jayensee.bsky.social
Hobby league of legends analyst, also talk about whatever topics catch my interest (usually related to data analysis)
Giving the butterfly app a try
Jayensee on YouTube and GitHub.
jayensee.github.io
241 posts
30 followers
83 following
Getting Started
Active Commenter
comment in response to
post
Sounds a lot like Path of Exile
Although in PoE there have been two apocalypses before and you kinda start the third
Doryani goes into the beast and wipes out the vaal, then Malachai goes into the beast and ends the eternal empire, then you go into the beast and destroy the order of the templars
comment in response to
post
Isa no céu azul!
comment in response to
post
Prices are never based on how profitable the company is, tho
"You already make so much money, give me some of it" doesn't usually work on companies
comment in response to
post
Wouldn't be a fami tweet without an error
comment in response to
post
And used the keyword xor for the bitwise xor operation
comment in response to
post
I'm not sure that using the currency is wasting it, but that's definitely what it feels like
Ironically, removing the possibility of crafting my own gear made me really not care about rare drops
In PoE1 I could turn an ok item into a great item, in PoE2 it has to drop great
comment in response to
post
The alt text is the text in the images, if you don't wanna read from screenshots
comment in response to
post
Touch snow
comment in response to
post
Grass
comment in response to
post
The weirdest thing is they still list gender identity as a protected category, they just added that exception
comment in response to
post
Holy shit, it's real
transparency.meta.com/policies/com...
comment in response to
post
I think you might have smeeched a little too hard with this one...
comment in response to
post
Congrats! 🥳
comment in response to
post
Ah, then what's the value in using the BT model at all in this scenario?
comment in response to
post
I thought each LLM was its own competitor
comment in response to
post
That wasn't its goal, tho
They're launching the Riot Cinematic Universe off of Arcane, the goal was to get people to watch their next shows
comment in response to
post
My arena recap looks a little better
comment in response to
post
We stan a Dlash king 👑
comment in response to
post
comment in response to
post
Goat on the manatee?
comment in response to
post
Gambling on stupid shit is always funny, because at least everyone's losing about the same
With stuff like sports betting you can lose more money by being bad at betting
But nobody know what's going on in Trump's head so it's an even playing field
comment in response to
post
Did something happen on the bad site?
comment in response to
post
Nah, they just thought you were a little too capable of opening it and wanted to play it safe
comment in response to
post
Blender has the preferences on the Edit menu
comment in response to
post
For Evelynn they'd have to blur the ccs
comment in response to
post
Damn
comment in response to
post
Those aren't done by the federal government, I assume
Federal government would be stuff like the USPS
comment in response to
post
The federal government employs a grand total of like 3M workers (~2% of the workforce)...
Him thinking that his department could have a significant positive impact on the economy by firing people is crazy
comment in response to
post
But you don't know the objective things, just your perception of them
comment in response to
post
Calling it statistics can be a bit misleading
The model doesn't have access to the training data after it's done training
But, yeah chatgpt is basically just a very advanced auto-complete
comment in response to
post
:(
comment in response to
post
What Elon is currently finding out is that he actually bought Hotmail and not email
comment in response to
post
Damn, I bought the book a few years back because I couldn't find it online
comment in response to
post
But those Rs don't apply to the transient players, they describe the players who stick around till they retire or die
The journey of the transient players is: start at their initial rating, go down a bunch because they don't do well, then quit with the lower rating, feeding points into the pool
comment in response to
post
I think what's happening is the distribution of player skill has genuinely changed a lot over time
We live in a time where everyone has easy access to videos from incredible players explaining how to play, so the baseline has probably gone up and the system takes a while to catch on
comment in response to
post
At least in The Rating of Chess Players Past & Present the only mention of players who quit early that I can find is this
(screenshot of the quote because it doesn't fit within the character limit)
comment in response to
post
Yea, but the players they lose to eventually retire too so it evens out
comment in response to
post
Yea, but he completely disregards the new players that lose a couple times then stop
If it's a zero-sum game, the top players had to have gotten those points from somewhere, and those people also stop playing eventually
In the steady state the rating distribution should stay mostly the same
comment in response to
post
From my understanding, a poisson regression assumes homoscedasticity (constant variance), which is definitely not the case in this scenario
comment in response to
post
Also, treating all the data points equally disregards the sample sizes of each point
Maybe the week-by-week number of patients admitted is very consistent, in which case this doesn't matter as much, but it would be nice to have a graph with this information
comment in response to
post
Doesn't the end of the mandate for universal testing significantly bias the results?
I think mixing data from before and after the mandate ended isn't helpful
comment in response to
post
The thing is with a zero-sum system you already don't get deflation
The absolute skill equivalent of a given rating always drifts over time, but it can drift both ways
comment in response to
post
Lasted longer than I expected, tbh
comment in response to
post
Do you have links for some of his worst comments?
My impression of him was that he was just the stereotypical reddit atheist before that even became a stereotype (and well before reddit was a thing)
comment in response to
post
The challenge of efficiently optimizing a "noisy" (no bound on gradients) multivariate function that has many local minima, isn't guaranteed to be continuous, has unbounded variables, and in which each variable has a different scale was more of a headache than I thought at first