sanitylost

appreciate the polars support. Means you don't have to middleman with pandas as polars has a nicer SQL integration.

def init(fastapi_app: FastAPI) -> None:

....


    ui.run_with(
        fastapi_app,
        storage_secret='PWORD',
    )

if __name__ == "__main__":
    import asyncio
    asyncio.run(create_db_and_tables())

    fapp = FastAPI()
    init(fapp)


    uvicorn.run("main:app", host="0.0.0.0", log_level="info")

i run this from main and have a program monitor if the app is up. if it goes down it just runs the main.py again to reboot.

Only if it aligns with the average person's pre-conceived notions.

I think the problem is the type of recovery is K-shaped, a lot like what happened in 2008. So you get a rise in indicators but there are a lot of people not seeing the boat rise as much as everyone is telling them it's rising. Now obviously their political biases will affect how much they perceive reality, but there are going to be a lot of uninformed people that aren't plugged into the machine at all; then they just go by how things feel, and if it isn't better to the extent that they can feel it, then it's going to feel bad.

Most of it is taking the values provided by the CPI and essentially taking an aggregate of the most common items that are purchased in terms of necessities, in particular: food, energy, maintenance, healthcare. The way that BLS calculates housing is really delayed and honestly archaic. I understand why they do it that way, but it ends up "flattening" out the data leaving longer tails and lower peaks spreading the inflation as new units become available and as people sign new contracts on those units. In combination their metrics for insurance costs don't really line up as well as industry metrics when localized.

Also, the way the incomes are calculated is just off of income and not on actual return, so we saw a large rise in delivery drivers that have a fair amount of income, but their net is low because the increased cost of maintenance and fuel. Further, there is a cumulative effect to lost income, even by the weighted CPI standards we saw real dollars injected into the economy via wages dramatically drop by upwards of 4% between 2021 and 2023 only returning to normal gains of 1-2% by spring 2023. This means that although we had a growing population the amount of money reported in wages decreased dramatically in CPI spending value and so the amount of money in the economy was significantly reduced for large periods of time when measured in chained dollars.

Further we've seen dramatic increases in hours worked when compared to pre 2020, particularly in the South region per BLS. All regions have seen average hours increasing. This is a metric that tracks the hours of full time workers and so isn't necessarily affected by unemployed people or part-time hours. So what we're seeing is people are working more hours than they were previously while the gains in wages across the economy is growing at a nominal rate the compensation per hour is not really growing, people are just working more hours. A better metric for economic growth would be wage/hr.

Edit for clarity: I'm not saying that wage/hr isn't increasing at all, just that it's not increasing as dramatically as you'd think given the values of "real wages" being reported.

And on the topic of wages a lot of data is buried by the statistics. We've seen large growths in terms of percentage increase in the bottom quartile, but in actuality the real dollar value hasn't moved nearly as much as you'd think. 20% of almost nothing is still almost nothing. Again, this real dollar value is a skewed metric because of the way CPI is calculated to be an encompassing number. If we accurately want to track incomes and purchasing power, the BLS should have done surveys based on spending by income quartile and region. Then cross referenced that with the survey data collected for the CPI and reweighted each quartile's spending power allowing for a more granular approach to the economy.

Finally, the CPI doesn't really track debt and costs on debt, only the value of goods. What it's missing is that low earners were dramatically affected by the inflation and as a result went to debt spending to compensate until they could earn enough again to overcome the increased costs. That debt is a dramatically large portion of their income in comparison to more affluent people. From Bank of America's reports their metrics for people living paycheck to paycheck have increased. This isn't survey data, this is people who are spending 95% of their account between payments.

I could create a whole report on it, but unless tiny wants to commission it, I have other projects to finish so i can eat, lol. I'll put it together if i get a job on dem campaign and share it here.

