This a testing version from the original repository: https://github.com/karpathy/makemore
We are making the full project step by step in the ipynb file for better understranding.
We only see what letter might come after another letter. Used a file with BAngla names. The database is not much sophisticated. Need to refine it later.
We first made a statistical model based on counting the occurences of letter pairs. Then we made a neural network model that learns to do the same thing.