changed the concepts to 768-embeddings

app.py

@@ -10,13 +10,13 @@ import os
 model_id = "CompVis/stable-diffusion-v1-4"
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
-# List of concept embeddings to use
+# List of concept embeddings compatible with SD v1.4
 concepts = [
-    "sd-concepts-library/sword-lily-flowers102",
-    "sd-concepts-library/azalea-flowers102",
-    "sd-concepts-library/samurai-jack",
-    "sd-concepts-library/wu-shi-art",
-    "sd-concepts-library/wu-shi"
+    "sd-concepts-library/cat-toy",
+    "sd-concepts-library/disco-diffusion-style",
+    "sd-concepts-library/modern-disney-style",
+    "sd-concepts-library/charliebo-artstyle",
+    "sd-concepts-library/redshift-render-style"
 ]
 
 def download_concept_embedding(concept_name):
@@ -50,6 +50,7 @@ def load_learned_embed_in_clip(learned_embeds_path, text_encoder, tokenizer):
 
 def generate_images(prompt):
     images = []
+    failed_concepts = []
     
     # Load base pipeline
     pipe = StableDiffusionPipeline.from_pretrained(
@@ -58,12 +59,13 @@ def generate_images(prompt):
     ).to(device)
     
     for concept in concepts:
-        # Download and load concept embedding
-        embed_path = download_concept_embedding(concept)
-        if embed_path is None:
-            continue
-            
         try:
+            # Download and load concept embedding
+            embed_path = download_concept_embedding(concept)
+            if embed_path is None:
+                failed_concepts.append(concept)
+                continue
+                
             token = load_learned_embed_in_clip(
                 embed_path,
                 pipe.text_encoder,
@@ -81,7 +83,7 @@ def generate_images(prompt):
             with autocast(device):
                 image = pipe(
                     concept_prompt,
-                    num_inference_steps=50,
+                    num_inference_steps=40,
                     generator=generator,
                     guidance_scale=7.5
                 ).images[0]
@@ -94,17 +96,24 @@ def generate_images(prompt):
             
         except Exception as e:
             print(f"Error processing concept {concept}: {str(e)}")
+            failed_concepts.append(concept)
             continue
     
-    return images if images else [None] * 5
+    if failed_concepts:
+        print(f"Failed to process concepts: {', '.join(failed_concepts)}")
+    
+    # Return available images, pad with None if some failed
+    while len(images) < 5:
+        images.append(None)
+    return images[:5]
 
 # Create Gradio interface
 iface = gr.Interface(
     fn=generate_images,
     inputs=gr.Textbox(label="Enter your prompt"),
-    outputs=[gr.Image() for _ in range(5)],
+    outputs=[gr.Image(label=f"Concept {i+1}") for i in range(5)],
     title="Multi-Concept Stable Diffusion Generator",
-    description="Generate images using 5 different concepts from the SD Concepts Library"
+    description="Generate images using 5 different artistic concepts from the SD Concepts Library"
 )
 
 # Launch the app