I want to use your code for finding the difference between two (training) text documents.
Kripa Ghosh

If the data has images of the documents, and you want to distinguish the writing, its possible to do so. 
If you want to do document similarity (which is an NLP task), you will need a totally different architecture. Perhaps you can try an LSTM with the same siamese structure

Like what you read? Give Harshvardhan Gupta a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.