zeryx

9 ms·

61.
Ask HN: Where do you go for discourse?
37 points by zeryx 4y ago|26 comments
62.
by zeryx 4y ago
I know this is a very contentious issue, and I'm by no means an expert. I don't think there are any correct paths, especially when it comes to tax payer expenses, but I can't help feeling like this is giving up. Very crude an
63.
by zeryx 4y ago
Insane that literally no US bank is actually focusing on this. Are retail investors so insignificant that none of them even tries to differentiate and steal customers? You'd think with the insane demand of electric cars, other orgs wou
64.
by zeryx 5y ago
Condos don't count.. unfortunately. Most condo boards don't support installing electric charging infrastructure.
65.
by zeryx 5y ago
Most people in Canada don't have a home, they rent - especially in BC.
66.
by zeryx 5y ago
Good call; yeah that's when we launch :) I agree on a 5 year horizon before it becomes "obvious" that some web3 projects have succeeded. Obviously if you see a crypto project that's cool and interesting, don't just
67.
by zeryx 5y ago
We launch in June, but there's plenty of other projects with builders who are actually building real sustainable businesses and organizations with roadmaps towards profitability as well. They aren't common, to your point the vast
68.
by zeryx 5y ago
As someone who extensively works in crypto (www.fractional.finance) and building web3 technologies - the vast majority of projects are scams, or scams that are trying to fake themselves into legitimacy. Most NFT projects fall into the categ
69.
by zeryx 5y ago
It's incredibly frusturating that this isn't part of the public discourse, where are the pro-nuclear candidates? Which parties are pro-nuclear?
70.
by zeryx 6y ago
My thought; not a scientist and this is all conjecture, but a potential other cause than what other folks are mentioning: Consent conversations and alcohol can be quite confusing and scary to navigate now. This seems to suggest to me, that
71.
by zeryx 6y ago
XPS are great but I would caution about poor thermal management, even after reseating the GPU and CPU cores it doesn't take much to hit the CPU temp limits with full fans.
72.
by zeryx 9y ago
Thanks, it's worth mentioning that we're entirely focused on the dataset for object detection here, but it has a massive classification dataset as well.
73.
by zeryx 9y ago
Yea sorry that was a mobile fix, thanks for catching that. I'll update shortly.
74.
by zeryx 9y ago
I would be lying if I didn't agree with you. The valuable part is that Tensorflow's extendability and functionality are best in class. Making an object detector in scikit-learn is incredibly difficult, same with torch. Keras defin
75.
by zeryx 9y ago
1) The biggest reason is label diversity. COCO is a great dataset, but at 90 labels you're going to be missing some potential. The problem with commercial machine learning development is access to data, and with Open Images and it&#x27
76.
by zeryx 9y ago
So the object detection extension is mostly focused on breaking down complex operations like anchor generation into functional steps that are compatible with the base tensorflow metagraph definition. You can see these builders here: https:
77.
by zeryx 9y ago
Author here, if you have any questions/comments I'll check back in periodically and answer them here!
78.
by zeryx 11y ago
wyldfire: good point I should have brought that up! For my use case (neural network design) there was no divergence between threads so each kernel ran exactly the same path (the for loop is unrolled at runtime by the compiler) but if my if&
79.
How to isolate an algorithm with CUDA(blog.zeryx.me)
2 points by zeryx 11y ago|2 comments