = "mertafacan/electra-tr-product-reviews-daft"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

text = "Ürün beklediğimden çok daha kaliteli, tavsiye ederim!"
inputs = tokenizer(text, return_tensors="pt")
outputs = model(**inputs)

pred = torch.argmax(outputs.logits, dim=1).item()
print("Positive" if pred == 1 else "Negative")

⚠️ Disclaimer

This model was fine-tuned as an experimental study and is not a general-purpose solution. Use with caution and consider other models for broader applications.

📬 Contact

Author: Mert Afacan Email: [email protected]