Ameo

10 ms·

91.
by Ameo 4y ago
> Or perhaps hardware will catch up before. I feel like that's been the pretty consistent lesson in computing over the past decades. New technologies start out as expensive, exotic, and specialized and become cheap and commonplace
92.
by Ameo 4y ago
So much dev time goes up in smoke because of pedantic and overzealous linters, formatters, etc. in PR checks and CI pipelines. Your whole PR gets held up because of a style nit like an unused import or trailing whitespace. That forces you
93.
by Ameo 4y ago
I recently came across a very interesting history of the early evolution of the Unix OS and some of these core utils during the 1970s and early 1980s: https://www.cs.dartmouth.edu/~doug/reader.pdf Lots of cool details
94.
by Ameo 4y ago
I think that the activation function definitely is important for this particular case, but it's not actually periodic; it saturates (with a configurable amount of leakyness) on both ends. The periodic behavior happens due to patterns
95.
by Ameo 4y ago
Oh nice catch on the typo - thanks!!
96.
Reverse engineering a neural network's clever solution to binary addition(cprimozic.net)
562 points by Ameo 4y ago|151 comments
97.
by Ameo 4y ago
Huge fan of just; I add a Justfile to pretty muc hevery new repo I create regardless of language or stack. My personal favorite feature is the ability to load environment variables from a `.env` file and set them for all commands run. Just
98.
by Ameo 4y ago
Networking is the place where I notice how tall modern stacks are getting the most. Debugging networking issues inside of Kubernetes feels like searching for a needle in a haystack. There are so, so many layers of proxies, sidecars, ingress
99.
by Ameo 4y ago
> Roughly speaking, the complexity of an object is the length of its shortest description. This has a ton of really interesting implications. Even though Kolmogorov complexity is technically incomputable due to the halting problem, you c
100.
by Ameo 4y ago
Hello, author here! Some helpeful people on Reddit pointed out that there’s actually an even easier way to achieve this functionality that makes use of built-in depth test functionality with OpenGL/WebGL: https://old.reddit.
101.
Google Helpful Content Update- More Content by People, for People in Search(developers.google.com)
4 points by Ameo 4y ago|0 comments
102.
by Ameo 4y ago
Thank you - and ty for checking out the post!
103.
Logic Through the Lens of Neural Networks(cprimozic.net)
11 points by Ameo 4y ago|3 comments
104.
by Ameo 4y ago
Film grain is similar to noise which by definition compresses extremely poorly. By actually modeling the way the film grain is produced, they can avoid sending what is essentially random data over the wire while still preserving the visual
105.
Middle-End Development with SvelteKit and Tensorflow.js(cprimozic.net)
2 points by Ameo 4y ago|0 comments
106.
Middle-End Development with SvelteKit and Tensorflow.js(cprimozic.net)
1 points by Ameo 4y ago|0 comments
107.
by Ameo 4y ago
This post really turned my stomach, but not because of the cheating. Yeah, obviously the cheating is wrong and students who cheated should be held accountable according to university policy. Yeah, it's nice that the professor was abl
108.
by Ameo 4y ago
You found that when they were required to come into the office, the underperforming employees+teams did more work? I wonder if you have any thoughts on why that is, I'm not necessarily surprised but I'm interested to hear more abo
109.
by Ameo 4y ago
I use bunny.net for my personal websites. I'm largely quite happy with it; it's definitely a great budget option and is very easy to set up. That being said, it's definitely a budget option. Their admin dashboard, features,
110.
by Ameo 4y ago
I've used this with FFmpeg with great success. Encoding time and video quality is comparable to other encoders, and file size almost always smaller.
111.
by Ameo 4y ago
Wow, they hot-swapped activation functions (GELU -> RELU) during training. They are indeed very similar activation functions, but it's kinda crazy to me that you can make that kind of a change to a model while it's training, p
112.
by Ameo 4y ago
I think that the browser is going to eat the desktop/OS and that most apps will eventually be browser-based. PWAs are the initial movement in that direction. As browser APIs expand and support more use-cases through WebAssembly, WebGPU
113.
by Ameo 5y ago
https://cprimozic.net/ Personal site and blog. Designed with the goal of being what I want people to see if they were to google me. My favorite part is my portfolio: https://cprimozic.net/portfolio/ I
114.
by Ameo 5y ago
For me personally, it really helps to be able to see things like how the responses compose to create the outputs. I know a lot of math-minded people are able to intuitively "get" concepts from symbolic representations, but yeah it
115.
Exploring Neural Networks Visually in the Browser(cprimozic.net)
84 points by Ameo 5y ago|8 comments
116.
by Ameo 5y ago
BunnyCDN (mentioned in the blog post) has this functionality. It is mentioned in the blog post and they're using it for image processing, but it works for videos too. I use it myself, but just as a normal static file CDN, but they hav
117.
by Ameo 5y ago
It's definitely full stack; it has endpoints[1] which are just arbitrary API endpoints where you can implement any logic you want (talk to a database, perform conditional logic, etc.). It may not be a fully-loaded solution like rails,
118.
by Ameo 5y ago
SvelteKit now has a static adapter[1] which is designed for those kinds of scenarios, defaulting to pre-rendering all pages and not requiring a server I think. I've not tried it out myself though, but it might be good for that if you
119.
by Ameo 5y ago
This seems to be the RAM they're provisioned it with: https://www.samsung.com/semiconductor/dram/module/M393A2K40B... It doesn't look to be ECC. > Why do you RAID-1 NVMe That's what they set
120.
by Ameo 5y ago
Hey, author here! - Request rate is pretty low on average and peaks at around 15 requests/second. That's nothing really and it would probably take 100x or more request rate before I saw any kind of CPU bottlenecking or similar; m

More ›