Update app.py

app.py

@@ -1,23 +1,24 @@
 import torch
-
+import gradio as gr
 from diffusers import UniDiffuserPipeline
 from diffusers.utils import load_image
+from accelerate import Accelerator
+Accelerator = Accelerator(cpu=True)
 
-device = "cuda"
-model_id_or_path = "thu-ml/unidiffuser-v1"
-pipe = UniDiffuserPipeline.from_pretrained(model_id_or_path, torch_dtype=torch.float16)
-pipe.to(device)
+pipe = accelerator.prepare(UniDiffuserPipeline.from_pretrained("thu-ml/unidiffuser-v1", torch_dtype=torch.bfloat16))
+pipe = accelerator.prepare(pipe.to("cpu"))
 
-# Image variation can be performed with an image-to-text generation followed by a text-to-image generation:
-# 1. Image-to-text generation
-image_url = "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/unidiffuser/unidiffuser_example_image.jpg"
+def plex(image_url,stips)
 init_image = load_image(image_url).resize((512, 512))
 
-sample = pipe(image=init_image, num_inference_steps=20, guidance_scale=8.0)
+sample = pipe(image=init_image, num_inference_steps=stips, guidance_scale=8.0)
 i2t_text = sample.text[0]
-print(i2t_text)
 
-# 2. Text-to-image generation
-sample = pipe(prompt=i2t_text, num_inference_steps=20, guidance_scale=8.0)
-final_image = sample.images[0]
-final_image.save("unidiffuser_image_variation_sample.png")
+sample = pipe(prompt=i2t_text, num_inference_steps=stips, guidance_scale=8.0)
+    for i, imge in enumerate(sample["images"]):
+        apol.append(imge)
+    return apol
+
+iface = gr.Interface(fn=plex, inputs=[gr.Image(label="img",type="filepath"), gr.Slider(label="num inference steps", minimum=1, step=1, maximum=5, value=5)], outputs=gr.Gallery(label="out", columns=2))
+iface.queue(max_size=1)
+iface.launch(max_threads=1)