Update app.py

app.py

@@ -1,187 +1,188 @@
-import gradio as gr
-import numpy as np
-import random
-#from diffusers import DiffusionPipeline
-from diffusers import AutoPipelineForText2Image, AutoPipelineForImage2Image
-#from diffusers.utils import load_image
-import torch
-from PIL import Image
-
-
-modelPath = "stabilityai/sdxl-turbo"
- 
-if torch.cuda.is_available():
-    device = "cuda"
-    torch.cuda.max_memory_allocated(device=device)
-    #pipe = DiffusionPipeline.from_pretrained(modelPath, torch_dtype=torch.float16, variant="fp16", use_safetensors=True)
-    pipeTex2Image = AutoPipelineForText2Image.from_pretrained(modelPath, torch_dtype=torch.float16, variant="fp16")
-    pipeImage2Image = AutoPipelineForImage2Image.from_pretrained(modelPath, torch_dtype=torch.float16, variant="fp16")
-    #pipe.enable_xformers_memory_efficient_attention()
-    pipeTex2Image.enable_xformers_memory_efficient_attention() 
-    pipeImage2Image.enable_xformers_memory_efficient_attention()
-    #pipe = pipe.to(device)
-else:
-    device = "cpu"
-    #pipe = DiffusionPipeline.from_pretrained(modelPath, use_safetensors=True)
-    pipeTex2Image = AutoPipelineForText2Image.from_pretrained(modelPath, use_safetensors=True)
-    pipeImage2Image = AutoPipelineForImage2Image.from_pretrained(modelPath, use_safetensors=True)
-    #pipe = pipe.to(device)
-
-#pipe = pipe.to(device)    
-pipeTex2Image.to(device)
-pipeImage2Image.to(device)
-
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024
-
-def infer(prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, use_as_input, strength, image):
-
-    if randomize_seed:
-        seed = random.randint(0, MAX_SEED)
-        
-    generator = torch.Generator().manual_seed(seed)
-    
-    if use_as_input:
-        print("Image to Image:")
-        pipe = pipeImage2Image
-        init_image = Image.fromarray(np.uint8(image)).resize((width, height)).convert("RGB")
-        init_image.save("input.png", format="PNG")
-        print(type(init_image), init_image.size)
-        image = pipe(
-            prompt = prompt, 
-            negative_prompt = negative_prompt,
-            guidance_scale = guidance_scale, 
-            num_inference_steps = num_inference_steps, 
-            width = width, 
-            height = height,
-            generator = generator,
-            strength=strength,
-            image=init_image
-        ).images[0] 
-    else:
-        print("Text to Image:")
-        pipe = pipeTex2Image
-        image = pipe(
-            prompt = prompt, 
-            negative_prompt = negative_prompt,
-            guidance_scale = guidance_scale, 
-            num_inference_steps = num_inference_steps, 
-            width = width, 
-            height = height,
-            generator = generator
-        ).images[0] 
-    
-    return image
-
-examples = [
-    "Face  of a modern woman of Balkan descent 25 years old",
-    "Blue car sandero stepway on dirt road",
-    "Cow in the skin of a dog of dalmatian breed",
-]
-
-css="""
-#col-container {
-    margin: 0 auto;
-    max-width: auto;
-}
-"""
-
-
-with gr.Blocks(css=css) as app:
-    
-    with gr.Column(elem_id="col-container"):
-        gr.Markdown(f"""
-        # Text-to-Image, Image-to-Image by Slavko Novak
-        Currently running on {device}.
-        """)
-        
-        with gr.Row():
-            
-            prompt = gr.Text(
-                label="Prompt",
-                show_label=False,
-                max_lines=1,
-                placeholder="Enter your prompt",
-                container=False,
-            )
-            
-            run_button = gr.Button("Generate", scale=0)
-        
-        result = gr.Image(label="Result", show_label=False)
-        use_as_input = gr.Checkbox(label="Use image as input", value=False)
-
-        with gr.Accordion("Advanced Settings", open=False):
-            
-            negative_prompt = gr.Text(
-                label="Negative prompt",
-                max_lines=1,
-                placeholder="Enter a negative prompt",
-                visible=True,
-            )
-            
-            seed = gr.Slider(
-                label="Seed",
-                minimum=0,
-                maximum=MAX_SEED,
-                step=1,
-                value=0,
-            )
-            
-            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
-            
-            with gr.Row():
-                
-                width = gr.Slider(
-                    label="Width",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=512,
-                )
-                
-                height = gr.Slider(
-                    label="Height",
-                    minimum=256,
-                    maximum=MAX_IMAGE_SIZE,
-                    step=32,
-                    value=512,
-                )
-            
-            with gr.Row():
-                
-                guidance_scale = gr.Slider(
-                    label="Guidance scale",
-                    minimum=0.0,
-                    maximum=10.0,
-                    step=0.1,
-                    value=0.0,
-                )
-                
-                strength = gr.Slider(
-                    label="Strength scale",
-                    minimum=0.0,
-                    maximum=1.0,
-                    step=0.1,
-                    value=0.5,
-                )
-                
-                num_inference_steps = gr.Slider(
-                    label="Number of inference steps",
-                    minimum=1,
-                    maximum=12,
-                    step=1,
-                    value=2,
-                )
-        
-        gr.Examples(
-            examples = examples,
-            inputs = [prompt]
-        )
-
-    run_button.click(
-        fn = infer,
-        inputs = [prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, use_as_input, strength, result],
-        outputs = [result]
-    )
-
-app.queue().launch(server_name="0.0.0.0", server_port=8080, share=True)
+import gradio as gr
