ultra-accurate-emotion-classifier

Multi-Label Emotion Classification Model

This model was trained for multi-label emotion classification with 14 emotion categories.

Model Details

• Base model: roberta-base
• Maximum sequence length: 128
• Best validation F1 score: 0.5082
• Training epochs: 4

Emotion Labels

['amusement', 'anger', 'annoyance', 'caring', 'confusion', 'disappointment', 'disgust', 'embarrassment', 'excitement', 'fear', 'gratitude', 'joy', 'love', 'sadness']

Usage

from transformers import AutoTokenizer, AutoModel
import torch
import json

# Load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("ax3lrod/ultra-accurate-emotion-classifier")
model = AutoModel.from_pretrained("ax3lrod/ultra-accurate-emotion-classifier", trust_remote_code=True)

# Load thresholds
with open("best_thresholds.json", "r") as f:
    thresholds = json.load(f)

# Inference
text = "Your text here"
inputs = tokenizer(text, return_tensors="pt", max_length=128, truncation=True, padding=True)
outputs = model(**inputs)
probabilities = torch.sigmoid(outputs.logits)

# Apply thresholds
predictions = []
for i, prob in enumerate(probabilities[0]):
    if prob > thresholds[i]:
        predictions.append(i)