Update app.py

app.py

@@ -1,43 +1,25 @@
 import gradio as gr
-import wget
 from transformers import pipeline
 from diffusers import StableDiffusionPipeline
 import torch
-import time
+import wget
 
 # Define the device to use (either "cuda" for GPU or "cpu" for CPU)
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
-# Function to load the models
-def load_models():
-    global caption_image, sd_pipeline, translator
-    start_time = time.time()
-    
-    # Load the image captioning model
-    caption_image = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large", device=device)
-    print(f"Caption model loaded in {time.time() - start_time:.2f} seconds")
-
-    # Load the Stable Diffusion model with low CPU memory usage
-    sd_pipeline = StableDiffusionPipeline.from_pretrained(
-        "runwayml/stable-diffusion-v1-5", 
-        low_cpu_mem_usage=True  # Enable low CPU memory usage
-    ).to(device)
-    print(f"Stable Diffusion model loaded in {time.time() - start_time:.2f} seconds")
-
-    # Load the translation model
-    translator = pipeline(
-        task="translation",
-        model="facebook/nllb-200-distilled-600M",
-        device=device
-    )
-    print(f"Translator model loaded in {time.time() - start_time:.2f} seconds")
-
 # Load the models
-load_models()
-
-# Download the images
-url1 = "https://github.com/Shahad-b/Image-database/blob/main/sea.jpg?raw=true"
-sea = wget.download(url1)
+# Image captioning model to generate captions from uploaded images
+caption_image = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large", device=device)
+# Stable Diffusion model for generating new images based on captions
+sd_pipeline = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5").to(device)
+
+# Load the translation model (English to Arabic)
+translator = pipeline(
+    task="translation",
+    model="facebook/nllb-200-distilled-600M",
+    torch_dtype=torch.bfloat16,
+    device=device
+)
 
 # Function to generate images based on the image's caption
 def generate_image_and_translate(image, num_images=1):
@@ -59,23 +41,22 @@ def generate_image_and_translate(image, num_images=1):
 
 # Set up the Gradio interface
 interface = gr.Interface(
-    fn=generate_image_and_translate,
+    fn=generate_image_and_translate,   # Function to call when processing input
     inputs=[
-        gr.Image(type="pil", label="Upload Image"),
-        gr.Slider(minimum=1, maximum=10, label="Number of Images", value=1, step=1)
+        gr.Image(type="pil", label="Upload Image"), # Input for image upload
+        gr.Slider(minimum=1, maximum=10, label="Number of Images", value=1, step=1) # Slider to select number of images
     ],
     outputs=[
-        gr.Gallery(label="Generated Images"),
-        gr.Textbox(label="Generated Caption (English)", interactive=False),
-        gr.Textbox(label="Translated Caption (Arabic)", interactive=False)
+        gr.Gallery(label="Generated Images"),  # Output for displaying generated images
+        gr.Textbox(label="Generated Caption (English)", interactive=False),  # Output for English caption
+        gr.Textbox(label="Translated Caption (Arabic)", interactive=False)# Output for Arabic caption
+
     ],
-    title="Image Generation and Translation",
-    description="Upload an image to generate new images based on its caption and translate the caption into Arabic.",
-    examples=[
-        ["sea.jpg", 3]
-    ]
+    title="Image Generation and Translation",  # Title of the interface
+    description="Upload an image to generate new images based on its caption and translate the caption into Arabic.",  # Description
 )
 
+
 # Launch the Gradio application within the main guard
 if __name__ == "__main__":
     interface.launch()