PyData Cambridge - 36th Meetup

We are happy to announce the 36th PyData Cambridge meetup!

We are looking for more volunteers to help running our meetup.

Our code of conduct has always reflected the diversity we want in our meetup. To continue this and improve our meetup this has to start by changing things at the top. So we especially would like to welcome organisers coming from a more diverse background.

Interested? Please get in touch with us on meetup or our email:
pydatacambridge@gmail.com

Agenda

18:45 - Doors open
19:00 - Introduction
19:15 - "The 7 lines of code you need to run faster real-time inference" - Adrian Boguszewski (Intel)
19:50 - Interval
20:15 - "Making Text-to-Pokemon using Deep Learning" - Justin Pinkney (Lambda Labs)
20:50 - End (Pub - Old Ticket Office, Station Square)

Talk: The 7 lines of code you need to run faster real-time inference

You've already trained your great neural network. It reaches 99.9% of accuracy and saves the world, so you would like to deploy it. However, it must run in real time and process data locally, and you don't want to build a web API. After all, you are a Data Scientist, not a Web Developer… So, is it possible to automatically optimize and run the network fast on the local hardware you have, not the hardware you wish you had? Absolutely! During the talk, I'll present the OpenVINO Toolkit. You'll learn how to automatically convert the model using Model Optimizer and run the inference with the Runtime. The magic with only seven lines of code. After all, you'll get a step-by-step jupyter notebook to try at home.

Bio: AI Software Evangelist at Intel. Adrian graduated from the Gdansk University of Technology in the field of Computer Science 6 years ago. After that, he started his career in computer vision and deep learning. For the previous two years, as a team leader of data scientists and Android developers, Adrian was responsible for an application to take a professional photo (for an ID card or passport) without leaving home. He is a co-author of the LandCover.ai dataset and he was teaching people how to do deep learning. His current role is to educate people about OpenVINO Toolkit. In his free time, he’s a traveller. You can also talk with him about finance, especially savings and investments.

Talk: "Making Text-to-Pokemon using Deep Learning"

Put in text, get out Pokemon! Recent advances in deep learning have made it possible for people to generate almost any image they can describe, using text to image models like DALLE2 and Stable Diffusion. Stable Diffusion in particular is open to everyone and with a bit of Python knowledge and a GPU you can build on it to make amazing things! Justin will talk about how he trained a "Text-to-Pokemon" generator using open source tools and less than £10 of GPU compute, he'll explain how these models work, how we got here and what the future might hold. There'll be a live demo on the day and in the meantime you can try out the model here: https://huggingface.co/spaces/lambdalabs/text-to-pokemon

Bio: Justin is a Senior Machine Learning Researcher at Lambda Labs specialising in image generation and editing applications using the latest generative methods from GANs to Diffusion models. As well as research he also applies these methods to real world problems as part of Lambda's consulting team, working with diverse clients from VFX studios to consumer apps.

Code of Conduct
PyData is dedicated to providing a harassment-free event experience for everyone, regardless of gender, sexual orientation, gender identity, and expression, disability, physical appearance, body size, race, or religion. We do not tolerate harassment of participants in any form.
The PyData Code of Conduct governs this meetup. ( http://pydata.org/code-of-conduct.html ) To discuss any issues or concerns relating to the code of conduct or the behaviour of anyone at a PyData meetup, please contact NumFOCUS Executive Director Leah Silen (leah@numfocus.org) or organizers.