Update README.md
Browse files
README.md
CHANGED
|
@@ -5,6 +5,9 @@ tags:
|
|
| 5 |
---
|
| 6 |
# Memleak.nude: a digital installation created through the nakedness of metal.
|
| 7 |
|
|
|
|
|
|
|
|
|
|
| 8 |
# Thesis and Methodology:
|
| 9 |
Humanity has switched intimacy and its entire realness into digital life, whereas public life has became a source of self-censorship. Our silicon collects it all, unfiltered, unlabeled and untethered by human boundaries.
|
| 10 |
The unfilteredness of data enticed me. Even though computers in the top levels have context about the data, like file headers, everything is clear and unfiltered binary when looked upon.
|
|
@@ -22,6 +25,9 @@ The embedding was overtrained with a small learning rate to capture as many deta
|
|
| 22 |
the model and the embeddings are cc0, any outputs that come out can be commercialized by the output's creator, the end user, with no reference needed to me, the author.
|
| 23 |
have fun, do anything you like with it. i can't impose restrictions on beauty.
|
| 24 |
|
|
|
|
|
|
|
|
|
|
| 25 |
|
| 26 |
## Intended uses & limitations
|
| 27 |
|
|
@@ -30,8 +36,8 @@ create compositions out of an artist's most sensitive data leak.
|
|
| 30 |
copy the contents to your inversion embedding folders for your local stable diffusion.
|
| 31 |
|
| 32 |
#### Limitations and bias
|
| 33 |
-
the embedding works best at creating init images and be used in AND statements with a weight. i.e. a
|
| 34 |
-
|
| 35 |
## Training data
|
| 36 |
|
| 37 |
the training data consists of 102 images of the artist's personal sensitive raw data. here are a few examples:
|
|
|
|
| 5 |
---
|
| 6 |
# Memleak.nude: a digital installation created through the nakedness of metal.
|
| 7 |
|
| 8 |
+
|
| 9 |
+
|
| 10 |
+
|
| 11 |
# Thesis and Methodology:
|
| 12 |
Humanity has switched intimacy and its entire realness into digital life, whereas public life has became a source of self-censorship. Our silicon collects it all, unfiltered, unlabeled and untethered by human boundaries.
|
| 13 |
The unfilteredness of data enticed me. Even though computers in the top levels have context about the data, like file headers, everything is clear and unfiltered binary when looked upon.
|
|
|
|
| 25 |
the model and the embeddings are cc0, any outputs that come out can be commercialized by the output's creator, the end user, with no reference needed to me, the author.
|
| 26 |
have fun, do anything you like with it. i can't impose restrictions on beauty.
|
| 27 |
|
| 28 |
+
here are some of the raw outputs:
|
| 29 |
+
  
|
| 30 |
+
|
| 31 |
|
| 32 |
## Intended uses & limitations
|
| 33 |
|
|
|
|
| 36 |
copy the contents to your inversion embedding folders for your local stable diffusion.
|
| 37 |
|
| 38 |
#### Limitations and bias
|
| 39 |
+
the embedding works best at creating init images and be used in AND statements with a weight. i.e. a cat AND nakedMetal::0.ab . Personally found 0.17-0.4 the best range but you are invited to experiment if using.
|
| 40 |
+
  
|
| 41 |
## Training data
|
| 42 |
|
| 43 |
the training data consists of 102 images of the artist's personal sensitive raw data. here are a few examples:
|