Third pass tweet collection on 100 most recent alert_gc tweets is complete.

11,112 Unique tweets
7,473 Unique user IDs

Still unfiltered, but now a large enough dataset to get working on.

0 2

MidJourney dataset is filled with so many beautiful faces!

0 2

First attempt to use Stable Diffusion textual Inversion on my own dataset

0 6

Resources and the park: MAMMALS IN PORTUGAL: A data set of terrestrial, volant, and marine mammal occurrences in Portugal ....... https://t.co/hSnZu4ral1

0 0

The alternative was, if you had a datasette, to plug it in and press "play". The game couldn't tell the difference. 😁

(Still works with https://t.co/gclbq1UT0i's C64 Preservation Project .g64 disk images of Leaderboard and 10th Frame.)

0 3

Testing New Vintage/Retro Diffusion Dataset...

This getting interesting.. 🔥

Going to get these release ASAP..
these new datasets needs to given public 💖

7 26

Feels kinda sad GBA DiffusionModel Dataset not looking good.
Needs redone seems getting that back at line!

there is some other cool stuff coming out SOON!



tested default simple prompts..

6 27

Sorry not doing much art.. but doing something for sorting some files... and data collecting 😎

This image is from a faulty SNES dataset that has already been trashed.

0 2

New to tests my latest SNES Diffusion dataset model..

In a few tests it gives too many artifacts and so I went through the whole dataset files again and now it should be much cleaner. so now its redone and new session started!.

10 22

Testing new Dataset Model.

Here is first outputs.

it is very early stage.
push this model further in a few days!
Seems promising. We'll see what's to come, I'll keep you posted!

14 30

https://t.co/llIP3PLG4X

GAN still create from a dataset including textile artworks and photographs taken during the Australian Black Summer 2019-20

2 10

Notice all who used my NOTEBOOK Forks & diffusion models.
all beta data has now been publicly purged.
I've started building bigger better and much higher quality datasets with new methods.
will focus on a bigger project, which includes many models.

9 32

welll its supposed to be Tommy Cash but I'm not sure midjourney knows who he is based on their dataset, this is still kind of close though in my humble opinion



0 1

♡ Just an AI girl ♡

a collection of images generated by DALL·E 2 that show that the AI model was trained on a dataset containing photos of the popular girly aesthetic found online on social media platforms like instagram, tumblr, & pinterest

3tez♡

https://t.co/d1BPO02t34

5 21

my AI now generate better anime image from text by adding mean(mean(dataset_clip_image_embbedings) + clip_text_embedding)

it shift the clip text embbeding from clip distribution to my dataset clip distribution and the second mean normalize it

1 8

Say hello new friends! conjured them on our world 🪄

This is also one of my newest dataset..
Default settings it can do cool stuff
Going release down what it can do too ;)

Testing artstyled portrait style.

5 21

it's so cool my AI is better than i expected wow this only trained with 99k images for 18 hours imagine if it was trained with bigger dataset and train longer

0 3

don't give niko to my AI because it's ouside what it learned which is anime dataset still interesting though

0 7

YESsss... it finally works!
The very first releases Diffusion Models
At the default prompt..

This is a very early stage!

And this is also small testing dataset 6k images, 65k/i 512x512.

15 60