ESG Scoring Model

This model performs ESG (Environmental, Social, Governance) scoring for text classification.

Model Description

Model Type: Sequence Classification for ESG Scoring
Language: English, Vietnamese
Task: ESG Factor Scoring (E, S, G)

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load model and tokenizer
model = AutoModelForSequenceClassification.from_pretrained("chungpt2123/esg-scoring", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained("chungpt2123/esg-scoring")

# Example usage
text = "The company has implemented renewable energy solutions to reduce carbon emissions."
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=8192)

# Get probabilities for each ESG factor
with torch.no_grad():
    outputs = model(**inputs)
    probabilities = torch.nn.functional.softmax(outputs.logits, dim=1)
    
# Get scores for each factor
e_score = probabilities[0, 0].item()  # Environmental score
s_score = probabilities[0, 1].item()  # Social score 
g_score = probabilities[0, 2].item()  # Governance score

print(f"Environmental: {e_score:.4f}")
print(f"Social: {s_score:.4f}")
print(f"Governance: {g_score:.4f}")

Training Details

Training Data: Custom ESG dataset
Training Approach: Fine-tuned for ESG factor scoring
Labels: E (Environmental), S (Social), G (Governance)

Model Performance

The model achieves strong performance on ESG scoring tasks across multiple languages.

Limitations

Trained primarily on English and Vietnamese text
Performance may vary on domain-specific or technical content
Best performance on texts similar to training data distribution

@misc{esg_scoring_model,
  title={ESG Scoring Model},
  author={Chung},
  year={2024},
  publisher={Hugging Face},
  url={https://huggingface.co/chungpt2123/esg-scoring}
}