Emotion-X: Fine-tuned DeBERTa-Xlarge Based Emotion Detection
This is a fine-tuned version of microsoft/deberta-xlarge-mnli for emotion detection on the dair-ai/emotion dataset.
Overview
Emotion-X is a state-of-the-art emotion detection model fine-tuned from Microsoft's DeBERTa-Xlarge model. Designed to accurately classify text into one of six emotional categories, Emotion-X leverages the robust capabilities of DeBERTa and fine-tunes it on a comprehensive emotion dataset, ensuring high accuracy and reliability.
Model Details
- Model Name:
AnkitAI/deberta-xlarge-base-emotions-classifier - Base Model:
microsoft/deberta-xlarge-mnli - Dataset: dair-ai/emotion
- Fine-tuning: This model was fine-tuned for emotion detection with a classification head for six emotional categories (anger, disgust, fear, joy, sadness, surprise).
Training
The model was trained using the following parameters:
- Learning Rate: 2e-5
- Batch Size: 4
- Weight Decay: 0.01
- Evaluation Strategy: Epoch
Training Details
- Evaluation Loss: 0.0858
- Evaluation Runtime: 110070.6349 seconds
- Evaluation Samples/Second: 78.495
- Evaluation Steps/Second: 2.453
- Training Loss: 0.1049
- Evaluation Accuracy: 94.6%
- Evaluation Precision: 94.8%
- Evaluation Recall: 94.5%
- Evaluation F1 Score: 94.7%
Usage
You can use this model directly with the Hugging Face transformers library:
from transformers import AutoModelForSequenceClassification, AutoTokenizer
model_name = "AnkitAI/deberta-xlarge-base-emotions-classifier"
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
# Example usage
def predict_emotion(text):
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=128)
outputs = model(**inputs)
logits = outputs.logits
predictions = logits.argmax(dim=1)
return predictions
text = "I'm so happy with the results!"
emotion = predict_emotion(text)
print("Detected Emotion:", emotion)
Emotion Labels
- Anger
- Disgust
- Fear
- Joy
- Sadness
- Surprise
Model Card Data
| Parameter | Value |
|---|---|
| Model Name | microsoft/deberta-xlarge-mnli |
| Training Dataset | dair-ai/emotion |
| Learning Rate | 2e-5 |
| Per Device Train Batch Size | 4 |
| Evaluation Strategy | Epoch |
| Best Model Accuracy | 94.6% |
Support the Project
If this model is useful in your work, you can support independent research:
License
This model is licensed under the MIT License.
More models: ankitaglawe.com