Computer Generated Tweets and Other Prose Using GPT-2

Location image of event venue

Details

Abstract:

Here we will go over how to apply GPT-2, a generative deep
learning model to generate text that mimics an input of a person’s tweets. We will focus on preparing your input text so that it can be understood by the model and also how to make sure you aren’t overfitting. Then we will look at using GPT-2 on Google Colab with GPT2-simple, which is an easy way to use the library that allows you to train, run your model and manage
refined models of GPT-2.

Speaker Bio:

Joshua Herman is a Python Developer at Bank of America working
on software infrastructure for currency trading. He has six years of
experience in software development using various languages such as Python, JavaScript, Java and Scheme. He has also worked in various fields such as Insurance, Computational Population Biology, Distributed Ledgers, CRM and web development. He got his Bachelors in Computer Science from University of Illinois at Chicago in 2012.

6:00 p.m - 6:30 p.m is time for social. Seminar will start at 6:30 p.m.

Our Sponsor: Metis Chicago ( https://www.thisismetis.com/ )