GuiltySpark/amikus_text_summarizer

Check out the configuration reference at https://huggingface.co/docs/hub/spaces#reference

Readme

For this application I have created three features: 1.sumarizer 2.keyword extraction

1. 1.paraphraze text

1.Sumarizer: for this I have used transformers library, sumarization pipeline ,to sumarize the text (see sumarize fuction) the parameters are maxlength = 200, minlenth = 30

1. 1.keyword extraction: this feature extracts 20 keyword from the text each of length of 3 or less tha 3 words, see the same sumarize function the lib I have used for this is yake and its function KeywordExtractor 3.paraphraze text: for this i have used PEGASUS from transformers library.I used senstence-splitter to split the text into a list of sentences. Then paraphrazed each of the sentences and then compiled them back into a paragraph. I am using a pre trained model = 'tuner007/pegasusparaphrase' for this.

The selection of example texts was based on my own areas of fascination like robotics, ancient history and astronomy.