cardiffnlp/twitter-roberta-base-2021-124m-topic-multi
This model is a fine-tuned version of cardiffnlp/twitter-roberta-base-2021-124m on the
cardiffnlp/tweet_topic_multi
via tweetnlp
.
Training split is train_all
and parameters have been tuned on the validation split validation_2021
.
Following metrics are achieved on the test split test_2021
(link).
- F1 (micro): 0.7528230865746549
- F1 (macro): 0.5564228688431104
- Accuracy: 0.535437760571769
Usage
Install tweetnlp via pip.
pip install tweetnlp
Load the model in python.
import tweetnlp
model = tweetnlp.Classifier("cardiffnlp/twitter-roberta-base-2021-124m-topic-multi", max_length=128)
model.predict('Get the all-analog Classic Vinyl Edition of "Takin Off" Album from {@herbiehancock@} via {@bluenoterecords@} link below {{URL}}')
Reference
@inproceedings{camacho-collados-etal-2022-tweetnlp,
title={{T}weet{NLP}: {C}utting-{E}dge {N}atural {L}anguage {P}rocessing for {S}ocial {M}edia},
author={Camacho-Collados, Jose and Rezaee, Kiamehr and Riahi, Talayeh and Ushio, Asahi and Loureiro, Daniel and Antypas, Dimosthenis and Boisson, Joanne and Espinosa-Anke, Luis and Liu, Fangyu and Mart{'\i}nez-C{'a}mara, Eugenio and others},
author = "Ushio, Asahi and
Camacho-Collados, Jose",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
month = nov,
year = "2022",
address = "Abu Dhabi, U.A.E.",
publisher = "Association for Computational Linguistics",
}
- Downloads last month
- 143
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Dataset used to train cardiffnlp/twitter-roberta-base-2021-124m-topic-multi
Evaluation results
- Micro F1 (cardiffnlp/tweet_topic_multi) on cardiffnlp/tweet_topic_multiself-reported0.753
- Macro F1 (cardiffnlp/tweet_topic_multi) on cardiffnlp/tweet_topic_multiself-reported0.556
- Accuracy (cardiffnlp/tweet_topic_multi) on cardiffnlp/tweet_topic_multiself-reported0.535