The main idea of this project is to classify memes on the basis of the offensive content present in the meme.
We are not Considering the OCR(image to text part) rather only using the text content present in the meme.
Fasttext is used in one of the version:
https://github.com/facebookresearch/fastText
The pre trained vector file that is used is : wiki-news-300D-1M.vec, this can be found in https://fasttext.cc/docs/en/english-vectors.html