Spaces:
Sleeping
Sleeping
update
Browse files
app.py
CHANGED
|
@@ -47,9 +47,12 @@ It is recommended to upgrade to GPU in Settings after duplicating this space to
|
|
| 47 |
<a href="https://huggingface.co/spaces/Ziqi/ReVersion?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
|
| 48 |
'''
|
| 49 |
DETAILDESCRIPTION='''
|
| 50 |
-
ReVersion
|
| 51 |
'''
|
| 52 |
# DETAILDESCRIPTION='''
|
|
|
|
|
|
|
|
|
|
| 53 |
# Custom Diffusion allows you to fine-tune text-to-image diffusion models, such as Stable Diffusion, given a few images of a new concept (~4-20).
|
| 54 |
# We fine-tune only a subset of model parameters, namely key and value projection matrices, in the cross-attention layers and the modifier token used to represent the object.
|
| 55 |
# This also reduces the extra storage for each additional concept to 75MB. Our method also allows you to use a combination of concepts. There's still limitations on which compositions work. For more analysis please refer to our [website](https://www.cs.cmu.edu/~custom-diffusion/).
|
|
|
|
| 47 |
<a href="https://huggingface.co/spaces/Ziqi/ReVersion?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
|
| 48 |
'''
|
| 49 |
DETAILDESCRIPTION='''
|
| 50 |
+
ReVersion
|
| 51 |
'''
|
| 52 |
# DETAILDESCRIPTION='''
|
| 53 |
+
# ReVersion: <R> represents the learned text token for a relation. Use <R> in your prompt for relation-specific generation.
|
| 54 |
+
# '''
|
| 55 |
+
# DETAILDESCRIPTION='''
|
| 56 |
# Custom Diffusion allows you to fine-tune text-to-image diffusion models, such as Stable Diffusion, given a few images of a new concept (~4-20).
|
| 57 |
# We fine-tune only a subset of model parameters, namely key and value projection matrices, in the cross-attention layers and the modifier token used to represent the object.
|
| 58 |
# This also reduces the extra storage for each additional concept to 75MB. Our method also allows you to use a combination of concepts. There's still limitations on which compositions work. For more analysis please refer to our [website](https://www.cs.cmu.edu/~custom-diffusion/).
|