It took precisely one challenge before I got sick of non-functional programming.
I feel like if they didn't want functional answers, they wouldn't give me vectors.
It took precisely one challenge before I got sick of non-functional programming.
I feel like if they didn't want functional answers, they wouldn't give me vectors.
ah yes, the hardest part of any HackerRank challenge - reading from some weird version of stdin.
do I know how to read from stdin in R? nope, because I've literally never had to do that.
@Gargron I can understand the avoidance of third party scripts, but I really don't get the point of avoiding tracking, especially when you just said it made it harder to understand effect?
Sorry if I've totally missed like a key part of Masto here, I actually searched when I joined and saw nothing about/against analytics.
Like even at an instance level it seems right to let people understand anonymised info about impressions/interactions?
I love Judah and PC's relationship.
This single moment of Judah knowing how PC felt about him, and the confidence in writing it on her behalf, and PC getting to express it fully and honestly just hit me with a lot of emotion.
I wish I could feel like that. Like I knew how other people felt about me with confident, and that it was good. I miss that kind of intimacy and BJ amazingly makes me cry sometimes by showing me what it could be.
@Gargron This did surprise me about Masto when I joined. Is there a reason analytics is particularly difficult to implement, or is it just not high enough compared to other priorities?
β€οΈ
(cw eye contact)
@cassolotl I remember there being a debate about this because it increased the following of key community members, but neglected growing connections of individuals with smaller followings.
I think it could be a good idea thhough, because thinking about it, if the Ambassador bot trawled through various hashtags and used the same algorithm, it could find positive members within subspaces, which would encourage small-medium growth too.
A good #comic about #ExecutiveDysfunction and #mania by @Andrea. :)
http://par-la-fenetre.tumblr.com/post/165137660420/216-know-the-difference-seriously-though-where
tired: programming
wired: hacking
glitch!explodingmind: HECKING COMPTOOTERS
Pet peeve: "should of"
punk
@grainloom
WHY DID I CLICK ON THIS
@ashkitten Oh no :(
I"ve been trying to mask the feeling by switching to Huel in the evening and just not having any snacks around me, because like Quet only seems to give you that craving just after you take it, I don't know if Abilify is the same?
@ashkitten Hadn't heard of that one, but it sounds really really similar in how it works, so I guess it's the same effect?
It's weird! It's like I want to say I have a lot of self control, but also it is changing what my body physically thinks is the case
I found out today that quetiapine gives you massive munchies when you take it.
in other words, I have totally removed my feelings of guilt from eating total crap after 10PM
@amphetamine Yes! But the person I heard it from turned out to have really butts-politics, so I never got into it.
Gonna read through again now, thanks :)
@amphetamine I just read the DIY HRT part of that article and just - holy fucking shit - right?!
It is exactly as awesome as I wanted this to be.
So last month my fear was basically that I didn't know anything about machine learning, but felt I should start to learn more.
And now my fear is that I basically don't know any maths to know the important parts of machine learning.
PROGRESS.
how about some more nice applications of machine learning? yes you say? ok here you go pal~
this one's free too
"Reconstructing cell cycle and disease progression using deep learning"
Nature Communications 8, Article number: 463 (2017)
doi:10.1038/s41467-017-00623-3
https://www.nature.com/articles/s41467-017-00623-3
i feel it's important people know about all the wonderful things we can do with this tech, and that people are out doing it
it's not all scary cybre shit you know?
Also, #machinelearning stuff.
I spent fucking hours last week doing grid search and random search for tuning hyperparameters.
Fucking Adaptive Resampling exists.
https://topepo.github.io/caret/adaptive-resampling.html
I was convinced that there had to be a better way of filtering through hyperparameters, and while this can be really risky when you're like for *the* optimal tune because it futility of parameters is a risky field, it is TWO TO FOUR TIMES FASTER than grid search for the same results.
Jonkman Microblog is a social network, courtesy of SOBAC Microcomputer Services. It runs on GNU social, version 1.2.0-beta5, available under the GNU Affero General Public License.
All Jonkman Microblog content and data are available under the Creative Commons Attribution 3.0 license.