Update app.py

app.py

@@ -4,281 +4,271 @@ from huggingface_hub import hf_hub_download
 from PIL import Image
 import zipfile
 import os
-import random
 
-# Global variables
-df = None
-images_dir = None
+# ==================== Configuration ====================
+REPO_ID = "Deva8/Generative-VQA-V2-Curated"
+CACHE_DIR = "./dataset_cache"
 
-def setup_dataset():
-    """Download and setup the dataset (called once on startup)"""
-    global df, images_dir
-    
-    print("Loading metadata...")
-    # Load metadata
-    csv_path = hf_hub_download(
-        repo_id="Deva8/Generative-VQA-V2-Curated",
-        filename="main_metadata.csv",
-        repo_type="dataset"
-    )
-    df = pd.read_csv(csv_path)
-    
-    print("Downloading images zip (this may take a few minutes)...")
-    # Download zip file
-    zip_path = hf_hub_download(
-        repo_id="Deva8/Generative-VQA-V2-Curated",
-        filename="gen_vqa_v2-images.zip",
-        repo_type="dataset"
-    )
-    
-    # Extract images
-    images_dir = "./extracted_images"
-    if not os.path.exists(images_dir):
-        print("Extracting images...")
-        os.makedirs(images_dir, exist_ok=True)
-        with zipfile.ZipFile(zip_path, 'r') as zip_ref:
-            zip_ref.extractall(images_dir)
-    
-    print(f"✓ Dataset ready! {len(df)} examples loaded.")
-    return f"Dataset loaded successfully! {len(df):,} examples available."
+# Global state
+dataset_state = {
+    "df": None,
+    "images_dir": None,
+    "loaded": False,
+    "error": None
+}
 
-def get_random_sample():
-    """Get a random sample from the dataset"""
-    if df is None:
-        return None, "Please wait, dataset is loading...", "", ""
-    
-    # Get random row
-    sample = df.sample(1).iloc[0]
-    
-    # Load image
-    img_path = os.path.join(images_dir, sample['file_name'])
-    img = Image.open(img_path)
-    
-    question = sample['question']
-    answer = sample['answer']
-    metadata = f"Image ID: {sample['image_id']} | Question ID: {sample['question_id']}"
-    
-    return img, question, answer, metadata
+# ==================== Dataset Loading ====================
 
-def search_by_question(query):
-    """Search for questions containing the query"""
-    if df is None:
-        return None, "Dataset not loaded yet", "", ""
+def initialize_dataset():
+    """Download and setup the dataset"""
+    if dataset_state["loaded"]:
+        return
+    
+    try:
+        print("📥 Downloading metadata...")
+        csv_path = hf_hub_download(
+            repo_id=REPO_ID,
+            filename="main_metadata.csv",
+            repo_type="dataset",
+            cache_dir=CACHE_DIR
+        )
+        dataset_state["df"] = pd.read_csv(csv_path)
+        print(f"✓ Loaded {len(dataset_state['df']):,} examples")
+        
+        print("📦 Downloading images (10GB, please wait)...")
+        zip_path = hf_hub_download(
+            repo_id=REPO_ID,
+            filename="gen_vqa_v2-images.zip",
+            repo_type="dataset",
+            cache_dir=CACHE_DIR
+        )
+        
+        dataset_state["images_dir"] = os.path.join(CACHE_DIR, "extracted")
+        if not os.path.exists(dataset_state["images_dir"]):
+            print("📂 Extracting images...")
+            os.makedirs(dataset_state["images_dir"], exist_ok=True)
+            with zipfile.ZipFile(zip_path, 'r') as zf:
+                zf.extractall(dataset_state["images_dir"])
+        
+        dataset_state["loaded"] = True
+        print("✅ Dataset ready!")
+        
+    except Exception as e:
+        dataset_state["error"] = str(e)
+        print(f"❌ Error: {e}")
+
+# Load dataset on startup
+initialize_dataset()
+
+# ==================== Helper Functions ====================
+
+def load_image(file_path):
+    """Load image with error handling"""
+    try:
+        full_path = os.path.join(dataset_state["images_dir"], file_path)
+        return Image.open(full_path).convert('RGB')
+    except Exception as e:
+        print(f"Error loading image: {e}")
+        return None
+
+def check_dataset():
+    """Check if dataset is loaded"""
+    if not dataset_state["loaded"]:
+        msg = dataset_state["error"] if dataset_state["error"] else "Dataset is loading..."
+        return None, f"⏳ {msg}", "", ""
+
+# ==================== Main Functions ====================
+
+def show_random():
+    """Display a random example"""
+    check = check_dataset()
+    if check:
+        return check
+    
+    sample = dataset_state["df"].sample(1).iloc[0]
+    img = load_image(sample['file_name'])
+    
+    return (
+        img,
+        sample['question'],
+        sample['answer'],
+        f"Image ID: {sample['image_id']} | Question ID: {sample['question_id']}"
+    )
+
+def search_question(query):
+    """Search by question keywords"""
+    check = check_dataset()
+    if check:
+        return check
     