+import numpy as np
+import random
+#from diffusers import DiffusionPipeline
+from diffusers import AutoPipelineForText2Image, AutoPipelineForImage2Image
+#from diffusers.utils import load_image
+import torch
+from PIL import Image
+
+
+modelPath = "stabilityai/sdxl-turbo"
+ 
+if torch.cuda.is_available():
+    device = "cuda"
+    torch.cuda.max_memory_allocated(device=device)
+    #pipe = DiffusionPipeline.from_pretrained(modelPath, torch_dtype=torch.float16, variant="fp16", use_safetensors=True)
+    pipeTex2Image = AutoPipelineForText2Image.from_pretrained(modelPath, torch_dtype=torch.float16, variant="fp16")
+    pipeImage2Image = AutoPipelineForImage2Image.from_pretrained(modelPath, torch_dtype=torch.float16, variant="fp16")
+    #pipe.enable_xformers_memory_efficient_attention()
+    pipeTex2Image.enable_xformers_memory_efficient_attention() 
+    pipeImage2Image.enable_xformers_memory_efficient_attention()
+    #pipe = pipe.to(device)
+else:
+    device = "cpu"
+    #pipe = DiffusionPipeline.from_pretrained(modelPath, use_safetensors=True)
+    pipeTex2Image = AutoPipelineForText2Image.from_pretrained(modelPath, use_safetensors=True)
+    pipeImage2Image = AutoPipelineForImage2Image.from_pretrained(modelPath, use_safetensors=True)
+    #pipe = pipe.to(device)
+
+#pipe = pipe.to(device)    
+pipeTex2Image.to(device)
+pipeImage2Image.to(device)
+
+MAX_SEED = np.iinfo(np.int32).max
+MAX_IMAGE_SIZE = 1024
+
+def infer(prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, use_as_input, strength, image):
+
+    if randomize_seed:
+        seed = random.randint(0, MAX_SEED)
+        
+    generator = torch.Generator().manual_seed(seed)
+    
+    if use_as_input:
+        print("Image to Image:")
+        pipe = pipeImage2Image
+        init_image = Image.fromarray(np.uint8(image)).resize((width, height)).convert("RGB")
+        init_image.save("input.png", format="PNG")
+        print(type(init_image), init_image.size)
+        image = pipe(
+            prompt = prompt, 
+            negative_prompt = negative_prompt,
+            guidance_scale = guidance_scale, 
+            num_inference_steps = num_inference_steps, 
+            width = width, 
+            height = height,
+            generator = generator,
+            strength=strength,
+            image=init_image
+        ).images[0] 
+    else:
+        print("Text to Image:")
+        pipe = pipeTex2Image
+        image = pipe(
+            prompt = prompt, 
+            negative_prompt = negative_prompt,
+            guidance_scale = guidance_scale, 
+            num_inference_steps = num_inference_steps, 
+            width = width, 
+            height = height,
+            generator = generator
+        ).images[0] 
+    
+    return image
+
+examples = [
+    "Face  of a modern woman of Balkan descent 25 years old",
+    "Blue car sandero stepway on dirt road",
+    "Cow in the skin of a dog of dalmatian breed",
+]
+
+css="""
+#col-container {
+    margin: 0 auto;
+    max-width: auto;
+}
+"""
+
+
+with gr.Blocks(css=css) as app:
+    
+    with gr.Column(elem_id="col-container"):
+        gr.Markdown(f"""
+        # Text-to-Image, Image-to-Image by Slavko Novak
+        Currently running on {device}.
+        """)
+        
+        with gr.Row():
+            
+            prompt = gr.Text(
+                label="Prompt",
+                show_label=False,
+                max_lines=1,
+                placeholder="Enter your prompt",
+                container=False,
+            )
+            
+            run_button = gr.Button("Generate", scale=0)
+        
+        result = gr.Image(label="Result", show_label=False)
+        use_as_input = gr.Checkbox(label="Use image as input", value=False)
+
+        with gr.Accordion("Advanced Settings", open=False):
+            
+            negative_prompt = gr.Text(
+                label="Negative prompt",
+                max_lines=1,
+                placeholder="Enter a negative prompt",
+                visible=True,
+            )
+            
+            seed = gr.Slider(
+                label="Seed",
+                minimum=0,
+                maximum=MAX_SEED,
+                step=1,
+                value=0,
+            )
+            
+            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+            
+            with gr.Row():
+                
+                width = gr.Slider(
+                    label="Width",
+                    minimum=256,
+                    maximum=MAX_IMAGE_SIZE,
+                    step=32,
+                    value=512,
+                )
+                
+                height = gr.Slider(
+                    label="Height",
+                    minimum=256,
+                    maximum=MAX_IMAGE_SIZE,
+                    step=32,
+                    value=512,
+                )
+            
+            with gr.Row():
+                
+                guidance_scale = gr.Slider(
+                    label="Guidance scale",
+                    minimum=0.0,
+                    maximum=10.0,
+                    step=0.1,
+                    value=0.0,
+                )
+                
+                strength = gr.Slider(
+                    label="Strength scale",
+                    minimum=0.0,
+                    maximum=1.0,
+                    step=0.1,
+                    value=0.5,
+                )
+                
+                num_inference_steps = gr.Slider(
+                    label="Number of inference steps",
+                    minimum=1,
+                    maximum=12,
+                    step=1,
+                    value=2,
+                )
+        
+        gr.Examples(
+            examples = examples,
+            inputs = [prompt]
+        )
+
+    run_button.click(
+        fn = infer,
+        inputs = [prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, use_as_input, strength, result],
+        outputs = [result]
+    )
+
+#app.queue().launch(server_name="0.0.0.0", server_port=8080, share=True)
+app.queue().launch()