Investigating different text generation methods using Fastai language model on the IMDB dataset.
This repo includes the examples shown and the implementations of the methods described in my blog post "Can We Generate High-Quality Movie Reviews Using Language Models?", link will be added here shortly.
Notebook can be run as-is - the first half of the notebook fine-tunes the Wikipedia language model on the IMDB dataset, and so a GPU is highly recommended (this step took ~4 hours on my mid-range GPU). After this step is performed for the first time, the relevant code can be safely commented out as detailed in the notebook and you can experiment with text generation freely!