Spaces:

ysharma
/

Low-rank-Adaptation

Configuration error

App Files Files Community

ysharma HF staff commited on Dec 10, 2022

Commit

0e7df92

•

1 Parent(s): ba8b2ea

update description

Browse files

Files changed (1) hide show

app.py +1 -1

app.py CHANGED Viewed

@@ -58,7 +58,7 @@ def accelerate_train_lora(steps):
 with gr.Blocks() as demo:
     gr.Markdown("""<h1><center>LORA - Low-rank Adaptation for Fast Text-to-Image Diffusion Fine-tuning</center></h1>
     """)
-     gr.Markdown(
         """**Main Features**<br>- Fine-tune Stable diffusion models twice as faster than dreambooth method, by Low-rank Adaptation.<br>- Get insanely small end result, easy to share and download.<br>- Easy to use, compatible with diffusers.<br>- Sometimes even better performance than full fine-tuning<br>Please refer the Github repo this Space is based on, here - <a href = "https://github.com/cloneofsimo/lora">LORA</a><br>You can also refer this tweet by AK over here to quote/retweet/like, here on <a href="https://twitter.com/_akhaliq/status/1601120767009513472">Twitter</a>.<br>This Gradio Space is an attempt to explore this novel LORA approach to fine-tune Stable diffusion models, using the power and flexibility of Gradio!<br><b>To use this Space well:</b>- First, upload your set of images (4-5), then enter the number of fine-tuning steps, and then press the 'Train LORA model' button.<br>- Enter a prompt, then set the alpha value using the Slider (nearer to 1 implies overfitting to the uploaded images), and then press the 'Inference' button.<br><b>Bonus:</b>Download your fine-tuned model weights from the Gradio file component. The smaller size of LORA models (around 3-4 mb files) is the main highlight of this 'Low-rank Adaptation' approach of fine-tuning.""")
     with gr.Row():

 with gr.Blocks() as demo:
     gr.Markdown("""<h1><center>LORA - Low-rank Adaptation for Fast Text-to-Image Diffusion Fine-tuning</center></h1>
     """)
+    gr.Markdown(
         """**Main Features**<br>- Fine-tune Stable diffusion models twice as faster than dreambooth method, by Low-rank Adaptation.<br>- Get insanely small end result, easy to share and download.<br>- Easy to use, compatible with diffusers.<br>- Sometimes even better performance than full fine-tuning<br>Please refer the Github repo this Space is based on, here - <a href = "https://github.com/cloneofsimo/lora">LORA</a><br>You can also refer this tweet by AK over here to quote/retweet/like, here on <a href="https://twitter.com/_akhaliq/status/1601120767009513472">Twitter</a>.<br>This Gradio Space is an attempt to explore this novel LORA approach to fine-tune Stable diffusion models, using the power and flexibility of Gradio!<br><b>To use this Space well:</b>- First, upload your set of images (4-5), then enter the number of fine-tuning steps, and then press the 'Train LORA model' button.<br>- Enter a prompt, then set the alpha value using the Slider (nearer to 1 implies overfitting to the uploaded images), and then press the 'Inference' button.<br><b>Bonus:</b>Download your fine-tuned model weights from the Gradio file component. The smaller size of LORA models (around 3-4 mb files) is the main highlight of this 'Low-rank Adaptation' approach of fine-tuning.""")
     with gr.Row():