An Absolute Beginner's Guide to Deep Learning with Keras | Dr. Brian Spiering @ PyBay2018

This talk was presented at PyBay2018 - the Bay Area Regional Python conference. See pybay.com for more details about PyBay and click SHOW MORE for more information about this talk
Abstract
Interest in deep learning and building artificial intelligence (AI) based applications has been growing in the past few years. The Keras library makes these techniques accessible by offering a high-level API capable of running on top of TensorFlow, CNTK, or Theano. Similar to Python in general, Keras puts user experience front and center by having an API designed for human beings (not machines). This makes Keras perfect for easy and fast prototyping. Keras also highly modular and easy to extend to more complex deep learning models. This talk will be an introduction to neural networks and the Keras API for building them. We'll walk step-by-step how to build a convolutional neural net (CNN).
Slides: goo.gl/qJ9i55
Speaker Bio
Dr. Brian Spiering is a Professor of Computer Science at the University of San Francisco and freelance consultant. He teaches humans the languages of computers (primarily Python) and teaches computers the languages of humans (through Natural Language Processing and Artificial Intelligence). He is active in the San Francisco tech community through volunteering and mentoring.
This and other PyBay2018 videos are brought to you by our Gold Sponsor Cisco!

[@RD-yv4cc]

Excellent introduction to keras. Really boosted my understanding to say the least!!

[@RiceBaiiKhao]

Amazing. I am currently learning exactly this in school and this man Dr. Ryan Spiering is elegantly explaining every single thing that I did not precisely understand. Also, great questions from the crowd. Just wow. Thank you.

[@Mal-nf2sp]

This guy is fantastic at teaching.

[@ravikumar-vr6zm]

awesome tutorial.......thanks

[@dirtymopwater2868]

Great talk

[@dompower500]

Excellent sir

[@simarkapoor4119]

the first professor who actually told why we need GPUs lol.

[@robertcohn8858]

Good if you know absolutely nothing about NNs. The keras part is somewhat worth seeing so you can get the flavor of what can be done - but it's a very cursory demonstration without giving the viewer a lot of background knowledge (which is understandable because of time constraints).

[@jcwfh]

The uptalk here is mind-numbing.

[@WahranRai]

forget to mention the bias and why we use it in neural network

[@nupurgautam1]

any one else counted occurrence of "what we gonna do?" in this lecture?

[@alkodjdjd]

my thought too

[@atomatopia1]

I counted "notion"

[@Ruhgtfo]

emm why used 512 units for the layers?

[@samerfk]

just a random number. you could increase/decrease the number if you wish until you find optimal/better performance

[@PeterRetief]

Tiny little mouse cursor!

[@oneboy8183]

good but slow down lol