my-emotion-classifier

My Emotion Classifier 🎭

This is a fine-tuned DistilBERT model for emotion classification. It can classify text into 6 different emotions:

• 😢 Sadness
• 😊 Joy
• ❤️ Love
• 😠 Anger
• 😨 Fear
• 😲 Surprise

Model Description

Base Model: distilbert-base-uncased
Fine-tuned on: emotion dataset
Task: Multi-class text classification
Language: English

Training Details

• Training epochs: 3
• Batch size: 16
• Learning rate: 2e-5
• Optimizer: AdamW
• Max sequence length: 128

Performance

The model achieves approximately:

• Accuracy: ~92%
• F1 Score: ~91%

Usage

Using Transformers Pipeline

from transformers import pipeline

# Load the model
classifier = pipeline("text-classification", model="YOUR_USERNAME/my-emotion-classifier")

# Make predictions
text = "I love this so much! Best day ever!"
result = classifier(text)
print(result)
# [{'label': 'joy', 'score': 0.95}]

Using AutoModel

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("YOUR_USERNAME/my-emotion-classifier")
model = AutoModelForSequenceClassification.from_pretrained("YOUR_USERNAME/my-emotion-classifier")

# Tokenize and predict
text = "This makes me so angry!"
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=128)
outputs = model(**inputs)
predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
predicted_class = torch.argmax(predictions, dim=-1).item()

print(f"Predicted emotion: {model.config.id2label[predicted_class]}")

Using the Hugging Face CLI

# Install the transformers library
pip install transformers

# Use the model in Python
python -c "from transformers import pipeline; classifier = pipeline('text-classification', model='YOUR_USERNAME/my-emotion-classifier'); print(classifier('I am so happy!'))"

Example Predictions

Limitations

• The model is trained on English text only
• Best performance on short texts (tweets, reviews, comments)
• May struggle with sarcasm or mixed emotions
• Limited to 6 emotion categories

Training Data

The model was fine-tuned on the emotion dataset which contains:

• 16,000 training examples
• 2,000 validation examples
• 2,000 test examples

Ethical Considerations

This model should not be used for:

• Medical or psychological diagnosis
• Making critical decisions about individuals
• Surveillance or monitoring without consent

Citation

If you use this model, please cite:

@misc{my-emotion-classifier,
  author = {Your Name},
  title = {My Emotion Classifier},
  year = {2025},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/YOUR_USERNAME/my-emotion-classifier}}
}

Contact

For questions or feedback, please open an issue on the model's Hugging Face page.