by Joohyun Park

What is happniess? Human asks, machine answers.

This project uses LSTM networks a with pre-trained model on a corpus of 100,000 crowd-sourced happy moments from HappyDB. You can see the visualization of original dataset here. The temperature controls the randomness of the output. A temperature of 0 will be relatively random but might not even look like English, while a temperature of 1.0 will probably be correct English but will also be very close to the original data, perhaps even straight quotations.

seed text:

temperature: 0.5

Loading Model