DistilRoBERTa Fine-Tuned for Emotion Classification
Fine-tuned from j-hartmann/emotion-english-distilroberta-base on the dair-ai/emotion dataset.
Part of the Mental Health Agentic AI Platform.
Performance
| Metric | Baseline | Fine-Tuned |
|---|---|---|
| Macro F1 | 0.869 | 0.892 |
| Accuracy | 0.914 | 0.930 |
Labels
- 0: sadness
- 1: joy
- 2: love
- 3: anger
- 4: fear
- 5: surprise
Usage
from transformers import pipeline
classifier = pipeline("text-classification", model="YDVJIYA/distilroberta-base-finetuned-emotion")
result = classifier("I feel hopeless and tired today")
print(result)
Training Configuration
Epochs: 3
Batch size: 32
Learning rate: 2e-5
Warmup ratio: 0.1
Weight decay: 0.01
Mixed precision (fp16)
Seed: 42
Author
Trained by Jiya Yadav as part of an end-to-end agentic AI platform.
- Downloads last month
- 16
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support