twitter masked-token-prediction bertweet election2020 politics

Pre-trained BERT on Twitter US Political Election 2020

Pre-trained weights for PoliBERTweet: A Pre-trained Language Model for Analyzing Political Content on Twitter, LREC 2022.

Please see the official repository for more detail.

We use the initialized weights from BERTweet or vinai/bertweet-base.

Training Data

This model is pre-trained on over 83 million English tweets about the 2020 US Presidential Election.

Training Objective

This model is initialized with BERTweet and trained with an MLM objective.

Usage

This pre-trained language model can be fine-tunned to any downstream task (e.g. classification).

from transformers import AutoModel, AutoTokenizer, pipeline
import torch

# choose GPU if available
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

# select mode path here
pretrained_LM_path = "kornosk/polibertweet-mlm"

# load model
tokenizer = AutoTokenizer.from_pretrained(pretrained_LM_path)
model = AutoModel.from_pretrained(pretrained_LM_path)

# fill mask
example = "Trump is the <mask> of USA"
fill_mask = pipeline('fill-mask', model=pretrained_LM_path, tokenizer=tokenizer)

outputs = fill_mask(example)
print(outputs)

# see embeddings
inputs = tokenizer(example, return_tensors="pt")
outputs = model(**inputs)
print(outputs)

# OR you can use this model to train on your downstream task!
# please consider citing our paper if you feel this is useful :)

Reference

Citation

@inproceedings{kawintiranon2022polibertweet,
  title     = {PoliBERTweet: A Pre-trained Language Model for Analyzing Political Content on Twitter},
  author    = {Kawintiranon, Kornraphop and Singh, Lisa},
  booktitle = {Proceedings of the Language Resources and Evaluation Conference},
  year      = {2022},
  publisher = {European Language Resources Association}
}