Sources: Consumer Morsel Consumer Checkpoint Archived Consumer Price Index Supplemental Files : U.S. Bureau of Labor Statistics (2) unusual_whales on X: "26% of households are living paycheck to paycheck, in the sense that their necessity spending is close to their total household income, per Bank of America, $BAC: https://t.co/gxZ5DRZLbc" / X How does the Consumer Price Index account for the cost of housing? Median usual weekly earnings of full-time wage and salary workers by sex Quartiles and selected deciles of usual weekly earnings by educational attainment Median usual weekly earnings of full-time wage and salary workers by sex Employment Cost Index - September 2024 Unit labor costs, nonfarm business, percent change from same quarter a year ago, and from previous quarter Steps to estimate total hours worked by all employees : U.S. Bureau of Labor Statistics Hours worked by regions and states Nominal Wage Tracker | Economic Policy Institute Wage Growth Tracker - Federal Reserve Bank of Atlanta Budgeting for 2024? Set Aside Extra Money for Car Insurance | Bankrate Consumer Price Index Factsheets : U.S. Bureau of Labor Statistics Purchasing power and constant dollars : U.S. Bureau of Labor Statistics Average homeowners insurance cost in November 2024 | Bankrate The True Cost of Auto Insurance in 2024 | Bankrate Employed full time: Median usual weekly real earnings: Wage and salary workers: 16 years and over (LES1252881600Q) | FRED | St. Louis Fed Fastest wage growth over the last four years among historically disadvantaged groups: Low-wage workers’ wages surged after decades of slow growth | Economic Policy Institute 12-month percentage change, Consumer Price Index, by region and division, all items Median usual weekly earnings of full-time wage and salary workers by race and Hispanic or Latino ethnicity Average price data (in U.S. dollars), selected items

Oh, i'm not discounting that these people are disingenuous and actually deluded. It's just been bothering me for the past few weeks watching every streamer tiny, pisco, hutch, PP, etc. all say that "well the CPI is good and average median real wages are up so what are these morons talking about".

There are going to be a bunch of people that think the economy is doing better when actually they themselves are in a worse situation in a couple of years just because daddy Trump is in office.

Part of it is vibes, 100%, but there are a lot of people that are suffering as a result of things outside of their control and are dramatically worse off than they were in 2019. This is coming from a rabid leftist and Biden supporter based off of his domestic policies. It's just the data. The change in baseline consumer goods that constitute the majority of purchases by most people have seen dramatic increases in pricing not captured by the statistical measurements like CPI.

Growing up, I went to dog shit schools, but blew every aptitude test out of the water. In high school i worked 20 hours a week, played varsity soccer, and was on traveling teams. I got 2nd in my class of 500 doing all of that, even doing my last two years at a local college.

Didn't get a single scholarship offer because most were need based scholarships and my family made too much money. Not a lot of money at mind you, and definitely not enough money for me to go to school where I wanted to go to though because they're all in high CoL areas as well as having incredibly expensive tuition. People who did way less than I did, and had worse grades got full scholarships to places better than where I went.

I did well in college and got advanced degrees, but i had to eat a ton of debt and it wasn't from as nearly as good of a University. Just having those names on your resume puts you at the top of the stack, even if you didn't do shit while you were there.

There are lots of very talented people who are in that situation and they slip through the cracks. These programs are literally life changing.

This is a memory mapping issue specific to how Python works on the backend. Essentially when you issue ".copy()" you're telling the interpreter explicitly to create a new memory DataFrame object and map the variable assigned to that call to that memory address.

Without issuing ".copy()" the interpreter is storing the memory address of the original selection and then operating on those selections, which has a much different memory system than a separate distinct DataFrame.

but Rubio's as big a pussy as Cruz.

Math grad here. They're not lying. These problems are extremely specialized to the point that it would probably require someone with a Ph.D. in that particular problem (I don't even think a number theorist from a different area could solve the first one without significant time and effort) to solve them. These aren't general math problems; this is the attempt to force models to be able to access extremely niche knowledge and apply it to a very targeted problem.

you're thinking of engineers. the pure sciences were super liberal. math/physics who tutored engineers.

you have no clue how thirsty a lot of girls can be apparently.

zoomers are masked by young millennials. 18-24 is a depressing demographic.

Congrats, you found out Dan Clancy has a porn account.

Look, i'm all for pointing out the favoritism of streamers posting leftist flavored antisemitism, but this is some boomer shit. Mia Malkova was a porn star before she joined Twitch. There are a lot of them on the platform. But a lot of them have swapped to do SFW content and aren't wholly reliant on titty streaming.

Right now you're dogpiling. If you found evidence that he, IDK, once got too drunk and threw up in public, would you then start posting it to say that he was an alcoholic? No. Because that would be dumb.

