A character-level LSTM is contructed with PyTorch to generate a horror plot on the ocassion of Halloween. The network will train character by character on some text, then generate new text character by character.
The input used to train this model was from a public dataset on kaggle that has more than 15k movie plots. This model will be able to generate new text based on the text from the lyrics!
This network is based off of Andrej Karpathy's post on RNNs and implementation in Torch.
Currently we are trying to deploy it to different platforms. Will be updated soon...
Author: Deepika Rana , Lead: DSC MAIT
Co-authored by: Kaustubh Gupta , ML Lead: DSC MAIT