LSTM-NumPy

Implementation of LSTM by only using NumPy

This is a Stanford deep learning exercise - Shakespeare as input.
Task: Try to predict the next character given a sequence of text.

Results

after ~30 min of training on 1 cpu core

 you? she what, boon how to same a call viness:
You anstion. 'twirt steers venolions,
And and the orentiop, and if pirtere, froth of injul.
Why fiffer the may a featelf;
The formuncbnce I know, upore c 

here you this aras loot leave'd his danth;
And mejely, God offory to omer, of time tein all sir.

CLANES:
The preater hear's mores of Prrean.
Coflipe thhire in hath s pefuly,
On entay and preet nom ho