README.md

---
language: en
license: mit
tags:
- emotion-classification
- multi-label
---

# Emotion Classifier (DeBERTa-v3-base)

F1: 0.8008092786408941

## Usage

import torch
import torch.nn as nn
from transformers import DebertaV2Tokenizer, DebertaV2Model

class TransformerClassifier(nn.Module):
def init(self, model_name, n_classes=5):
super().init()
self.transformer = DebertaV2Model.from_pretrained(model_name)
self.classifier = nn.Linear(768, n_classes)

def forward(self, input_ids, attention_mask):
    outputs = self.transformer(input_ids=input_ids, attention_mask=attention_mask)
    return self.classifier(outputs.last_hidden_state[:, 0])
tokenizer = DebertaV2Tokenizer.from_pretrained('YOUR_USERNAME/emotion-classifier-deberta')
model = TransformerClassifier('YOUR_USERNAME/emotion-classifier-deberta')
checkpoint = torch.load('pytorch_model.bin')
model.load_state_dict(checkpoint['model_state_dict'])
model.eval()


Anish Sharma (23f1001021) - IIT Madras 2025