//=time() ?>
New week, new #CleanNFT. Generated with #machinelearning #AI from a custom dataset + original sound. Available for 24hrs.
Collision - sound 🔊
10 editions / 1.0 $XTZ
https://t.co/ynzicaPw06
#salyaku #cryptoart #Tezos #HIC #nftart #nftcollector #cleanft #NFT #hicetnunc #tezosart
So I gave @ai_curio a dataset of all my B&W pictures plus some real creepy ones I gathered and let their Looking Glass AI generate a bunch of images!
So without further ado, here are some of my favourite "this Valdevia picture does not exist"! https://t.co/6Ydn9IjiNv
Meet the #NovelAI AI Model family!
Our AI storytellers:
Euterpe: Fairseq 13b (Our latest addition, still experimental)
Sigurd: GPT-J 6b
Calliope: GPT-Neo 2.7b
Experimental AI models:
Genji: GPT-J 6b finetuned on a Japanese storytelling dataset
Snek: GPT-J 6b for python coding
Check out our January issue! Covering #DeepLearning for #kidney #transplant pathology; characteristics of publicly available #skin #cancer image datasets; the implications of remote GP consultations during #COVID19; and more.
https://t.co/fMSiontUsg
New @LandInfoNZ #lidar point cloud dataset covering Nelson, New Zealand 🇳🇿 now available from OT. This v. high pt density (~37 pts/sq m) dataset covers 475 sq km of the S. Island: https://t.co/3hLzKFlYC5 📷 1) Pt. cloud colored by classification; 2) pt. cloud colored by elevation
@blahjedi I was going to say 10th Frame or Leaderboard, but was running out of tweet-space. 😋 Do you want me to specify which Leaderboard? Coz I think they all had it. 🤔
I did try it for a few of these games with Datasette. Coz we never gave it up when we got the 1541. Fun times. 😎
Trained @ai_curio ‘s Looking Glass AI on a dataset of hundreds of my D&D character illustrations, it did frighteningly well!
#keplergl is a great tool when you want to quickly graph a geographic dataset without any code! 🔥
Just upload your CSV, Json or GeoJSON!
👉 https://t.co/pepFoSlgna
#geospatial #DataScience #dataviz
Went the fairly traditional route to get this @lukhash program from the album onto my 64. Played it onto a tape and into the datasette. Other than having to hook up one my PAL 64s, was a snap. Very very cool. But I NEED this on Vinyl now ;-) #commodore
#LookingGlassAI some cherry picked pics of the 20 just produced. The dataset contains only 10 faces from my favorites comic artists...
although I also trained a GAN on the dataset because why not so there's also a huge untapped latent space (these are from the trained model)
The unified IFSAR topography dataset of Alaska is ~0.44TB.
Here it is at 0.00818% in #rayshader with the imhof1 palette and snow_hs transformation applied. img is 635kb.
It had tile artifacts that were cleaned with the "<=0 = https://t.co/pP2DFHkMgL" transformation.
DID SOMEBODY ASK FOR MORE PARALLEL UNIVERSE WOJAKS
this time with a larger dataset
alright let's do some AI beksinskis in this thread. top left is "ground truth" (i.e, one of the actual images from the dataset), everything else is generated.
New postdoc call to start during 2022. A fully funded postdoc position to study population genomics in a big dataset of primate WGS. Previous experience with theory and popgen required. Funded by
@ERC_Research
Please RT. Thanks! Contact me by email if interested.
Theres so much furry fan characters in the dataset that you can get Toriel, Renamon, and Judy
Discriminator Dreaming with layers D.b8.conv0, D.b8.conv1, D.b4.conv and D.b4.fc, respectively. Were the spirals curly hair all along, overrepresented in the MetFaces dataset? https://t.co/YQuYOaM4Pb