|
--- |
|
tags: |
|
- text-to-image |
|
- lora |
|
- diffusers |
|
- template:diffusion-lora |
|
widget: |
|
- text: 'Logo Design, A close-up of a bottle cap with the word "HUNTER" carved into the center of the cap. The cap is made up of a silver metal frame with ridges along the edges. The background is a dark gray color. The word is carved into a wood-like pattern. There are tall pine trees in the middle of the frame. There is a mountain range in the background.' |
|
output: |
|
url: images/LD1.png |
|
- text: 'Logo Design, a round, teal-colored metal sign is adorned with the words "RANGER" in bold, capitalized letters. The sign is set against a stark white backdrop, creating a striking contrast. The text, "PARK RANGER", is prominently displayed in the center of the sign, adding a pop of color to the otherwise monochromatic image.' |
|
output: |
|
url: images/LD2.png |
|
- text: 'Logo Design, An eye-level view of a dark gray house with the words "House Cleaning" written in the center of the image. The house is cracked, and the sky is filled with dark gray clouds. There are three spires sticking out of the top of the roof of the house, and there are small white stars scattered across the sky.' |
|
output: |
|
url: images/LD3.png |
|
base_model: black-forest-labs/FLUX.1-dev |
|
instance_prompt: Logo Design |
|
license: creativeml-openrail-m |
|
--- |
|
# Logo-Design-Flux-LoRA |
|
|
|
<Gallery /> |
|
|
|
**The model is still in the training phase. This is not the final version and may contain artifacts and perform poorly in some cases.** |
|
|
|
## Model description |
|
|
|
**prithivMLmods/Logo-Design-Flux-LoRA** |
|
|
|
Image Processing Parameters |
|
|
|
| Parameter | Value | Parameter | Value | |
|
|---------------------------|--------|---------------------------|--------| |
|
| LR Scheduler | constant | Noise Offset | 0.03 | |
|
| Optimizer | AdamW | Multires Noise Discount | 0.1 | |
|
| Network Dim | 64 | Multires Noise Iterations | 10 | |
|
| Network Alpha | 32 | Repeat & Steps | 22 & 2.2K| |
|
| Epoch | 10 | Save Every N Epochs | 1 | |
|
|
|
Labeling: florence2-en(natural language & English) |
|
|
|
Total Images Used for Training : 14 [ Hi-RES ] |
|
|
|
## Best Dimensions |
|
|
|
- 1024 x 1024 (Default) |
|
|
|
## Setting Up |
|
``` |
|
import torch |
|
from pipelines import DiffusionPipeline |
|
|
|
base_model = "black-forest-labs/FLUX.1-dev" |
|
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16) |
|
|
|
lora_repo = "prithivMLmods/Logo-Design-Flux-LoRA" |
|
trigger_word = "Logo Design" |
|
pipe.load_lora_weights(lora_repo) |
|
|
|
device = torch.device("cuda") |
|
pipe.to(device) |
|
``` |
|
|
|
## Data source |
|
|
|
- https://playground.com/ |
|
|
|
## Trigger words |
|
|
|
You should use `Logo Design` to trigger the image generation. |
|
|
|
|
|
## Download model |
|
|
|
Weights for this model are available in Safetensors format. |
|
|
|
[Download](/prithivMLmods/Logo-Design-Flux-LoRA/tree/main) them in the Files & versions tab. |
|
|
