Update app.py

app.py

@@ -177,6 +177,13 @@ with gr.Blocks(css=css) as demo:
             
         with gr.Row():
 
+            strength2 = gr.Slider(
+                    label="Strength",
+                    minimum=.1,
+                    maximum=1,
+                    step=0.1,
+                    value=.5,
+                )
             guidance_scale2 = gr.Slider(
                     label="Guidance Scale",
                     minimum=1,
@@ -209,23 +216,40 @@ with gr.Blocks(css=css) as demo:
                 return f"Failed to load image from URL: {e}"
         return None
 
-    @spaces.GPU
     def image2image(uploaded_image, image_url, use_generated=False):
-        image = select_image(uploaded_image, image_url, use_generated=False)
-        prompt = "one awesome dude"
-        generator = torch.Generator(device=device).manual_seed(1024)
-        image = pipeline2Image(prompt=prompt, image=image, strength=0.75, guidance_scale=7.5, generator=generator).images[0]
+        image = select_image(uploaded_image, image_url, use_generated=use_generated)
+        #prompt = "one awesome dude"
+        #generator = torch.Generator(device=device).manual_seed(1024)
+        #image = pipeline2Image(prompt=prompt, image=image, strength=0.75, guidance_scale=7.5, generator=generator).images[0]
         return image
     
     use_generated_image.click(fn=lambda: image2image(None, None, True), inputs=[], outputs=additional_image_output)
     uploaded_image.change(fn=image2image, inputs=[uploaded_image, image_url, gr.State(False)], outputs=additional_image_output)
     image_url.submit(fn=image2image, inputs=[uploaded_image, image_url, gr.State(False)], outputs=additional_image_output)
 
+    @spaces.GPU
+    def infer2(prompt, seed=42, image, randomize_seed=False, width=1024, height=1024, strength=.5, guidance_scale=5.0, num_inference_steps=28, progress=gr.Progress(track_tqdm=True)):
+        if randomize_seed:
+            seed = random.randint(0, MAX_SEED)
+        generator = torch.Generator(device=device).manual_seed(seed)
+        image = pipeline2Image(prompt=prompt, image=image, strength=strength, guidance_scale=guidance_scale, num_inference_steps=num_inference_steps, generator=generator).images[0]
+#    generator = torch.Generator().manual_seed(seed)
+#    image = pipe(
+#        prompt=prompt, 
+#        width=width,
+#        height=height,
+#        num_inference_steps=num_inference_steps, 
+#        generator=generator,
+#        guidance_scale=guidance_scale
+#    ).images[0] 
+        return image, seed
+    final_image_output = gr.Image(label="Final Image", show_label=False)
+    
     gr.on(
         triggers=[run2_button.click, prompt2.submit],
-        fn=infer,
-        inputs=[prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
-        outputs=[result, seed]
+        fn=infer2,
+        inputs=[prompt2, seed2, additional_image_output, randomize_seed2, width2, height2, guidance_scale2, num_inference_steps2],
+        outputs[final_image_output, seed2]
     )
 
 demo.launch()