Update app.py

app.py

@@ -1,3 +1,96 @@
+'''
+from gradio_client import Client, file
+from datasets import load_dataset
+import os
+from tqdm import tqdm
+from PIL import Image
+
+# Initialize Gradio client
+client = Client("http://localhost:7861")
+
+# Load the dataset
+dataset = load_dataset("svjack/Dont_be_your_lover_Images")
+
+# Create directories for saving images and results
+os.makedirs("Dont_be_your_lover_Images_Captioned", exist_ok=True)
+
+# Process each image in the dataset
+for i, item in enumerate(tqdm(dataset["train"], desc="Processing images")):
+    try:
+        # Get the PIL Image object
+        pil_image = item["image"]
+
+        # Save the image locally with 000i.png format
+        img_filename = f"{i:04d}.png"
+        img_path = os.path.join("Dont_be_your_lover_Images_Captioned", img_filename)
+        pil_image.save(img_path)
+
+        # Process the image through the API
+        result = client.predict(
+            input_image=file(img_path),
+            prompt="Write a long detailed description for this image.",
+            temperature=0.6,
+            top_p=0.9,
+            max_new_tokens=512,
+            log_prompt=True,
+            api_name="/chat_joycaption"
+        )
+
+        # Save the result as a text file with the same name
+        result_filename = f"{i:04d}.txt"
+        result_path = os.path.join("Dont_be_your_lover_Images_Captioned", result_filename)
+
+        with open(result_path, "w", encoding="utf-8") as f:
+            f.write(str(result))
+
+    except Exception as e:
+        print(f"Error processing image {i}: {str(e)}")
+        continue
+
+print("Processing complete!")
+
+# Load the dataset
+dataset = load_dataset("svjack/Origin_Images")
+
+# Create directories for saving images and results
+os.makedirs("Origin_Images_Captioned", exist_ok=True)
+
+# Process each image in the dataset
+for i, item in enumerate(tqdm(dataset["train"], desc="Processing images")):
+    try:
+        # Get the PIL Image object
+        pil_image = item["image"]
+
+        # Save the image locally with 000i.png format
+        img_filename = f"{i:04d}.png"
+        img_path = os.path.join("Origin_Images_Captioned", img_filename)
+        pil_image.save(img_path)
+
+        # Process the image through the API
+        result = client.predict(
+            input_image=file(img_path),
+            prompt="Write a long detailed description for this image.",
+            temperature=0.6,
+            top_p=0.9,
+            max_new_tokens=512,
+            log_prompt=True,
+            api_name="/chat_joycaption"
+        )
+
+        # Save the result as a text file with the same name
+        result_filename = f"{i:04d}.txt"
+        result_path = os.path.join("Origin_Images_Captioned", result_filename)
+
+        with open(result_path, "w", encoding="utf-8") as f:
+            f.write(str(result))
+
+    except Exception as e:
+        print(f"Error processing image {i}: {str(e)}")
+        continue
+
+print("Processing complete!")
+'''
+
 import spaces
 import gradio as gr
 from transformers import LlavaForConditionalGeneration, TextIteratorStreamer, AutoProcessor