In this project, I built a RNN based mini-app that generates new 'fake' Seinfeld TV series scripts based on the patterns it 'learns' from the training data. My Recurent Neural Network is trained on (part of) the Seinfeld dataset of scripts from 9 seasons.
The main artefacts are:
-dlnd_tv_script_generation.ipynb : Jupuyter notebook that goes through the process, including:
- Importing Datasets
- Pre-processing data
- Building the RNN
- Selecting hyper-parameters
- Training the RNN
- Generating TV scripts.