This is how everyone picks raffle winners, right?

Congrats to the first Welch Labs Patreon quarterly book raffle winner! In the mail to you today is a copy of one of my favorite books of all time, including some annotations around my favorite passages. Enjoy!

The HackathonCLT winning team, Welch Labs, created a heatmap of #MecklenburgCounty for each of MedLink Mecklenburg six clinics to represent where patients were coming from, and in turn where the highest number of disease cases originated. ow.ly/yOhz30oektI

Woah, that’s fun!

Looking forward to this one!

Working on a new series on GANs, thought it would be fun to see if I could map a video of me to some clips of @CaseyNeistat - here's the results for far. Pretty crazy, right!? #NeuralNetworks #DeepLearning

This was fun, thanks to Sharon Smith for the invitation!

Really like this #PyTorch tutorial, nicely builds a simple neural network from pure python into pytorch piece by piece. pytorch.org/tutorials/begi…

Really looking forward to catching up on these when I have some time! The animations look incredible.

So learning about #GANs has been really tough, but I at least have some cool pictures to show for it. :)

First attempt at removing cars off the roads with neural nets. Will have to dream harder.

Conclusions after a couple weeks with fast.ai: the library is under-documented and buggy, but delivers astonishingly good performance, quickly. I’ve achieved better performance in 3 days in #fastai than I did after working on the same problem in #keras for ~1 month.

I had a great time chatting with Kirill Eremenko on the SuperDataScience podcast about #NeuralNetworks, #SelfDrivingCars, and #ComputerVision, available here: superdatascience.com/podcast/buildi…. Thanks for the invitation!

Perspective matters.

Moving in the age direction of the latent space of #StyleGan gets a little creepy. Using nice repo from Dmitry Nikitko: github.com/Puzer.

Looking forward to teaching the Deep Learning workshop at Data Science North Carolina later this month. Hoping to include some new material on GANs! dsncconf.com/deep-learning/

Awesome!

I'm doing a TEDx talk at Berkeley this February. If you want to come, they gave one of those fancy discount codes for my followers (i.e. you): SPEAKER35 Hope to see you there!

There’s no safety in pure empiricism.

I’m hearing comments that Grid AI (Lightning) seem to have copied fastai's API without credit, and claimed to have invented it. We wrote a paper about our design; it's great it's inspiring others. Claiming credit for other's work? NOT great 1/ mdpi.com/2078-2489/11/2…