Description
The BERT-Emotions-Classifier is a fine-tuned BERT-based model designed for multi-label emotion classification. It has been trained on the sem_eval_2018_task_1 dataset, which includes text samples labeled with a variety of emotions, including anger, anticipation, disgust, fear, joy, love, optimism, pessimism, sadness, surprise, and trust. The model is capable of classifying text inputs into one or more of these emotion categories.
Overview
- Model Name: BERT-Emotions-Classifier
- Task: Multi-label emotion classification
- Dataset: sem_eval_2018_task_1
- Labels: ['anger', 'anticipation', 'disgust', 'fear', 'joy', 'love', 'optimism', 'pessimism', 'sadness', 'surprise', 'trust']
- Base Model: BERT (Bidirectional Encoder Representations from Transformers)
Input Format
The model expects text input in the form of a string.
Output Format
- The model provides a list of labels and associated scores, indicating the predicted emotions and their confidence scores.
Example Applications
- Emotion analysis in social media posts
- Sentiment analysis in customer reviews
- Content recommendation based on emotional context
Ethical Considerations
When using this model, it's essential to consider the ethical implications of emotion analysis. Ensure that the use of emotional data respects privacy and consent, and avoid making decisions that could have adverse effects based solely on emotion analysis.
Inference
from transformers import pipeline
# Load the BERT-Emotions-Classifier
classifier = pipeline("text-classification", model="ayoubkirouane/BERT-Emotions-Classifier")
# Input text
text = "Your input text here"
# Perform emotion classification
results = classifier(text)
# Display the classification results
print(results)