cl3misch

12 ms·

181.
by cl3misch 1y ago
In desktop Firefox it maxes out one core while OP's link is the active tab. When the tab is in the background, CPU usage looks negligible.
182.
DotDotGoose: Free open-source tool for manually counting objects in images(github.com)
1 points by cl3misch 1y ago|0 comments
183.
signal-export: Export your Signal chats to Markdown files with attachments(github.com)
3 points by cl3misch 1y ago|0 comments
184.
MergerFS: Combine numerous filesystems into a single mount point(github.com)
2 points by cl3misch 1y ago|0 comments
185.
Apptainer: Application Containers for Linux(apptainer.org)
123 points by cl3misch 1y ago|106 comments
186.
Awkward Array: library for variable data using NumPy-like idioms(awkward-array.org)
2 points by cl3misch 1y ago|0 comments
187.
SPlus: A Stable Whitening Optimizer for Neural Network Optimization(github.com)
1 points by cl3misch 1y ago|0 comments
188.
NumPy 2.3 Introduces OpenMP Parallelization Support(phoronix.com)
1 points by cl3misch 1y ago|0 comments
189.
by cl3misch 1y ago
> In NumPy, A B works if A and B are both scalar. Or if A is 5×1×6 and B is 5×1×6×1. But not if A is 1×5×6 and B is 1×5×6×1. Huh? > So, I removed it. In DumPy you can only do A B if one of A or B is scalar or A and B have exactly the
190.
by cl3misch 1y ago
How does this compare to xarray, which also introduces named axes on numpy arrays? https://github.com/pydata/xarray
191.
by cl3misch 1y ago
I could never understand why people use dstack, hstack and the like. I think plain np.stack and specifying the axis explicitely is easier to write and to read. For transposes, np.einsum can be easier to read as it let's you use (single
192.
Astroturfing(en.wikipedia.org)
7 points by cl3misch 1y ago|1 comments
193.
PageCrypt – Password Protect HTML(pagecrypt.maxlaumeister.com)
2 points by cl3misch 1y ago|1 comments
194.
Differentiable Programming in Mojo(github.com)
1 points by cl3misch 1y ago|0 comments
195.
Unitful Quantities in Jax(github.com)
1 points by cl3misch 1y ago|0 comments
196.
Adwaita Sans and Mono Typefaces(gitlab.gnome.org)
4 points by cl3misch 1y ago|0 comments
197.
by cl3misch 1y ago
I think the multiplicative midpoint (i.e geometric mean) is more sensible for such a large range, which gives 10k. Still not as bad!
198.
Interrupt: Linux Powered Hacking Gadget(interrupt-tech.com)
3 points by cl3misch 1y ago|1 comments
199.
Why do Indian researchers accept to publish in predatory journals?(academia.stackexchange.com)
7 points by cl3misch 1y ago|0 comments
200.
by cl3misch 1y ago
As mentioned in this discussion: https://news.ycombinator.com/item?id=43769486
201.
More-itertools: More Python routines beyond itertools(github.com)
5 points by cl3misch 1y ago|1 comments
202.
by cl3misch 1y ago
Care to elaborate? Where is the complexity hidden?
203.
ZimaBoard 2(zimaspace.com)
3 points by cl3misch 1y ago|1 comments
204.
by cl3misch 1y ago
Neither in (German) high school nor in the many math courses of a physics B.Sc. have I ever used the secant function. I am surprised the article does not explain it in the beginning. I assume for other people it must be a common function?
205.
by cl3misch 1y ago
Apparently it's only available in MS Office: > The basic styles of Kermit (Regular, Bold, Italic, and Bold Italic) are available today in Office, with the remaining 38 styles arriving in early May. ...from the last paragraph of the
206.
by cl3misch 2y ago
> there is no evaluation of the truth value of any statement at any point You could argue that the attention part of the network is some form of truth validation of the next predicted token? But indeed, the current chat interfaces don&#x
207.
by cl3misch 2y ago
> It's also immediately clear to me when I look at the architecture of transformers that reasoning is not in the cards. I'm not saying that's incorrect, but thb that's exactly the tautology I was talking about!
208.
by cl3misch 2y ago
Are you not reasoning on lossy abstractions? I'm still not on board with the (seemingly prevalent) notion that LLM's can't reason. What's reasoning, anyway? I'm not actively advocating for any side, but the argument
209.
by cl3misch 2y ago
According to other comments stating how responsive Discord is to reports, it might be better to not delete these webhooks but instead report the connected users/servers.
210.
by cl3misch 2y ago
Wow, very addictive loop! I instantly played a couple of rounds. Fyi, the perk to self-destruct the last block didn't seem to work.

More ›