Gospel of LSTM : How I wrote 5th Gospel of Bible using LSTMs

Sleeba Paul (~sleeba)




Language modeling was a complex task of previous days. But advancements in Deep Learning has solved this problem very effectively. Using Recurrent Neural Networks architecture, I've built a language model which can effectively generate the fifth gospel of bible by learning from four existing gospels. This model is also able to divide verses and chapters along with meaningful passages.


  1. Recurrent Neural Networks basics
  2. Deep learning basics
  3. Language modeling basics
  4. Familiarization with PyTorch

Content URLs:

GitHub Repo: https://github.com/sleebapaul/gospel_of_rnn.git

My blog post on Language Modelling using RNNs Part-1: https://sleebapaul.github.io/rnn-tutorial/

My blog post on Language Modelling using RNNs Par-2: https://sleebapaul.github.io/rnn-tutorial-2/

Google Colab Notebook: https://drive.google.com/file/d/12pEy-aOS0_PiVkFgxyINmBbtuvB5TqV5/view?usp=sharing

Speaker Info:

Sleeba Paul is a Power System graduate and published researcher who loves intelligent machines. He currently works as a Machine Learning Engineer at Refly; an AI startup in India where he works on content enhancement and video analytics.

Speaker Links:

Personal website: http://sleebapaul.github.io/

LinkedIn: https://www.linkedin.com/in/sleebapaul/

Github: https://github.com/sleebapaul

Id: 803
Section: Data science
Type: Talks
Target Audience: Intermediate
Last Updated: