Generate-TV-Scripts

Udacity Deep Learning Nanodegree Project #3.

Objective: Generate your own Seinfeld TV scripts using RNNs. The input data was Seinfeld dataset of scripts from 9 seasons. The Neural Network you'll build will generate a new ,"fake" TV script, based on patterns it recognizes in this training data.

What I learnt:

1. Implemening Pre-processing Functions on the dataset:

    * Lookup Table
    * Tokenize Punctuation
   

2. Building the Neural Network:

    * Batching the data
    * Creating dataloaders
    * Initializing RNN model and defining layers
    * Forward propagation of the neural network
    * Initialize the hidden state of an LSTM/GRU
    * Apply forward and back propagation
   

3. Traning the Neural Network:

    * Setting the hyperparameters for optimal loss values.
   

4. Generating new TV scripts using the trained model.