shivvamm/emotion-distilbert-finetuned

😄 Emotion Classifier — Fine-tuned DistilBERT

This is a DistilBERT-based model fine-tuned for emotion classification on the dair-ai/emotion dataset.

It predicts one of 6 emotion labels for a given text:

• sadness
• joy
• love
• anger
• fear
• surprise

---

🧠 Model Details

• Base model: distilbert-base-uncased
• Fine-tuned on: dair-ai/emotion
• Task: Multi-class single-label text classification
• Framework: 🤗 Transformers

---

📊 Dataset

Dataset: dair-ai/emotion

Split	Samples
Train	16,000
Validation	2,000
Test	2,000
Total	20,000

Classes (labels):

• 0 → sadness
• 1 → joy
• 2 → love
• 3 → anger
• 4 → fear
• 5 → surprise

---

🚀 Usage

You can use the model directly with the 🤗 Transformers pipeline:

from transformers import pipeline

classifier = pipeline("text-classification", model="shivvamm/emotion-distilbert-finetuned", top_k=None)

text = "I feel hopeful and excited about the future."
results = classifier(text)
print(results)

Example output:

[{'label': 'joy', 'score': 0.9876}]

---

✅ Intended Use

• Social media and product review emotion analysis
• Sentiment & psychological tone detection
• Personal journaling or well-being apps
• Academic research in NLP and affective computing

---

⚠️ Limitations

• Handles English language only
• Classifies only one dominant emotion per text
• May misclassify mixed emotions, sarcasm, or idioms

---

📄 Citation

@misc{shivvamm2025emotion,
  title={Emotion Classification using DistilBERT},
  author={Shivvamm},
  year={2025},
  howpublished={\url{https://huggingface.co/shivvamm/emotion-distilbert-finetuned}},
  note={Fine-tuned on the dair-ai/emotion dataset}
}

---

👤 Author

Shivvamm
Model: shivvamm/emotion-distilbert-finetuned
License: MIT

---

💡 Fine-tuned with 🤗 Hugging Face Transformers and Accelerate