zo7

11 ms·

61.
by zo7 10y ago
I wonder if taking a hot bath can have the same effect. I've been taking them more frequently recently and I've definitely noticed that my heart rate goes up (and that I smell a bit different than after taking a shower, presumably
62.
by zo7 10y ago
Also by this group: "Doggiewoggiez! Poochiewoochiez!", a remake of Holy Mountain made entirely out of found footage of dog films. It's really astounding, especially when you consider how much effort it took to go through all
63.
by zo7 10y ago
Maybe the fact that RocketAI didn't have a product, wasn't solving any problem, or was touting a bogus algorithm with absolutely zero details should've set off some flags. They didn't base their interest off of any value
64.
by zo7 10y ago
I've heard about this a lot from people I know who work in healthcare. It seems that one could make a successful business simply by hiring a bunch of data scientists to offer analytics and data processing services for healthcare, but w
65.
by zo7 10y ago
One way to do PCA is using SVD to find a transformation matrix of eigenvectors to project your data with, so they're similar.
66.
by zo7 10y ago
Is it possible to swap the link on this post? That's kind of silly that they just copy the original post. Understanding different optimizers and why they work is important though, especially since often people will just use them withou
67.
by zo7 10y ago
My intuition is that these patterns "hijack" the ReLU activations in the lower levels, causing either important features to not fire or features that shouldn't fire to do so. Usually the lower layers learn very primitive sh
68.
by zo7 10y ago
I would imagine you could print the pattern onto a film that you stick on the lens of a camera to throw off its classifier.
69.
by zo7 10y ago
What the OP described would be extremely low bandwidth -- you only need three parameters to describe what to draw per keyframe (maybe one or two more if you want to describe mixtures of emotions or identities or more complex things), wher
70.
by zo7 10y ago
Yeah, Keras especially has a really intuitive and easy-to-use api, so definitely start there if you want to do more with deep learning. It's almost too easy, really...
71.
by zo7 10y ago
Possibly, but one thing the article doesn't address is the computational costs of doing the convolution. Because you have to store an intermediate representation of the image for every convolution kernel that you use, you start to eat
72.
Generating Faces with Deconvolution Networks(zo7.github.io)
172 points by zo7 10y ago|20 comments
73.
Modeling Alvin Lucier's “I Am Sitting in a Room”(zo7.github.io)
1 points by zo7 10y ago|0 comments
74.
Creating Ghostly Videos with Temporal Median Filters(zo7.github.io)
2 points by zo7 10y ago|0 comments
75.
by zo7 10y ago
Ah, so I guess he'd approve of the way I stumbled through university :) I get what he's saying, but I can't imagine adhering strictly to this would be healthy. I've waited until the last minute to finish enough projects
76.
by zo7 10y ago
I was going to say something similar but this is a really good analogy. One thing I might say differently however is that most people on Reddit don't come to the mall to go to their favorite shops, they come to the mall to come to the
77.
by zo7 10y ago
Yeah, after that I told him (as calmly as I could coming off the adrenaline from almost getting hit) something to the effect of "You have your kids with you, I'm just trying to get home like you and I'm required to ride on th
78.
by zo7 10y ago
Damn... I can't imagine the mindset people need to get that angry at people passively going their own way.
79.
by zo7 10y ago
I don't think non-cyclists realize how much risk you take on when you ride in the city. In my town we really only have bike lanes on the two main roads that go up and down the bulk of the city, and they ride dangerously close to lines
80.
by zo7 10y ago
That is beautiful! I didn't realize you could pipe into OSX's `open` util. (or how big the man page for bash really is!) Here's zsh: https://drive.google.com/file/d/0B_jWp8d11aB5Wi15eVdJS1Nhdk0...
81.
by zo7 10y ago
I'm just glad that they think I'm stable!
82.
by zo7 10y ago
My name is Flynn, reading about all of this is endlessly entertaining :)
83.
by zo7 10y ago
This looks fantastic and is something I've been wanting to make myself for Keras! (although this way better implemented than I had imagined) I have a model to finish and start training tonight, will definitely try this out.
84.
by zo7 10y ago
Yeah you can't really be patronizing to people who prefer Apple products when you're contrasting a consumer tech company with a genocidal fascist regime.
85.
by zo7 11y ago
The title makes better sense for the original movie, where dummy cyborg bodies (shells) are taken over by a rogue AI. The movie asks what role our biological consciousness plays when our lives are so intertwined with technology (the cyborgs
86.
by zo7 11y ago
> Personally, I really do believe the throttling here is not intended behavior of Binge On. I honestly do. Could you elaborate on this? I wasn't aware about Binge On before, but going just off this incident it appeared to me like de
87.
by zo7 11y ago
I'm surprised that there's so much "I don't get it" sentiment here. You would think that everyone on HN would be clamoring to figure out how Snapchat is able to get as many users as it has.
88.
by zo7 11y ago
Just to clarify, the "social dick-measuring contest" the author was referring to is being able to see how many followers or friend a user has. I can look up how many followers some celebrity has on Twitter or compare how many frie
89.
by zo7 11y ago
From Snapchat's privacy policy, "Snapchat captures what it’s like to live in the moment. So in many cases the messages sent through our services are automatically deleted from our servers once we detect that they have been viewed
90.
by zo7 11y ago
Oh sorry! I completely misunderstood what DetectIndent was doing, that's a good tip. I did want to point out that there's some sense to forcing the code to look the same on every machine though. Of course converting tabs/spac

More ›