Baseline Predictions for Stanford Sentiment Treebank (SST-5)

We obtained model predictions for the Stanford Sentiment Treebank benchmark (SST-5) from eight open-source baselines: dictionary-based methods VADER and TextBlob, traditional machine learning methods like logistic regression and support vector machine (SVM), fastText classifier, and deep learning classifiers: BERT and ELMo with Flair and fine-tuned BERT with Hugging Face.