Movie Script Database generator & Applications
This collection of scripts allows you to collect thousands of real movie scripts with their genres from IMSDb using their RSS feed and a little bit of HTML scraping
from acquire_script import load_scripts
#load scripts into ./data folder as JSON
load_scripts()
To illustrate a few applications of the Database, I've written two (as of today) Machine-Learing systems:
The classifier is written in Python using the Tensorflow framework. It is able to classify the scripts into the provided categories (based on the first 1000 words). (can be found here, it incluides loading the dataset into a Tensorflow dataset)
The generator is still in progress and is currently not able to produce correct text output due to difficulties generating that with the standard DCGAN approach. Those two approaches are located