|
Fangz posted:The 2018 ADTRW AOTY chart, with my annotations! I'd really like to know what you use to make these. Weights were 5 to 1 for first to last? Any chance you can run it with logarithmic weights; just convert with log(weight+1) to take all the #1s down a bit? I can see my #1 and #2 choices dominating the rest and I don't think I was alone in being able to flip them freely. I'd like to see what the rest of the choices do.
|
# ? Jan 15, 2019 21:52 |
|
|
# ? Jun 10, 2024 13:11 |
|
Fangz posted:The 2018 ADTRW AOTY chart, with my annotations! Happy to be holding down the comedy flank.
|
# ? Jan 15, 2019 21:53 |
|
i guess me and nephytys are the same person
|
# ? Jan 15, 2019 23:36 |
|
Ranzear posted:I'd really like to know what you use to make these. I do these in R and yeah you're right. Running with log weights would be a bit different but not to my eye in any terribly interesting or systematic way.
|
# ? Jan 15, 2019 23:57 |
|
not surprised i ended up adjacent to beato believer since pripri is one of my favorite animedogsicle posted:i just always spin too many plates when it comes to seasonal anime and don't have any real reminder system or calendar so i miss new episodes and lost my place etc i just use the senpai.moe calendar to remind me what shows im watching and/or am behind on, i usually finish most or all of the shows i pick up a given season unless they just utterly collapse to a degree i cant even hate watch them like juuni taisen did
|
# ? Jan 16, 2019 02:03 |
|
Fangz I'm going to need you to make it really clear where I am
|
# ? Jan 16, 2019 02:44 |
Namtab posted:Fangz I'm going to need you to make it really clear where I am in the pile with squark, where people who voted for only one thing went
|
|
# ? Jan 16, 2019 02:46 |
|
dogsicle posted:in the pile with squark, where people who voted for only one thing went Ah, I see it.
|
# ? Jan 16, 2019 02:53 |
|
So, I have to find Erg and get them to recommend me stuff?
|
# ? Jan 16, 2019 22:21 |
|
Incidentally, I am reassured to see that Karen's power has been restored.
|
# ? Jan 16, 2019 22:35 |
|
|
# ? Jun 10, 2024 13:11 |
|
Fangz posted:The short version is that the Biplot is a 2D representation that just best preserved the variability in the dataset, so you have to go somewhere, even if it doesn't make total sense. In your case, note that *not listing* any of the more popular shows is interpreted as a choice, so you are grouped together with a bunch of other people who only gave one pick and it was a pick no one else chose. E.g. Namtab and Ibraham Gaunt with Slow Start and Killing Bites. I prefer to think of us as independent spirits.
|
# ? Jan 16, 2019 22:53 |