Model Card for EnvironmentalBERT-extremeweather

Model Description

Using the EnvironmentalBERT-base model as a starting point, the EnvironmentalBERT-extremeweather Language Model is additionally fine-tuned on a 4k train dataset to detect whether a text addresses one or more of the topics between storm, flood, heatwave, drought, wildfire, coldwave.

How to Get Started With the Model

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 = "extreme-weather-impacts/environmentalBERT-extremeweather"
model_name = "extreme-weather-impacts/environmentalBERT-extremeweather"
 
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(tokenizer_name, max_len=512)
 
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer, device=0, top_k=None) # set device=0 to use GPU
 
# See https://huggingface.co/docs/transformers/main_classes/pipelines#transformers.pipeline
print(pipe("Scope 1 emissions are reported here on a like-for-like basis against the 2013 baseline and exclude emissions from additional vehicles used during repairs.", padding=True, truncation=True))
print(pipe("Hurricanes play a significant role in our yearly risk assessment.", padding=True, truncation=True))
print(pipe("Droughts increase the risk of severe wildfires that can additionally damage our crops.", padding=True, truncation=True))

More details can be found in the paper

@article{Schimanski25extremeweatherimpacts,
    title={{What Firms Actually Lose (and Gain) from Extreme Weather Event Impacts}},
    author={Tobias Schimanski and Glen Gostlow and Malte Toetzke and Markus Leippold},
    year={2025},
    journal={Soom available on SSRN},
}