@davemark @TheChrisPaduan Yep, I think that’s why I like the big screen too … these old eyes can’t handle small screens any longer 😛

@RandallHawes @AngelaPreston I discovered a couple of days ago that I can follow particular Twitter accounts right here from the Fediverse using ActivityPub and this site:

https://bird.makeup/

Has been a game changer for me since now I can follow tech and news sites directly here and not wonder if I’m missing out on something important 🙂

@davemark I enjoyed it, but never got to finish it. I forget why though … I think we were moving countries a lot at that point and then my wife started playing it as well and we were both playing on one Switch and I just stopped — or something 🙂

Do you play just in hand-held mode? Or do you play on the TV? I think I finish more games on TV than in handheld mode and that might have had something to do with things when it came to “Breath of the Wild” …

@TheChrisPaduan

@davemark I’m torn … to buy the game, I’d have to buy a new Switch since we left our old one in Canada when we moved the last time. But it’s a new Zelda game … Plus, “Diablo IV” will be coming out soon after that … Do I have time to play two games? I don’t know …

#DevilsAdvocate

@TheChrisPaduan

I've always wondered why barns are red.
https://patreon.com/lowqualityfacts

People who don't speak a word of German are far too confident in their ability to assassinate baby Hitler if given the opportunity to time travel.

I swear typos are gonna be the depth of me.

And as if we didn’t have enough misinformation around (and lots of people going ga-ga over #ChatGPT and how great it is at beating search engines …) apparently Google now has their own misinformation engine — Bard.

Predictably, on one of its first public appearances, it spouted misinformation. But then again, probably a lot of people didn’t notice. Heck, people at Google probably didn’t notice since they showed it off proudly 😛

I’m going to call all these AIs (which is a misnomer anyway) MEs (Misinformation Engine) from now on …

https://www.engadget.com/google-bard-chatbot-false-information-twitter-ad-165533095.html

It’s only going to get worse …

#ChatGPT #Bard #ConversationalME

@trode Have fun and if you wanna talk about any of this stuff, I’m always around — I don’t like larger communities and so don’t frequent Discord etc. but do miss the shop-talk 😛

At the moment the one I like the most out of the web clients is InvokeAI but there are a lot of nice web clients around. Apparently there’s also a bunch of native macOS clients but they seem to be not so well-known, at least in the circles where I read stuff …

@trode I’m sorry it didn’t work for you. I had major issues with the Apple code and was going to spend the weekend breaking it down and re-working it to see why it failed. Guernika saved me all that … but then I realized that even that didn’t do what I wanted since I wanted an EINSUM compiled version of Dreamlike Diffusion but apparently the EINSUM version doesn’t work on macOS … I don’t know why but I saw a note to that effect on HuggingFace.

Ah, well 🙂

@trode From what I can tell based on my very limited research, it doesn’t look as if they are very interested at all. I tried several conversions the Apple way (or using other approaches based on the same approach) and all of them failed for me.

I then found the Guernika Converter (https://huggingface.co/Guernika/CoreMLStableDiffusion) and that actually worked correctly … for some models. Some would convert but didn’t work correctly with (at least) my Swift code.

But it is slightly better overall than anything else I’d tried before.

Since I’ve moved instances, I guess it’s time for another #introduction!

I’m a #data scientist and #MachineLearning engineer based in #NYC. I don’t really like making my employer part of my identity, but I currently work at Flatiron Health, and I was previously a quantitative finance researcher at a hedge fund. My career has brought me into communities for #OpenSource, #bayes and #NLP, where I usually go by Eigenfoo 🧑‍💻📚

Red Flowers

2023-01-16/17 day 217/218 of doing art every day for a year (mostly - I took sick days off).

This is me having fun. I think it could be done better, but I'm not there yet. I'll work on it.

I've uploaded a Timelapse for this piece. You can view it here:

https://youtube.com/shorts/YrdI0iCoauk

If you want to view older artworks I've done, I'm posting them to my blog at https://LMAshton.com. It'll be a little while until I've got them all posted.

#ArtMatters #AYearForArt #Procreate #ProcreateArt #DigitalArt #WomensArt #WomenArtists #ArtistsOnMastodon
A digital painting. In the fore…

Just getting back from #Photo #Trip to #Banff and the #CanadianRockies. What an amazing trip. This was an #ice #formation on #AbrahamLake.

#LandscapePhotography #Photography #Nature #NaturePhotography #Winter #Snow #mountains #PhotoOfTheDay

Colorful Dragon
Midjourney 4 AI Art
#stablediffusion #dalle2 #dalle #aiart #nightcafe #midjourney

Oranges in Oil

2023-01-14/15 day 215/216 of doing art every day for a year (mostly - I took sick days off).

This was another experiment using oil paints in Procreate. Obviously not real oil paint, but brushes that are designed to simulate oil paint. And the more playing I do with these brushes, the more I love them. They're fun!

#ArtMatters #AYearForArt #Procreate #ProcreateArt #DigitalArt #WomensArt #WomenArtists #ArtistsOnMastodon
A digital oil painting. Oranges…

Lotus

2023-01-12/13 day 213/214 of doing art every day for a year (mostly - I took sick days off).

This was another experiment, ie me having fun.

#ArtMatters #AYearForArt #Procreate #ProcreateArt #DigitalArt #WomensArt #WomenArtists #ArtistsOnMastodon
A pink lotus in front, green le…

Debug neural network layers without changing your code!

pip install surgeon-pytorch

Inspect layers, debug, and even train separately with this tool to make your machine learning life easier!

#machinelearning #datascience #python #deeplearning #kaggle #career #tech

Looks as if the #StableDiffusion image generation on the #Swift side of the fence for #macOS has been progressing faster than I thought 🙂

It feels as if it’s been barely any time since I last took a look but there are a lot more #CoreML StableDiffusion models out there now and a lot more GUIs too. I took a look at a few, combined features from a few, added my own features too and came up with something which lets me maintain my image gallery and to generate new images too.

Given that on my 32GB M1 MBP it takes about 13 seconds to generate an image at 20 steps (except for the first time) it is pretty fast too. At least, fast enough for my needs 🙂

So it looks as if I’m back to messing with StableDiffusion on Swift rather than #Python, at least for the time being …
Showing a GUI for StableDiffusi…

Yellow-bellied Marmot, A.K.A. Rock Chuck, letting loose a mighty roar that reverberated against the granite of Donner Pass. Who am I kidding, it was the cutest little echoing ‘squeak’… #Photography #WildlifePhotography