Update app.py

app.py

@@ -52,45 +52,46 @@ pipe.to(device)
 
 @spaces.GPU(duration=150)
 def generate_caption_and_image(image, f, p, d):
-    img = image.convert("RGB")
-    # reader = easyocr.Reader(['en'])  
-    # # result = reader.readtext(img)
-    import random
+    if f and p and d:
+        img = image.convert("RGB")
+        # reader = easyocr.Reader(['en'])  
+        # # result = reader.readtext(img)
+        import random
+        
     
-
-
-
-
+    
+    
+    
+            
+    
+     
+        text = "a picture of "
+        inputs = processor(img, text, return_tensors="pt").to(device)
+        
+        out = model2.generate(**inputs, num_beams = 3)
+       
         
-
- 
-    text = "a picture of "
-    inputs = processor(img, text, return_tensors="pt").to(device)
     
-    out = model2.generate(**inputs, num_beams = 3)
-   
+        caption2 = processor.decode(out[0], skip_special_tokens=True)
+        
+        # Generate caption
+        inputs = processor(image, return_tensors="pt", padding=True, truncation=True, max_length=250)
+        inputs = {key: val.to(device) for key, val in inputs.items()}
+        out = model.generate(**inputs)
+        caption1 = processor.decode(out[0], skip_special_tokens=True)
     
-
-    caption2 = processor.decode(out[0], skip_special_tokens=True)
+        prompt = f"Generate a clothing item using the following details: 1. {caption1} 2. {caption2} 3. Fabric: {f} 4. Pattern: {p} 5. Design Style: {d}. "
     
-    # Generate caption
-    inputs = processor(image, return_tensors="pt", padding=True, truncation=True, max_length=250)
-    inputs = {key: val.to(device) for key, val in inputs.items()}
-    out = model.generate(**inputs)
-    caption1 = processor.decode(out[0], skip_special_tokens=True)
-
-    prompt = f"Generate a clothing item using the following details: 1. {caption1} 2. {caption2} 3. Fabric: {selected_fabric} 4. Pattern: {selected_pattern} 5. Design Style: {selected_textile_design}. "
-
-  
-    prompt +="The image should have a clean, minimalistic grey or white background, with realistic lighting and fine details, ensuring a sophisticated and polished appearance"
-
-
-  
-
-    # Generate image based on the caption
-    generated_image = pipe(prompt).images[0]
-
-    return prompt, generated_image
+      
+        prompt +="The image should have a clean, minimalistic grey or white background, with realistic lighting and fine details, ensuring a sophisticated and polished appearance"
+    
+    
+      
+    
+        # Generate image based on the caption
+        generated_image = pipe(prompt).images[0]
+    
+        return prompt, generated_image
 
 # Gradio UI
 iface = gr.Interface(