55 lines
1.8 KiB
Markdown
55 lines
1.8 KiB
Markdown
---
|
|
language: "en"
|
|
tags:
|
|
- sentiment
|
|
- emotion
|
|
- twitter
|
|
|
|
widget:
|
|
- text: "Oh wow. I didn't know that."
|
|
- text: "This movie always makes me cry.."
|
|
|
|
---
|
|
|
|
## Description
|
|
|
|
With this model, you can classify emotions in English text data. The model was trained on 6 diverse datasets and predicts 7 emotions:
|
|
|
|
1) anger
|
|
2) disgust
|
|
3) fear
|
|
4) joy
|
|
5) neutral
|
|
6) sadness
|
|
7) surprise
|
|
|
|
The model is a fine-tuned checkpoint of DistilRoBERTa-base.
|
|
|
|
## Application
|
|
|
|
a) Run emotion model with 3 lines of code on single text example using Hugging Face's pipeline command on Google Colab:
|
|
|
|
[](https://colab.research.google.com/github/j-hartmann/emotion-english-distilroberta-base/blob/main/simple_emotion_pipeline.ipynb)
|
|
|
|
b) Run emotion model on multiple examples and full datasets (e.g., .csv files) on Google Colab:
|
|
|
|
[](https://colab.research.google.com/github/j-hartmann/emotion-english-distilroberta-base/blob/main/emotion_prediction_example.ipynb)
|
|
|
|
## Contact
|
|
|
|
Please reach out to jochen.hartmann@uni-hamburg.de if you have any questions or feedback.
|
|
|
|
Thanks to Samuel Domdey and chrsiebert for their support in making this model available.
|
|
|
|
## Appendix
|
|
|
|
Please find an overview of the datasets used for training below. The table summarizes which emotions are available in each of the datasets.
|
|
|
|
|Name|anger|disgust|fear|joy|neutral|sadness|surprise|
|
|
|---|---|---|---|---|---|---|---|
|
|
|Crowdflower (2016)|Yes|-|-|Yes|Yes|Yes|Yes|
|
|
|Emotion Dataset, Elvis et al. (2018)|Yes|Yes|Yes|Yes|-|Yes|Yes|
|
|
|GoEmotions, Demszky et al. (2020)|Yes|Yes|Yes|Yes|Yes|Yes|Yes|
|
|
|ISEAR, Vikash (2018)|Yes|Yes|Yes|Yes|-|Yes|-|
|
|
|MELD, Poria et al. (2019)|Yes|Yes|Yes|Yes|Yes|Yes|Yes|
|
|
|SemEval-18|Yes|-|Yes|Yes|-|Yes|-| |