| --- |
| license: apache-2.0 |
| language: |
| - en |
| - vi |
| metrics: |
| - f1 |
| base_model: |
| - microsoft/deberta-v3-small |
| pipeline_tag: fill-mask |
| tags: |
| - finance |
| - esg |
| - text-classification |
| - fill-mask |
| - bert |
| library_name: transformers |
| datasets: |
| - nguyen599/EnVi-ESG-200 |
| widget: |
| - text: "Over three chapters, it covers a range of topics from energy efficiency and renewable energy to the circular economy and sustainable transportation." |
| --- |
| |
| ESG analysis can help investors determine a business' long-term sustainability and identify associated risks. MaskESG-DeBERTa-v3-small is a [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) model fine-tuned on [EnVi-ESG-200](https://huggingface.co/nguyen599/EnVi-ESG-200) dataset, include 200,000 annotated sentences from Vietnam, English news and ESG reports. |
|
|
| **Input**: A financial text. |
|
|
| **Output**: Environmental, Social, Governance or Neural. |
|
|
| **Language support**: English, Vietnamese |
|
|
| # How to use |
| You can use this model with Transformers pipeline for ESG classification or fill mask task. |
| ```python |
| # tested in transformers==4.53.0 |
| from transformers import AutoTokenizer, AutoModelForMaskedLM, pipeline |
| |
| maskesg = AutoModelForMaskedLM.from_pretrained('nguyen599/MaskESG-DeBERTa-v3-small') |
| tokenizer = AutoTokenizer.from_pretrained('nguyen599/MaskESG-DeBERTa-v3-small') |
| nlp = pipeline("fill-mask", model=maskesg, tokenizer=tokenizer) |
| # Classification as fill-mask |
| results = nlp(f'Over three chapters, it covers a range of topics from energy efficiency and renewable energy to the circular economy and sustainable transportation. This sentence is {tokenizer.mask_token}') |
| print(results) |
| # [{'score': 0.9015821814537048, |
| # 'token': 444, |
| # 'token_str': ' E', |
| # 'sequence': 'Over three chapters, it covers a range of topics from energy efficiency and renewable energy to the circular economy and sustainable transportation. This sentence is E'}, |
| # {'score': 0.09723947197198868, |
| # 'token': 427, |
| # 'token_str': ' N', |
| # 'sequence': 'Over three chapters, it covers a range of topics from energy efficiency and renewable energy to the circular economy and sustainable transportation. This sentence is N'}, |
| # {'score': 0.0010556845227256417, |
| # 'token': 322, |
| # 'token_str': ' S', |
| # 'sequence': 'Over three chapters, it covers a range of topics from energy efficiency and renewable energy to the circular economy and sustainable transportation. This sentence is S'}, |
| # {'score': 0.0001152529803221114, |
| # 'token': 443, |
| # 'token_str': ' G', |
| # 'sequence': 'Over three chapters, it covers a range of topics from energy efficiency and renewable energy to the circular economy and sustainable transportation. This sentence is G'}, |
| # {'score': 1.14425779429439e-06, |
| # 'token': 299, |
| # 'token_str': ' e', |
| # 'sequence': 'Over three chapters, it covers a range of topics from energy efficiency and renewable energy to the circular economy and sustainable transportation. This sentence is e'}] |
| |
| ``` |
|
|
