Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
neoncontrails
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
22 ms
·
61.
▲
by
neoncontrails
5y ago
I won't lie, the mere possibility that Google had rescued TF-IDF from antiquity, grafted some neural transformer layers onto it, and perhaps even advanced its own search capabilities as a result was extremely intriguing. Although in hi
62.
▲
by
neoncontrails
5y ago
That stackoverflow library is so gloriously bonkers. Is it useful in practice?
63.
▲
by
neoncontrails
5y ago
A coworker once insisted on fondling discarded drives with extremely strong magnets, then taking a power sander to the magnetized part. I wasn't sure if this was overkill or not. Is there any wisdom in this?
64.
▲
by
neoncontrails
5y ago
Genuinely curious, is "25 seconds" a typo? The bar graph suggests Dark Mode adds an additional 3000ms, presumably measured in wall clock time, but IIRC extensions are executed in a single thread. So I'm having trouble attribu
65.
▲
by
neoncontrails
5y ago
Highly recommend looking into Munch, "a dictionary that supports attribute-style access, a la JavaScript." https://github.com/Infinidat/munch In theory , implementing attribute-style access on user-defined
66.
▲
by
neoncontrails
5y ago
> The UserDict class initializer makes a dictionary which it stores in self.data. All of the methods on this dictionary-like UserDict class wrap around this self.data dictionary. Incredible. This appears to be exactly the solution to m
67.
▲
by
neoncontrails
5y ago
I might have balked at some of this advice until a few days ago, when I tried to create an FHE-encrypted key/value store that could be used as a drop-in replacement for `dict`. As it turns out, subclassing `dict` is awkward — especiall
68.
▲
by
neoncontrails
5y ago
You've hit the nail on the head: the early Word2Vec models treat everything as a word, but words are polysemous and fluid. Mutual information is useful for discovering n-grams that can be informatively mapped to the same vector space a
69.
▲
by
neoncontrails
5y ago
Did this example confuse anyone else? > For instance, given an Array in JavaScript, > window.x = ["a", "b", "c"]; > We can access it in Python as, ''' >>> from js import x # imp
70.
▲
by
neoncontrails
6y ago
Thanks. I was a bit puzzled earlier why AWS was so insistent about enabling object locking, my specific use case doesn't profit from remote versioning at all. But I can see how this would mitigate log integrity concerns. I'll defi
71.
▲
by
neoncontrails
6y ago
The book I usually recommend is Two Scoops of Django with a small caveat that while it remains the most lucid guide to a fairly complex framework, Django itself has evolved since its publication (most notably in its support for async). But
72.
▲
by
neoncontrails
6y ago
These are interesting epistemological questions. Is there a compelling reason to label a certain neural architecture as a "siamese network"? I'm guessing that this is a reference to conjoined twins, a biological phenomenon wi
73.
▲
by
neoncontrails
6y ago
I agree, GPT-2 makes an excellent writing companion. I actually developed my own version of talktotransformer.com to introduce some features that were missing from the public demo, including: * support for custom/fine-tuned/intern
74.
▲
by
neoncontrails
6y ago
No. GPT-2 is available to the public, at least up to so many millions of parameters. GPT-3 has so far been released exclusively to a small number of beta testers with strict rules on what you can do with it.
75.
▲
by
neoncontrails
6y ago
Wow. Could you embed these files in a GitHub README.md, for instance, or does the playback require a specific decoder? I've long wondered how developers manage to capture snazzy terminal demos of their projects, as the Linux tools I&#x
76.
▲
by
neoncontrails
6y ago
Huge fan of your gpt2-simple library, which I used to train a satirical news generator in a Colab notebook: https://colab.research.google.com/drive/1buF7Tju3DkZeL-EV4Ft... > Generates text faster than gpt-2-simple a
77.
▲
by
neoncontrails
6y ago
Earlier today, on the subject of why I made a particular design choice, I went straight for tier 6. I absolutely could have promoted my response to tier 5, or even 4, but I was too focused on prevailing as correct to see that one can affirm
78.
▲
by
neoncontrails
7y ago
How does one inspect WAMP payloads in the browser? I played with the demo app that syncs the position of a remote cursor with a local one, generating >100 state updates in the process. I was curious what those messages consisted of, but
79.
▲
by
neoncontrails
7y ago
> I learn a lot through examples, and seeing so many examples of functioning code was very beneficial. Quick sidebar: if you're looking for an intermediate/advanced-level book with a similar style of pedagogy, I can't reco
80.
▲
by
neoncontrails
7y ago
One of the nice, well-thought things about $ (end of line) and its inverse, ^ (beginning of line), is they have exactly the same meaning in normal mode as they do in regular expressions (with a small caveat that ^ ignores whitespace, so it&
81.
▲
by
neoncontrails
7y ago
Oh my goodness. I've committed some naughty deeds with the ast module before, but this is truly, wonderfully, heinous. You push this idea so far that I wouldn't even recognize some of these examples as syntactically valid, let alo
82.
▲
by
neoncontrails
8y ago
The Atlantic has a reputation for thoughtful, prescient journalism -- often with a focus on technology, but more often its surprising intersections with the world at large. I think it's frequently regarded as artsy and literary, like a
83.
▲
by
neoncontrails
8y ago
In high school I made a tic-tac-toe game that implemented what I thought was a clever trick for its has-winner? routine: starting with a 3x3 grid of ones, I multiplied each of its rows, columns, and diagonals by a unique prime number. This
84.
▲
Ask HN: Where does one find modular CSS components?
1 points
by
neoncontrails
8y ago
|
0 comments
85.
▲
by
neoncontrails
8y ago
As a longtime MBP user who just bought a Dell XPS 15, it’s the little things I miss. File previews in finder. The seamless trackpad experience without having to write a conf file to modify an open source driver. I don’t exactly miss the UI
86.
▲
by
neoncontrails
8y ago
I too have noticed this and have also gotten in the habit of prefixing technical queries with site:news.ycombinator.com. I’m glad it’s not just me. It’s an imperfect solution and I hope Google is invested in offering a better one.
87.
▲
by
neoncontrails
9y ago
Are document vectors supported analogously to word vectors? It's a bit peculiar to me that gensim withholds useful word2vec methods from the Doc2Vec class, like most_similar_from. I'd really enjoy having a common interface to both
88.
▲
by
neoncontrails
9y ago
This is much more nuanced criticism of language vectors than Chomsky's impatient hatred of them. There are interviews worth watching in which he loathes them for all the reasons you mention. Thankfully, I suppose, they rarely disappoin
89.
▲
by
neoncontrails
9y ago
Word2Vec actually employs a default strategy of pruning the most common 0.0001% of words in the corpus. My understanding is that this permits the RNN layer to train on the unadulterated text rather than an unnatural concentrate of tokens, w
90.
▲
by
neoncontrails
9y ago
I started wondering if I even needed to steal cars in GTA5: the Metro was such a convenient way to get around downtown pseudo-LA.
More ›