Update README.md
Browse files
README.md
CHANGED
|
@@ -4,14 +4,31 @@ tags:
|
|
| 4 |
- lora
|
| 5 |
- diffusers
|
| 6 |
- template:diffusion-lora
|
|
|
|
| 7 |
widget:
|
| 8 |
-
- text:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
output:
|
| 10 |
url: images/AC1.png
|
| 11 |
-
- text:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
output:
|
| 13 |
url: images/AC2.png
|
| 14 |
-
- text:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
output:
|
| 16 |
url: images/AC3.png
|
| 17 |
base_model: black-forest-labs/FLUX.1-dev
|
|
@@ -22,14 +39,53 @@ license: creativeml-openrail-m
|
|
| 22 |
|
| 23 |
<Gallery />
|
| 24 |
|
|
|
|
| 25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
## Trigger words
|
| 27 |
|
| 28 |
You should use `Abstract Cartoon` to trigger the image generation.
|
| 29 |
|
| 30 |
-
|
| 31 |
## Download model
|
| 32 |
|
| 33 |
Weights for this model are available in Safetensors format.
|
| 34 |
|
| 35 |
-
[Download](/prithivMLmods/Abstract-Cartoon-Flux-LoRA/tree/main) them in the Files & versions tab.
|
|
|
|
| 4 |
- lora
|
| 5 |
- diffusers
|
| 6 |
- template:diffusion-lora
|
| 7 |
+
- Abstract Cartoon
|
| 8 |
widget:
|
| 9 |
+
- text: >-
|
| 10 |
+
Abstract Cartoon, An eye-level painting of a monsters face, painted in a
|
| 11 |
+
vibrant orange, yellow, red, and blue color scheme. The monsters eyes are
|
| 12 |
+
covered in white paint, adding a pop of color to the scene. The eyes are
|
| 13 |
+
squinted with black paint, while the mouth is adorned with red and white
|
| 14 |
+
paint. The background is a dark blue, creating a contrast with the orange
|
| 15 |
+
and yellow of the monsters head.
|
| 16 |
output:
|
| 17 |
url: images/AC1.png
|
| 18 |
+
- text: >-
|
| 19 |
+
Abstract Cartoon, An eye-level view of a cats face is depicted in a close-up
|
| 20 |
+
view. The cats eyes are black, and the cats nose is white, with a red nose
|
| 21 |
+
and mouth painted in white. Its whiskers are white, and they are painted in
|
| 22 |
+
a variety of colors, including orange, yellow, and blue. The background is a
|
| 23 |
+
dark blue, and there is a star-shaped object in the top right corner of the
|
| 24 |
+
image.
|
| 25 |
output:
|
| 26 |
url: images/AC2.png
|
| 27 |
+
- text: >-
|
| 28 |
+
Abstract Cartoon, a portrait of Boat, painted by a dadaist abstract artist.
|
| 29 |
+
a colorful character with a whismical theme, the background color is dark
|
| 30 |
+
blue. All depicted by very colorful brush strokes. It all looks like a
|
| 31 |
+
cartoon from 90
|
| 32 |
output:
|
| 33 |
url: images/AC3.png
|
| 34 |
base_model: black-forest-labs/FLUX.1-dev
|
|
|
|
| 39 |
|
| 40 |
<Gallery />
|
| 41 |
|
| 42 |
+
- Hosted Here🧨: https://huggingface.co/spaces/prithivMLmods/FLUX-LoRA-DLC
|
| 43 |
|
| 44 |
+
**The model is still in the training phase. This is not the final version and may contain artifacts and perform poorly in some cases.**
|
| 45 |
+
|
| 46 |
+
## Model description
|
| 47 |
+
|
| 48 |
+
**prithivMLmods/Abstract-Cartoon-Flux-LoRA**
|
| 49 |
+
|
| 50 |
+
Image Processing Parameters
|
| 51 |
+
|
| 52 |
+
| Parameter | Value | Parameter | Value |
|
| 53 |
+
|---------------------------|--------|---------------------------|--------|
|
| 54 |
+
| LR Scheduler | constant | Noise Offset | 0.03 |
|
| 55 |
+
| Optimizer | AdamW | Multires Noise Discount | 0.1 |
|
| 56 |
+
| Network Dim | 64 | Multires Noise Iterations | 10 |
|
| 57 |
+
| Network Alpha | 32 | Repeat & Steps | 22 & 2200|
|
| 58 |
+
| Epoch | 10 | Save Every N Epochs | 1 |
|
| 59 |
+
|
| 60 |
+
Labeling: florence2-en(natural language & English)
|
| 61 |
+
|
| 62 |
+
Total Images Used for Training : 11 [ Hi-RES ]
|
| 63 |
+
|
| 64 |
+
## Best Dimensions
|
| 65 |
+
|
| 66 |
+
- 1024 x 1024 (Default)
|
| 67 |
+
|
| 68 |
+
## Setting Up
|
| 69 |
+
```
|
| 70 |
+
import torch
|
| 71 |
+
from pipelines import DiffusionPipeline
|
| 72 |
+
|
| 73 |
+
base_model = "black-forest-labs/FLUX.1-dev"
|
| 74 |
+
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
|
| 75 |
+
|
| 76 |
+
lora_repo = "prithivMLmods/Abstract-Cartoon-Flux-LoRA"
|
| 77 |
+
trigger_word = "Abstract Cartoon"
|
| 78 |
+
pipe.load_lora_weights(lora_repo)
|
| 79 |
+
|
| 80 |
+
device = torch.device("cuda")
|
| 81 |
+
pipe.to(device)
|
| 82 |
+
```
|
| 83 |
## Trigger words
|
| 84 |
|
| 85 |
You should use `Abstract Cartoon` to trigger the image generation.
|
| 86 |
|
|
|
|
| 87 |
## Download model
|
| 88 |
|
| 89 |
Weights for this model are available in Safetensors format.
|
| 90 |
|
| 91 |
+
[Download](/prithivMLmods/Abstract-Cartoon-Flux-LoRA/tree/main) them in the Files & versions tab.
|