sakhawat-hossen/bangla-sentiment-banglabert

BanglaBERT Sentiment Classifier

This is a fine‑tuned BanglaBERT model trained on the BANEmo Dataset (https://ieeexplore.ieee.org/document/11171926) for Bangla sentiment analysis.
It classifies Bangla comments into two categories:

• 0 → Sadness 😢
• 1 → Happiness 😀

---

📊 Model Details

• Base model: sagorsharma/banglabert
• Task: Sequence Classification (Sentiment Analysis)
• Labels: 2 (Sadness, Happiness)
• Dataset: BANEmo (https://ieeexplore.ieee.org/document/11171926)
• Evaluation Metrics: Accuracy, F1 Score
• Validation Accuracy: ~84%

---

🛠️ Training Setup

• Framework: Hugging Face Transformers
• Optimizer: AdamW
• Learning Rate: 5e‑5
• Epochs: 3
• Batch Size: 8
• Loss Function: CrossEntropyLoss

---

🚀 Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load model and tokenizer
model = AutoModelForSequenceClassification.from_pretrained("sakhawat-hossen/bangla-sentiment-banglabert")
tokenizer = AutoTokenizer.from_pretrained("sakhawat-hossen/bangla-sentiment-banglabert")

# Example prediction
text = "আজ আমি খুব খুশি।"
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
with torch.no_grad():
    outputs = model(**inputs)
prediction = torch.argmax(outputs.logits, dim=-1).item()

print("Prediction:", "Happiness 😀" if prediction == 1 else "Sadness 😢")