metadata
license: mit
pipeline_tag: token-classification
Description:
climateattention-ctw classifies if a given sequence is related to climate topics. As a fine-tuned classifier based on climatebert/distilroberta-base-climate-f (Webersinke et al., 2021), it is using the following ClimaText dataset (Varini et al., 2020):
- Wiki-doc corpus, with 115847 samples (57922 positives, 57925 negatives)
For company disclosures or news articles you might want to check the 10k model: kruthof/climateattention-10k-upscaled
How to use:
from transformers import AutoTokenizer, pipeline,RobertaForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("climatebert/distilroberta-base-climate-f")
climateattention = RobertaForSequenceClassification.from_pretrained('kruthof/climateattention-ctw',num_labels=2)
ClimateAttention = pipeline("text-classification", model=climateattention, tokenizer=tokenizer)
ClimateAttention('Emissions have increased during the last several months')
>> [{'label': 'Yes', 'score': 0.9993829727172852}]
Performance:
Performance tested on the balanced ClimaText Wiki-doc test set, featuring 3826 samples (1913 positives, 1913 negatives) (Varini et al., 2020)
| Accuracy | Precision | Recall | F1 |
|---|---|---|---|
| 0.8834 | 0.8717 | 0.8991 | 0.8852 |
# References:
Varini, F. S., Boyd-Graber, J., Ciaramita, M., & Leippold, M. (2020).
ClimaText: A dataset for climate change topic detection. arXiv preprint arXiv:2012.00483.
Webersinke, N., Kraus, M., Bingler, J. A., & Leippold, M. (2021).
Climatebert: A pretrained language model for climate-related text. arXiv preprint arXiv:2110.12010.
------------------------------
https://kruthof.github.io