I finished the model and it worked so well that it cost me $8 to rent the graphics card, but the results made it worth the cost. thanks for dataset.

8 63

We release our weights from recreated Stanford Alpaca 7B - LLaMA finetuned on synthetic instruction dataset.

https://t.co/nHT3XjMovb

And it's surprisingly good:

(Keep in mind, the following results is just from the smallest 7B model; GPT-3 is 175B)

9 33

Happy New Year one and all...and we are off to a roaring start for 2023. Globally, at +0.7°C above the 1979-2000 baseline, this is the highest anomaly I have witnessed for this dataset.

236 517

I hope that any AI monstrosity that scrapes through my art just gets random brushes, paint tubes, and my hands (hey, at least those are correct) for their dataset.

7 60

GM!

Still working on captioning this dataset. Hope your day is pleasant and productive!

5 26

Exact same prompt. *Almost* the same training dataset.

To the right, I added a few more pictures to add a bit of variability. The results are much better! (details, colors).

Prediction: "dataset engineering" will be more critical than "prompt engineering".

32 294

So Deviantart is making their own AI graphics tool DreamUp, that uses art in deviatntart. You can from image settingas ask for your art to not be included in the dataset. (Tho DA is not responsible even if 3rd party still uses it)

0 0

Testing New Vintage/Retro Diffusion Dataset...

This getting interesting.. 🔥

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

7 26

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

Recognize this Artwork? I can train using your Artwork as Dataset.

“Girl with tiger face in the mountain”

So.. can i sell it? Thats the kind of question this article provide. 😉

0 1

Add residual to latent diffusion first stage training and train with danbooru dataset. Now it can encode 256x256x3 pictures to eight ordered 16x16x4 latent codes. I don't know if this first stage model generated latent codes can be used in diffusion. https://t.co/5NFcO64qDK

0 6

【MNIST dataset】

The MNIST dataset is
an acronym
that stands
for the Modified National Institute
of Standards
and Technology dataset.

It is a dataset
of 60,000 small square
28×28 pixel grayscale images
of handwritten single digits
between 0 and 9.

2 3

yep!!! working to train my own image dataset... its proving challenging as i do not have a dedicated GPU -- but here were some results. You can kinda see the skulls trying to be formed

0 2

☾ *✧. . sound on, please . . ✧* ☽
-
Look! Now these generated eyes can blink because I added closed eyes to my dataset.

2 14

Egok360: A 360 Egocentric Kinetic Human Activity Video Dataset. https://t.co/oPe9wylkJ7

0 0

I have trained custom DenseNet like lightweight model the increases the size of image by the factor of 2. I trained my model on an anime episode and used it on faces dataset. Here is comparison between bilinear interpolation and my super-resolution.

5 12

Currently working on a sketch coloriser using Deoldify, using Danbooru dataset. Current results. pretty good so far!

4 15