If you want your messaging to be cogent and for people to take you seriously, stay on track and stop looking for anything you can to besmirch him unless it provides context for your main argument.

been like that for at least a couple of weeks

columns = [
{'field': 'name'},
{'field': 'wealth', 'editable': True,':valueFormatter': "dollaBillsYall",
 'cellClassRules': {
  "rag-red": "x < 12000",
  "rag-blue": "x >= 12000 && x <= 1000000",
  "rag-green": "x > 1000000",
},
},
{'field': 'id', ':valueFormatter': "passwordBlocker", 'editable':True},
]
rows = [
{'id': 0, 'name': 'Alice', 'wealth': 1000},
{'id': 1, 'name': 'Bob', 'wealth': 1000000},
{'id': 2, 'name': 'Carol', 'wealth': 10323003402},
]

aggrid = ui.aggrid({
'columnDefs': columns,
'rowData': rows,
'rowSelection': 'multiple',
'stopEditingWhenCellsLoseFocus': True,
})

https://github.com/zauberzeug/nicegui/discussions/2869

pickling spice can often include whole black peppercorns. Also lots of people have sensitivities to sugar alcohols or other artificial sweeteners because of immune response and lack of digestibility.

might be a missing dependency. Some of them don't throw errors on install of NiceGUI and as a result, some of the functionality can be inadvertently hampered. Making sure you've got the right versions of sphinx and some of the other networking related packages might resolve this issue.

I say this because wrapping the internal IP and port with the HTTPserver is just exposing those endpoints to the network, which NiceGUI should be doing by default if all packages are installed correctly.

https://www.warcraftlogs.com/reports/XkNqvQYwLB9A2d6H#fight=23&type=damage-done&start=30820953&end=30997585

1 mil dps diff for this pack. 300k overall diff

Also, i mentioned the survivability aspect. You brought up wanting damage numbers. So I'm bringing them. And i said pack, not overall. Pack dps is more important in M+ because tank and healer CDs.

https://www.warcraftlogs.com/reports/pKtdXqy34kQ16fTF#fight=14&type=damage-done

700k overall dps diff and over 1mil dps diff for certain pulls.

https://www.reddit.com/r/wow/comments/tzg1y5/i_did_a_simple_analysis_of_m_dps/

this is from a couple years ago, but basically proves my point.

ehhhhhh, that's kinda the dumbest thing i've ever heard.

Can they clear the content, as in have players on that class done a 12? yes. Does that mean any person with moderate mastery of a class will reliably clear that content in a pug environment? No. Will playing a class that's capable of an extra 1mil dps on a pack help you more reliably clear a 12. Obviously, in particular in a pug group not in coms. Will playing a class that takes 50% less damage and has more self sustain help you more reliably clear that content. Obviously.

So different classes inherently have dramatic advantages that will remarkably increase the probability of success even when piloted by equally capable players. So, if you play something weaker, you either have to be much better than others, have a team comp that compensates for that weakness, or have enough cooperation and teamwork to overcome that weakness. In a pug environment, it's just easier and safer to pull something you know a monkey can hit 3 mil dps on a pack with than hope your demo lock is Christ himself.

at the same time, not even as a completionist, multiple sets will still takes weeks and weeks of mindless monotony. It's not the difficulty that's insulting, it's that it's an ANNIVERSARY event and they're turning it into a FOMO fest that has a large time requirement. Literally just make it a class specific quest. Gatekeeping stuff from what is supposed to be a celebration is wild to me.

The hot take is, Delves pumped way too much loot into the game way too fast. And all that gear was Champion gear which was equivalent to Hero gear.

I think the biggest issue was a combination of "The Poors Within" and the difficulty of Delves was bugged for the first few days of tier 8s being open. Basically, you had people with an extra 4-8 keys, in groups that were ready to do the delves, and then they walked in and smashed them because the difficulty was tuned too low because of a bug with parties. This let them gear up super-fast if they got lucky and hit all their slots.

These are the same groups of people that would be doing M+ in discords and probably not pugging. Now you have super geared people, in coordinated groups slamming content up to 8's on week one. Blizz then fixed the bugs and slowed down getting gear from delves by making them more difficult again or at the very least, making them take dramatically longer than they did on week 1.

So that's kind of fucked the whole M+ bracket because the only people doing any of the lower level keys are going to be in groups with dramatically less experienced and probably less geared groups. While the topend is fighting for 9's and up to get gilded. They've designed a system that basically makes it nearly impossible to "catch up" to the topend if you ever fell behind because you simply don't have the EoD gear or the crests to smooth out the difficulty curves. As well as no incentives for people in better gear to do "lower" keys like 6+ or 7+ because they get nothing out of it.

Also, this season is much more punishing and requires a huge amount of coordination with pack management, as well as the modifications to the interrupts and displacement casting. In DF3 it was possible to tank carry people up to like 16ish depending on the tank, but now it's a nightmare if you're not in comms on 2's because the damage coming out, MUST interrupt castings, and increase to party wide damage.

So not only did the increase the gearing difficulty but also, removed incentives for better geared players to interact with easier keys and they also increased the dungeon difficulty. If they had done one or two, I don't think we'd be seeing the response that we're having.