Deep learning techniques and NLG
In this section, we are going build a very simple but intuitive application for NLG. We are going to generate a one-line summary from shot articles. We will see all the details about summarization in this section.
This application took a lot of training time so you can put your model to train on CPU and meanwhile, you can do some other task. If you don't have any other task, then let me give you one.
Exercise
Try to figure out how you can generate a Wikipedia article by just providing some starting character sequences. Don't take me wrong! I'm serious! You seriously need to think on this. This is the dataset that you can use: https://einstein.ai/research/the-wikitext-long-term-dependency-language-modeling-dataset. Jump to the download section and download this dataset named Download WikiText-103 word level (181 MB)
.
(Hint: See this link, https://github.com/kumikokashii/lstm-text-generator.)
Don't worry ;after understanding the concepts of summarization, you can...