jldugger

31 ms·

511.
by jldugger 3y ago
> This is not for everyone, probably yourself included, but it does work for me. Sounds a bit like accrual based accounting, which indeed is not for me ^_^
512.
by jldugger 3y ago
> I do my budget in double-entry. I have a short term budget in GNUCash in the form future dated scheduled transactions, but for more strategic stuff I have an annual budget spreadsheet, which among other things, models income taxes. Inc
513.
by jldugger 3y ago
You are already being spammed by AI. You think those well written followups to your non-response were handcrafted? > Final follow up, I'm excited to hear your thoughts regarding LibertyRock Forest. Any interest in finding out more?
514.
by jldugger 3y ago
Plus, 90 percent of what recruiters do on LinkedIn is basically already automated by LLMs.
515.
by jldugger 3y ago
Yes, apologies, what I meant by "favored" was that in every department, women applicants were more likely to get an admission than men. But I'm pretty sure the admission rate can still be lower for women overall than men over
516.
by jldugger 3y ago
> Say I have a simple table of outdoor temperatures and ice cream sales. What can the machinery of causal inference do for me in this situation? Not much. Causal inference works over networks of variables, specifically a DAG. But usually
517.
by jldugger 3y ago
from UT's press office, https://news.utexas.edu/2023/10/05/ai-driven-earthquake-fore... > The researchers said that their method had succeeded by following a relatively simple machine learning approac
518.
by jldugger 3y ago
actual paper? https://pubs.geoscienceworld.org/ssa/bssa/article-abstract/d... the abstract: > The proposed algorithm is trained using the available data from 2016 to 2020 and evaluated using real‐time dat
519.
by jldugger 3y ago
My read on this was that you can just dump the lagged values as inputs and let the network figure it out just as well as the other, time series specific models do, not that time doesn't matter.
520.
by jldugger 3y ago
The M7 forecasting challenge makes this goal explicit. Not the only use of forecasting, and IMO it would be good to have other timeseries data to present to models.
521.
by jldugger 3y ago
> It would be nice to provide a non-DL approach as a baseline, but I would expect it to lag behind the DL methods. The M# competitions have usually shown very old forecasting algorithms work quite well, with frankly, way less training ov
522.
by jldugger 3y ago
https://drive.google.com/file/d/1gqtOfZG2sSanWgUdzn-lx-pwSXZ... has the source report. > Participants were asked to rate their mental health on a scale of 0 to 10... average the full sample of participants repo
523.
by jldugger 3y ago
> Isn't it very uneconomical? On the contrary, the terms of credit cards include a "grace period": if you buy something on a card in the first day of a statement period, you will get a bill due to pay it off 45-50 days lat
524.
by jldugger 3y ago
Counterpoint: autopay gives me peace of mind knowing that I won't miss a payment, which has huge downsides: credit score dings, interest charges, and potentially declined charges. What I think trips people up is accounting for future e
525.
by jldugger 3y ago
Yea, and it's a pretty regular occurrence that the recipient in the US doesn't answer the phone call, because it's like 3am and the number is unrecognized.
526.
by jldugger 3y ago
It's not just inflation; actual real interest rates are now the highest they've been in a decade[1]. This is how the fed works -- raise real rates until demand for goods drops. Today's real rate is still not particularly high
527.
by jldugger 3y ago
I wonder what cloudflare saw in the emergency test this week.
528.
by jldugger 3y ago
I never understood why Walmart bought jet.com, but that was extremely good for Amazon. It's not really possible for the FTC to unwind that approval without suing amazon though, right?
529.
by jldugger 3y ago
> Given at most 3 can share a Nobel Prize, Proposed title of your spy thriller novel: The Four Body Problem.
530.
by jldugger 3y ago
I figure there's bound to be a parser in Python that can load my graphviz files and from there it's easy enough to transform objects in memory as required.
531.
by jldugger 3y ago
I assume career page quality is related to how important immigrant visas are to staffing, where "unable to find qualified local labor" is a criteria. No idea how to quantify that though.
532.
by jldugger 3y ago
Well, right now I have graphviz dot which should be enough to load in and transform as needed into whatever DSL desired.
533.
by jldugger 3y ago
Three reasons: 1. I want changes to the diagram to be peer reviewable and understandable. 2. I want something else to manually lay out things, rather than have to adjust it personally on every addition or change. 3. I want a parsable struct
534.
by jldugger 3y ago
In my experience with graphviz, the two layout problems are: 1. the lines are not distinct -- they can bunch up and overlap, making them untraceable 2. Logical groupings are hard to maintain unless they are highly connected.
535.
by jldugger 3y ago
Isn't the entire wind turbine regenerative breaking?
536.
by jldugger 3y ago
lightdm, looks like. I'm not entirely sure how, but I've got some hacky version of unity still chugging along.
537.
by jldugger 3y ago
Okay, now do colemak in login screen so I can type my password in without having to use a visual keyboard.
538.
by jldugger 3y ago
Or it signals that Google limits bids to what's profitable, and Yahoo outbid them. Which would suggest it does care, but only up to a limit. I don't know offhand how the contracts work, if there's a minimum traffic sent claus
539.
by jldugger 3y ago
"Die a hero or live long enough to become the villian"
540.
by jldugger 3y ago
Okay, then why did Yahoo outbid Google for placement?

More ›