omgavali26/emotion-analysis

IndicBERTv2 Marathi Emotion Classifier

This model is a fine-tuned version of ai4bharat/IndicBERTv2-MLM-only for multiclass Marathi Emotion Classification across 27 fine-grained emotion categories, with selected categories mapped into 6 core emotion groups.

Model Details

• Base Model: ai4bharat/IndicBERTv2-MLM-only
• Architecture: IndicBERTv2
• Task: Multiclass Emotion Classification
• Number of Labels: 27
• Target Language: Marathi (mr)
• Framework: PyTorch
• Library: Hugging Face Transformers
• Maximum Sequence Length: 128

27 Fine-Grained Emotion Classes

admiration, amusement, anger, annoyance, approval, caring, confusion, curiosity, desire, disappointment, disapproval, disgust, embarrassment, excitement, fear, gratitude, joy, love, nervousness, neutral, optimism, pride, realization, relief, remorse, sadness, surprise.

How to Use

The model can be used with the Hugging Face Transformers pipeline API.

Replace the placeholder model repository with your actual Hugging Face model ID.

import torch
from transformers import pipeline

model_id = "omgavali26/emotion-analysis"

emotion_pipeline = pipeline(
    "text-classification",
    model=model_id,
    tokenizer=model_id,
    device=0 if torch.cuda.is_available() else -1,
    top_k=None
)

text = "मुंबईतील कोस्टल रोड परिसरात पर्यटक आणि नागरिकांसाठी अत्याधुनिक बायो टॉयलेट उभारण्यात येत आहे."

result = emotion_pipeline(text)

print(result)

License

Apache License 2.0

The license should be verified against the base model, training dataset, and any other applicable licensing requirements before distribution.