You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Tone and Sentiment Classification Model

A fine-tuned XLM-RoBERTa model for tone classification in customer service communications.

Model Details

Base Model: xlm-roberta-base
Task: Multi-label tone classification
Languages: English, Malay, Indonesian
Labels: helpful, empathetic, friendly, dismissive, unprofessional

Usage

from transformers import AutoTokenizer, AutoModel
from src.inference_tone_only import TonePredictor

predictor = TonePredictor("your-username/tone-sentiment-model")
result = predictor.predict("Thank you for contacting us!")
print(result)

Training Details

Training data: 25,538 examples
Validation data: 3,188 examples
Test data: 3,207 examples
Training epochs: 4
Learning rate: 2e-5

Downloads last month: 274

Safetensors

Model size

0.3B params

Tensor type

F32