-    if not query or len(query.strip()) < 3:
-        return None, "Please enter at least 3 characters to search", "", ""
+    if not query or len(query.strip()) < 2:
+        return None, "Enter at least 2 characters", "", ""
     
-    # Search for matching questions
-    matches = df[df['question'].str.contains(query, case=False, na=False)]
+    matches = dataset_state["df"][
+        dataset_state["df"]['question'].str.contains(query, case=False, na=False)
+    ]
     
     if len(matches) == 0:
-        return None, f"No questions found containing '{query}'", "", ""
+        return None, f"No matches for '{query}'", "", ""
     
-    # Get random match
     sample = matches.sample(1).iloc[0]
+    img = load_image(sample['file_name'])
     
-    # Load image
-    img_path = os.path.join(images_dir, sample['file_name'])
-    img = Image.open(img_path)
-    
-    question = sample['question']
-    answer = sample['answer']
-    metadata = f"Image ID: {sample['image_id']} | Question ID: {sample['question_id']} | Found {len(matches)} matches"
-    
-    return img, question, answer, metadata
+    return (
+        img,
+        sample['question'],
+        sample['answer'],
+        f"Found {len(matches):,} matches | Showing random example"
+    )
 
-def search_by_answer(query):
-    """Search for specific answers"""
-    if df is None:
-        return None, "Dataset not loaded yet", "", ""
+def search_answer(query):
+    """Search by answer"""
+    check = check_dataset()
+    if check:
+        return check
     
-    if not query or len(query.strip()) < 1:
-        return None, "Please enter an answer to search", "", ""
+    if not query:
+        return None, "Enter an answer", "", ""
     
-    # Search for matching answers
-    matches = df[df['answer'].str.lower() == query.lower().strip()]
+    matches = dataset_state["df"][
+        dataset_state["df"]['answer'].str.lower() == query.lower().strip()
+    ]
     
     if len(matches) == 0:
-        return None, f"No examples found with answer '{query}'", "", ""
+        return None, f"No examples with answer '{query}'", "", ""
     
-    # Get random match
     sample = matches.sample(1).iloc[0]
+    img = load_image(sample['file_name'])
     
-    # Load image
-    img_path = os.path.join(images_dir, sample['file_name'])
-    img = Image.open(img_path)
-    
-    question = sample['question']
-    answer = sample['answer']
-    metadata = f"Image ID: {sample['image_id']} | Question ID: {sample['question_id']} | Found {len(matches)} examples with this answer"
-    
-    return img, question, answer, metadata
+    return (
+        img,
+        sample['question'],
+        sample['answer'],
+        f"Found {len(matches):,} examples | Showing random"
+    )
 
-def get_statistics():
+def get_stats():
     """Get dataset statistics"""
-    if df is None:
-        return "Dataset not loaded yet"
-    
-    stats = f"""
-    # 📊 Dataset Statistics
-    
-    - **Total Examples**: {len(df):,}
-    - **Unique Images**: {df['image_id'].nunique():,}
-    - **Unique Answers**: {df['answer'].nunique():,}
-    
-    ## Top 10 Most Common Answers:
-    
-    """
+    if not dataset_state["loaded"]:
+        return "Dataset loading..."
     
