Model Card for EnvironmentalBERT-water

Model Description

Based on this paper, this is the EnvironmentalBERT-water language model. A language model that is trained to better classify water texts in the ESG/nature domain.

Using the EnvironmentalBERT-base model as a starting point, the EnvironmentalBERT-water Language Model is additionally fine-trained on a 2.2k water dataset to detect water text samples.

How to Get Started With the Model

It is highly recommended to first classify a sentence to be "environmental" or not with the EnvironmentalBERT-environmental model before classifying whether it is "water" or not.

See these tutorials on Medium for a guide on model usage, large-scale analysis, and fine-tuning.

You can use the model with a pipeline for text classification:

from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline
 
tokenizer_name = "ESGBERT/EnvironmentalBERT-water"
model_name = "ESGBERT/EnvironmentalBERT-water"
 
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(tokenizer_name, max_len=512)
 
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer) # set device=0 to use GPU
 
# See https://huggingface.co/docs/transformers/main_classes/pipelines#transformers.pipeline
print(pipe("Water scarcity plays an increasing role in local communities in the South-West of the US.", padding=True, truncation=True))

More details can be found in the paper

@article{Schimanski23ExploringNature,
    title={{Exploring Nature: Datasets and Models for Analyzing Nature-Related Disclosures}},
    author={Tobias Schimanski and Chiara Colesanti Senni and Glen Gostlow and Jingwei Ni and Tingyu Yu and Markus Leippold},
    year={2023},
    journal={Available on SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4665715},
}