Instructions to use launch/POLITICS with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use launch/POLITICS with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="launch/POLITICS")# Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("launch/POLITICS") model = AutoModelForMaskedLM.from_pretrained("launch/POLITICS") - Notebooks
- Google Colab
- Kaggle
Fine-tuned POLITICS for political leaning classification
#5
by matous-volf - opened
Hi! Just wanted to let you know that we have fine-tuned this model for classifying political leaning (left, center, right), producing our political-leaning-politics. From our research (described in the paper), it is obvious that your work has paid off, as POLITICS consistently scores the best among the benchmarked pretrained models (BERT, RoBERTa, DeBERTa) of the same size, accross all our experiments. :)
We have also produced a second, larger model based on DeBERTa large for the same task.