File size: 2,181 Bytes
4407894 bd65e17 4407894 a582fdf 4407894 1c8afdf 4407894 1c8afdf a582fdf 4407894 1de8b6c d868d8f 4407894 9ddbd7b d868d8f 4407894 7e55d61 547eb1a a582fdf |
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 |
from PIL import Image
import PIL.Image
import cv2
import gradio as gr
import numpy as np
import torch, os, random
from accelerate import Accelerator
from transformers import pipeline
from diffusers.utils import load_image
from diffusers import KandinskyV22PriorPipeline, KandinskyV22Pipeline
accelerator = Accelerator(cpu=True)
pope_prior = accelerator.prepare(KandinskyV22PriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float32))
pope_prior = pope_prior.to("cpu")
pope = accelerator.prepare(KandinskyV22Pipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder", torch_dtype=torch.float32))
pope = pope.to("cpu")
generator = torch.Generator(device="cpu").manual_seed(random.randint(1, 4876364))
def plex(img, cook, one, two, three):
goof = load_image(img).resize((512, 512))
prompt = cook
negative_prior_prompt = "lowres,text,bad quality,low quality,jpeg artifacts,ugly,bad hands,bad face,blurry,bad eyes,watermark,signature"
img_emb = pope_prior(prompt=prompt, guidance_scale=0.85, num_inference_steps=5, generator=generator)
negative_emb = pope_prior(prompt=negative_prior_prompt, guidance_scale=1, num_inference_steps=5, generator=generator)
imags = pope(image_embeds=img_emb.image_embeds,negative_image_embeds=negative_emb.image_embeds,num_inference_steps=5,generator=generator,height=512,width=512,).images[0]
images_texts = [cook, goof, imags]
weights = [one, two, three]
primpt = ""
prior_out = pope_prior.interpolate(images_texts, weights, num_inference_steps=5)
imas = pope(**prior_out, height=512, width=512, num_inference_steps=5).images[0]
return imas
iface = gr.Interface(fn=plex,inputs=[gr.Image(label="drop", type="filepath"), gr.Textbox(label="prompt"), gr.Slider(label="Text Guide",minimum=0.01,step=0.01,maximum=0.99,value=0.5), gr.Slider(label="Your Image Guide",minimum=0.01,step=0.01,maximum=0.99,value=0.5),gr.Slider(label="Generated Image Guide",minimum=0.01,step=0.01,maximum=0.99,value=0.3)], outputs=gr.Image(), title="Ksky22 Cntrl Gdd Interp", description="ksky22 Cntrl Gdd Interp")
iface.queue(max_size=1,api_open=False)
iface.launch(max_threads=1) |