I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
1–10 of 18 posts
Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#2That's pretty neat. If you're interested, the creator[0] of the real NotHotdog app posted some details here on HN[1], apparently a blog post will follow.
[0] https://twitter.com/timanglade [1] https://news.ycombinator.com/item?id=14347865
Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#3is a hotdog a sandwich?
Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#4is a hotdog a sandwich?
Err—good question. No, not really, but it's difficult to find a reason why it's not. It's meat between bread.
Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#5is a hotdog a sandwich?
What else would it be? A pastry? I think because the elements (hot dog and bun) are cooked separately it has to be a sandwich. If you baked the hotdog within the bun and severed it then it would be a pastry to me.
Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#6Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#7If this last week's episode does one good thing it will be to convince programmers that it's not that hard to get started doing ML stuff. Not to say that it's all trivial or straightforward, if you look at the linked image below you can see that a simple NN will miss lots of stuff. In this example the image is being examined for hotdogness as a whole, not being segmented or searching for a bounding box of hotdogness.
Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#8is a hotdog a sandwich?
In my vernacular, a hotdog sandwich is a bit different preparation than a regular hotdog.
You slice the hotdog meat lengthwise so it will sit flat, then in half so that you have 4 pieces that fit onto regular sliced bread.
https://i.imgur.com/5XbzXOr_d.jpg?maxwidth=640&shape=thumb&f...
Note that is usually done not because it's a better idea. Driven by either lack of funds, or the idiotic practice where buns come in packs of 8, but hotdogs come in packs of 10.
Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#9If this last week's episode does one good thing it will be to convince programmers that it's not that hard to get started doing ML stuff. Not to say that it's all trivial or straightforward, if you look at the linked image below you can see that a simple NN will miss lots of stuff. In this example the image is being examined for hotdogness as a whole, not being segmented or searching for a bounding box of hotdogness.…
I think your link proves that the NN is smarter than you give it credit for. The image is of a laptop screen, so NOT HOTDOG ;)
Re: I Build My Own Not-Hotdog Classifier from HBO's Silicon Valley
#10If this last week's episode does one good thing it will be to convince programmers that it's not that hard to get started doing ML stuff. Not to say that it's all trivial or straightforward, if you look at the linked image below you can see that a simple NN will miss lots of stuff. In this example the image is being examined for hotdogness as a whole, not being segmented or searching for a bounding box of hotdogness.…
Good luck with the wedding! :)