The React Toxicity Classifier is importing a pre-trained model from tensorFlow.js.
The project was bootstrapped with Create React App.
The pre-trained model is a toxicity classifier that will detect if the text you write in the input field is classified within 7 different labels:
| Labels |
|---|
| toxic |
| severe_toxic |
| obscene |
| threat |
| insult |
| identity_hate |
The toxicity classifier will return a true (is toxic) or false (not toxic) depending on the input sentence.
- The sentence in the input field is submitted through a button and then sent to the pre-trained model.
- The input field can be cleared through a button.
- For every sentence that is labelled true (within the 7 different labels) a toxic point will be added underneath the input field.
- There is a button that can clear the toxic points.
A submitted sentence will be displayed in the table with the following information:
- How much % of toxicity the sentence contains
- An emoji that will switch between a happy one/love eyes (not toxic) and a skull (toxic)
- Toxic points that will increase every time the input sentence is labelled “true” (toxic)
- A button to delete one sentence at a time
- A button to delete all sentences from the table
At the top of the page the following links are displayed:
- Source code
- Dataset
- Toxicity classifier
- Tensorflow.js
- Contact information: E-mail, LinkedIn and GitHub
- Add requirements
- Implement the pre-trained toxicity classifier in the code
- Add React
You can learn more in the Create React App documentation.
To learn React, check out the React documentation.