This repository contains the code used for the baselines in the paper Happy Dance, Slow Clap: Using Reaction GIFs to Predict Induced Affect on Twitter
Included are the following Python files:
sentiment.py
: code for majority, LR, and RoBERTa for sentiment binary classificationreactions.py
: code for majority, LR, and RoBERTa for reaction multi-class classificationemotions.py
: code for majority, LR, and RoBERTa for emotions multi-label classificationCNN.py
: CNN code for sentiment, reaction, and emotions classification
In addition, the reaction to emotions mappings are available in:
Reactions2GoEmotions.csv