File size: 2,448 Bytes
5cd29b0 cfca7b8 34df01d 8ba122d 5cd29b0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 |
import torch
from diffusers import StableDiffusionPipeline, DDIMScheduler
import gradio as gr
pipe = StableDiffusionPipeline.from_pretrained("MVRL/GeoSynth").cuda()
pipe.enable_xformers_memory_efficient_attention()
# scheduler = DDIMScheduler.from_pretrained("stabilityai/stable-diffusion-2-1-base")
# pipe.scheduler = scheduler
def process(prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta):
generator = torch.manual_seed(seed)
output_images = pipe(prompt,
height=image_resolution,
width=image_resolution,
num_inference_steps=ddim_steps,
guidance_scale=scale,
negative_prompt=n_prompt,
num_images_per_prompt=num_samples,
eta=eta,
generator=generator,
).images
return output_images
block = gr.Blocks().queue()
with block:
with gr.Row():
gr.Markdown(
"""
# GeoSynth: Contextually-Aware High-Resolution Satellite Image Synthesis
Srikumar Sastry*, Subash Khanal, Aayush Dhakal, Nathan Jacobs (*Corresponding Author)<br>
"""
)
with gr.Row():
with gr.Column():
prompt = gr.Textbox(label="Prompt")
run_button = gr.Button(value="Run")
with gr.Accordion("Advanced options", open=True):
num_samples = gr.Slider(label="Images", minimum=1, maximum=12, value=1, step=1)
image_resolution = gr.Slider(label="Image Resolution", minimum=256, maximum=768, value=512, step=64)
ddim_steps = gr.Slider(label="Steps", minimum=1, maximum=100, value=20, step=1)
scale = gr.Slider(label="Guidance Scale", minimum=0.1, maximum=30.0, value=7.5, step=0.1)
seed = gr.Slider(label="Seed", minimum=-1, maximum=2147483647, step=1, randomize=True)
eta = gr.Number(label="eta (DDIM)", value=0.0)
n_prompt = gr.Textbox(label="Negative Prompt",
value='')
with gr.Column():
result_gallery = gr.Gallery(label='Output', show_label=False, elem_id="gallery")
ips = [prompt, n_prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta]
run_button.click(fn=process, inputs=ips, outputs=[result_gallery])
block.launch() |