Today I took a Deep Dream Workshop as part of this weekend's Gray Area Festival. I learned a lot of the principles back in grad school and I've been wanting to get my hands dirty with the code ever since I saw @mtyka's work last summer.

Turns out this material is kind of gnarly! Just installing the required code and libraries is highly non-trivial and I have major respect to Erin the instructor for helping us all get through it. Here are her notes about how to install and run the Deep Dream code.

it took me most of the workshop to get the installation up and running but it was pretty smooth (if not particularly speedy) when it finally worked. Here's my Deep Dream Hello World:

So I did not get far into the code but I've already learned some iPython tricks and look forward to poking at it more in the future. While I'm here, I'm going to do a small linkdump of neural network resources that are great for learning more:

• Chris Olah's blog post about visualizing the topology of the NN transform
• Google's amazing TensorFlow neural training demonstration
• The "Inceptionism" blog post that started it all
• Gatys et al. "Style Transfer" paper http://arxiv.org/abs/1508.06576