cardiffnlp/twitter-roberta-base-dec2021-tweet-topic-multi-all

This model is a fine-tuned version of cardiffnlp/twitter-roberta-base-dec2021 on the tweet_topic_multi. This model is fine-tuned on train_all split and validated on test_2021 split of tweet_topic. Fine-tuning script can be found here. It achieves the following results on the test_2021 set:

• F1 (micro): 0.7647668393782383
• F1 (macro): 0.6187022581213811
• Accuracy: 0.5485407980941036

Usage

import math
import torch
from transformers import AutoModelForSequenceClassification, AutoTokenizer

def sigmoid(x):
 return 1 / (1 + math.exp(-x))
 
tokenizer = AutoTokenizer.from_pretrained("cardiffnlp/twitter-roberta-base-dec2021-tweet-topic-multi-all")
model = AutoModelForSequenceClassification.from_pretrained("cardiffnlp/twitter-roberta-base-dec2021-tweet-topic-multi-all", problem_type="multi_label_classification")
model.eval()
class_mapping = model.config.id2label

with torch.no_grad():
 text = #NewVideo Cray Dollas- Water- Ft. Charlie Rose- (Official Music Video)- {{URL}} via {@YouTube@} #watchandlearn {{USERNAME}}
 tokens = tokenizer(text, return_tensors='pt')
 output = model(**tokens)
 flags = [sigmoid(s) > 0.5 for s in output[0][0].detach().tolist()]
 topic = [class_mapping[n] for n, i in enumerate(flags) if i]
print(topic)

Reference

@inproceedings{dimosthenis-etal-2022-twitter,
 title = "{T}witter {T}opic {C}lassification",
 author = "Antypas, Dimosthenis and
 Ushio, Asahi and
 Camacho-Collados, Jose and
 Neves, Leonardo and
 Silva, Vitor and
 Barbieri, Francesco",
 booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
 month = oct,
 year = "2022",
 address = "Gyeongju, Republic of Korea",
 publisher = "International Committee on Computational Linguistics"
}