In this file, you will find a multi-class text classification model based on the combination of extracted keywords and text using the Bert model and the Tensorflow library.
Furthermore, using the stremalite library, you can build a simple and easy interface and deploy your model without any knowledge of web development.
In our simple web application using streamlite, we have implemented various easy-to-use techniques to allow the user to check the most relevant topic of a text, get its keywords, make a summary and obtain a basic paraphrase with the help of transformers and hugging face models.