I mostly read non-fiction, but am pumped to have received the bicentennial edition of the Brothers Karamazov. Although it is one of my favorite books, itβs been so long, I cant say why.
Of to read-iscover : )
2/2
@clearnoise.bsky.social
Scientist Formerly at Pitt Med, HHMI Janelia and Umass Med, but have jumped head first from the ivory tower cryo-Electron Microscopy && HPC aficionado OpenAccess (mostly) Free Software
I mostly read non-fiction, but am pumped to have received the bicentennial edition of the Brothers Karamazov. Although it is one of my favorite books, itβs been so long, I cant say why.
Of to read-iscover : )
2/2
One surprise Iβve encountered while watching the internet drown in #LLM garbage, is a renewed appreciation for books.
Iβve always maintained a fairly well stocked library; but now Iβm noticing that I take more time to relish the flow of thought, unique and flavored by the given author.
1/2
unfortunately, this oneβs not on scihub
21.04.2024 18:13 β π 0 π 0 π¬ 0 π 0oh yikes! would you share a link or paper title please?
20.04.2024 12:48 β π 0 π 0 π¬ 1 π 0sounds better than using βhallucinatedβ and the scare quotes around βstructureβ I used, but same vibe here
20.04.2024 02:15 β π 1 π 0 π¬ 1 π 0so what would you say as the reviewer?
20.04.2024 01:51 β π 0 π 0 π¬ 1 π 0have you seen anyone use a reslog plot or similar to asses fictional results? I assume the trend from 10k to 10M particles is not the same for real vs hallucinated βstructuresβ
20.04.2024 01:28 β π 0 π 0 π¬ 1 π 0Reminder the switch in terminology from "refereeing" to "peer review" in the 1970s was partly due to concern about government interference journals.plos.org/plosbiology/...
17.04.2024 22:26 β π 4 π 3 π¬ 1 π 0Friendly reminder that if you live or have a business in RI your state taxes are not due until
July 15
And if you are in Providence, Kent or Washington county, the same applies for federal taxes.
#taxday #taxes #RI #irs
Can we start a rumor that chatGPT uses βelucidateβ at a high rate and it is a trigger for review of a.i. gen content?
#academia #academicchatter
#cryoEM scientist 10 or more years ago - βI have to write an entire software package to do my research, okay then!β
Kids these days - βI have to install a new package and follow the simple il instructions someone took time to share on a mailing list?! How unfair!β
this is a fun experiment - looking forward to updates!
06.04.2024 12:38 β π 1 π 0 π¬ 0 π 0thatβs not what i want my electrons doing π₯Ί
www.972mag.com/lavender-ai-...
Brains do not come with gdb stubs. All we can do is printf debugging (aka talk about stuff).
HT: derf
oi, so much green already!!
02.04.2024 10:53 β π 1 π 0 π¬ 1 π 0also .. great idea for a kids book. Coyote and pitty become best friends, public has to acknowledge both are cute and misunderstood after watching happy hijinx.
02.04.2024 10:52 β π 0 π 0 π¬ 0 π 0Just chased a coyote off b/c I was worried about my β¦ pitbull. Glad it was a wiley coyote and not one that decided to ask
βoh yeah, you, the guy with the half rotten stick in your hand, what are yooouuu gonna do?β
#wildlife #coyote
Please donβt tip over your TEMs.
#cryoEM
phys.org/news/2024-03...
Diving Grand Cayman in a week and canβt wait to listen a little closer : )
HT Albert Cardona
#ScienceOutreach #Fish #scuba
youtu.be/POITH02VVrw?...
I was just bemoaning a similar feeling. Itβs so easy to overlook many successes in the face of a recent failure. Thanks for sharing; itβs good to remember weβre all doing the same song and dance!
28.03.2024 18:57 β π 1 π 0 π¬ 0 π 0Update:
Trying a reverse calendar put spread (which sounds more like a fun sex act than a derivatives strategy)
FWIW, i tried to short DJT but there were no available shares at $70 on IBKR
27.03.2024 16:02 β π 0 π 0 π¬ 1 π 0An important reminder for all of us freelancers:
If I fix your #gpu #code in an hour, thatβs because Iβve spent 13 years learning how to do it in an hour.
You owe me for the years, not the hours.
#consulting
When doing subtomo PCA in #cryoET, the MW bias can be seen in the latent space mapping
If you can see the bias in the eigenvectors, it sure seems like a directional FSC with the e-vectors could allow you to ignore or regularize them prior to projecting feature vectors.
Anyone in #TeamTomo tried?
In explaining my research/funding model to some tenured professors recently, one of the more interesting replies was -
βOh, so youβre kinda like a research mercenary?β
In so far as I left a broken #academic system to do what Iβm *really* good at
AND π get π paid π
I guess so
#PostdocExodus
I wonder how many T-mobile wifi customers follow the instructions to put their router by a window, but donβt change or cover up the admin passwords printed on the router?
#infosec #tmobile
β¦ but factoring in the peripherals a node of A5000 was less cost efficient than A6000 b/c they have the same footprint in a datacenter but yield lower return (assuming you keep them busy all the time)
22.03.2024 16:29 β π 0 π 0 π¬ 0 π 0Are they?
Do you relative to A6000 or the consumer cards? In my tests, A6000 performed a bit slower than a rtx 3090 on a (template matching) task given the same wattage.
I didnβt compare a5000 to consumer, but to A6000 they were ~ linear in scaling cost vs performance at the card level β¦ cont
3/3
π² How do I figure the cost in not upgrading but running my hardware > 50% of the time (my time and effort to schedule)?
π² What is the env. cost in manuf./disposing of a gpu?
π² The β‘οΈ savings ignores all supporting hardware.
*my avg in RI USA is ~$0.25 when all charges/taxes etc considered
2/3
An #nvidia rtx 4090 does 1.5x the math than a 3090 at the same wattage.
So upgrading saves $100/year β‘οΈ, but cost ~$2000
Of course there are other costs to figure, but it gets harder really quickβ¦