00:45 <Qboid> 0d 1h 0m 0s left to event #30: [at 2018-10-19 01:45:00]. Say '!kountdown 30' for details

00:45 <iximeow> !wpn whitequark

00:45 <whitequark> !wpn egg

00:49 <egg|cell|egg> ;8ball did bofh get eaten by moths

00:49 <kmath> egg|cell|egg: Nope

00:49 <egg|cell|egg> Hmm

01:15 <Qboid> 0d 0h 30m 0s left to event #30: [at 2018-10-19 01:45:00]. Say '!kountdown 30' for details

01:35 <Qboid> 0d 0h 10m 0s left to event #30: [at 2018-10-19 01:45:00]. Say '!kountdown 30' for details

01:42 <Qboid> 0d 0h 2m 30s left to event #30: [at 2018-10-19 01:45:00]. Say '!kountdown 30' for details

01:45 <Qboid> 0d 0h 0m 0s left to event #30: [at 2018-10-19 01:45:00]. Say '!kountdown 30' for details

02:45 <SilverFox> if I do some neural net stuff, can I run it on my GPU if it has some AI shit on it?

02:45 <SilverFox> there are GPUs that are good at neural stuff

02:47 <UmbralRaptop> With the right software, drivers, etc.

02:51 <SilverFox> lets say I train a neural net for a solid 3 days straight or whatever, does that progress get saved and then I can just copy paste that network into other applications or whatever?

02:51 <SilverFox> or do i have to train it every time?

02:52 <SnoopJeDi> a trained network is deployable, sure

02:52 <SilverFox> perfect

02:53 <SilverFox> I have two things here, a Ryzen 1700 16 thread CPU, and a GTX 1060 6GB, which do you think would perform better given image recognition?

02:56 <SnoopJeDi> probably the GPU

02:57 <SilverFox> that's what I thought

02:57 <SnoopJeDi> I wouldn't recommend guessing though, just try both.

02:57 <SilverFox> but perhaps i should get a CPU instance going first and see how it performs

02:58 <SilverFox> if CPU does well then I'm going to try and see if it can do real time image processing with a raspi 3+

02:58 <SilverFox> hopefully the 1.4GHz of ARM strength will do

02:59 <SilverFox> and because of how power efficient it is I should be able to just easily chuck it into my car and away I go

10:08 <bofh^28> oh wow the BepiColombo launch is in <24hrs.

11:31 <egg|work|egg> bofh^28: although Qboid thinks it was yesterday

13:06 <egg|work|egg> !wpn bofh^28

13:29 <UmbralRaptop> Seen while hurrying to the bus stop: 3 cats, 1 fox, 1 hawk. o_O

13:30 <egg|work|egg> UmbralRaptop: yay

13:30 <egg|work|egg> UmbralRaptop: catpics?

13:41 <UmbralRaptop> egg|work|egg: was hurrying, so no time =\

13:44 <egg|work|egg> meow

15:27 <kmath> <rrika9> building dwarf debug info from shell scripts https://t.co/AzfCvys6Ar

15:41 <bofh^28> egg|work|egg: pong

15:41 <bofh^28> just woke up

15:42 <egg|work|egg> O_o

15:42 <egg|work|egg> bofhtime

15:50 <egg|work|egg> bofh^28: wait, do you tweet when asleep then O_o

15:54 <SnoopJeDi> bofh^28, re: yesterday's "math is hard" I just last night discovered this commencement address by my math advisor (while looking up his new address), it's very relevant: http://www.utdallas.edu/~zweck/UNM_Graduation.pdf

15:55 <SnoopJeDi> "Before we understand a piece of mathematics we feel stupid because we feel we are not clever enough to work out what is going on. We struggle for hours, days, or even years trying to solve a problem. One day we finally work it out. For about two hours we are on Cloud Nine. So we take a walk in the park to rejoice in our brilliant insights. But the longer we walk the more stupid we feel, 1By John

15:55 <SnoopJeDi> Zweck 1 because we gradually realize that the solution to the problem is so simple we should have known it all along. How could we be so stupid not to see the solution straight away? I have this experience every time I do math."

16:00 <bofh^28> egg|work|egg: kinda. more like briefly woke up, looked at twitter, fell back asleep. my sleep schedule's been... interesting.

16:01 <egg|cell|egg> Chat

16:08 <SnoopJeDi> egg|cell|egg, chat? https://d3gqasl9vmjfd8.cloudfront.net/491c0e9d-edb9-400a-81a6-a4ccc91c0a44.png

16:47 <UmbralRaptop> TFW surreal memes are relatable https://twitter.com/MemesSurreal/status/1053309495152201728

16:47 <kmath> <MemesSurreal> https://t.co/TdAdG9p2EK

17:05 <SnoopJeDi> pfft, what's surreal about convex regular polyhedra

17:13 <UmbralRaptop> hah

17:25 <egg|cell|egg> Bofh: meow

17:26 <UmbralRaptop> cat? https://www.sdss.org/surveys/orangespider-smaller/

17:26 <SnoopJeDi> !pet le chat

17:27 <SnoopJeDi> confusing image is confusing

17:27 <SnoopJeDi> I like the name though, reminds me of the DES "tank"

17:38 <UmbralRaptop> aaaaaaaaaaa https://twitter.com/PlavchanPeter/status/1053338306170249216

17:38 <kmath> <PlavchanPeter> Time for a stiff drink, and not the celebratory kind. 16/117 (13%) proposals were selected for funding from the NAS… https://t.co/DnC3K6Lp11

20:06 <SnoopJeDi> ooh, next week's colloquium is on Mu2e

22:17 <UmbralRaptop> … you found a planet where? https://aasnova.org/2018/10/05/aldebaran-b/

22:20 <UmbralRaptop> Er, "As Farr and collaborators have shown, it’s possible to extract accurate stellar and orbital parameters from irregularly sampled radial velocity data"

22:20 <UmbralRaptop> Behold the exciting analysis known as Lomb-Scargle, first published in the 1970s!