+    df = dataset_state["df"]
     top_answers = df['answer'].value_counts().head(10)
-    for i, (answer, count) in enumerate(top_answers.items(), 1):
-        stats += f"{i}. **{answer}** - {count} examples\n"
-    
-    stats += f"\n## Question Length Distribution:\n"
-    stats += f"- Average: {df['question'].str.split().str.len().mean():.1f} words\n"
-    stats += f"- Min: {df['question'].str.split().str.len().min()} words\n"
-    stats += f"- Max: {df['question'].str.split().str.len().max()} words\n"
     
-    stats += f"\n## Answer Length Distribution:\n"
-    stats += f"- Average: {df['answer'].str.split().str.len().mean():.2f} words\n"
-    stats += f"- Single word answers: {(df['answer'].str.split().str.len() == 1).sum():,} ({(df['answer'].str.split().str.len() == 1).sum() / len(df) * 100:.1f}%)\n"
+    stats = f"""# 📊 Dataset Statistics
+
+**Total Examples:** {len(df):,}  
+**Unique Images:** {df['image_id'].nunique():,}  
+**Unique Answers:** {df['answer'].nunique():,}
+
+## Top 10 Answers
+
+"""
+    for i, (ans, count) in enumerate(top_answers.items(), 1):
+        stats += f"{i}. **{ans}** - {count:,} examples\n"
     
     return stats
 
-# Initialize dataset on startup
-print("Starting dataset setup...")
-setup_status = setup_dataset()
-print(setup_status)
+# ==================== Gradio Interface ====================
 
-# Create Gradio interface
-with gr.Blocks(title="Generative VQA v2 Dataset Explorer", theme=gr.themes.Soft()) as demo:
-    gr.Markdown("""
-    # 🎯 Generative VQA-V2-Curated Dataset Explorer
+with gr.Blocks(
+    title="VQA Dataset Explorer",
+    theme=gr.themes.Soft(primary_hue="blue")
+) as demo:
     
-    Explore the **Generative VQA v2 Curated** dataset - a balanced, cleaned version of VQA v2 
-    optimized for generative visual question answering.
-    
-    **Dataset**: [Deva8/Generative-VQA-V2-Curated](https://huggingface.co/datasets/Deva8/Generative-VQA-V2-Curated)
+    gr.Markdown("""
+    # 🎯 Generative VQA-V2 Dataset Explorer
     
-    ---
+    Explore 135K+ curated visual question-answer pairs from the 
+    [Generative-VQA-V2-Curated](https://huggingface.co/datasets/Deva8/Generative-VQA-V2-Curated) dataset.
     """)
     
     with gr.Tabs():
-        # Tab 1: Random Samples
-        with gr.Tab("🎲 Random Samples"):
-            gr.Markdown("### Click the button to see random examples from the dataset")
-            
-            with gr.Row():
-                random_btn = gr.Button("🔄 Get Random Sample", variant="primary", size="lg")
+        
+        # Random Samples Tab
+        with gr.TabItem("🎲 Random"):
+            gr.Markdown("### Browse random examples")
+            btn_random = gr.Button("🔄 Show Random Example", variant="primary", size="lg")
             
             with gr.Row():
-                with gr.Column(scale=1):
-                    random_image = gr.Image(label="Image", type="pil")
-                
-                with gr.Column(scale=1):
-                    random_question = gr.Textbox(label="❓ Question", lines=2)
-                    random_answer = gr.Textbox(label="✅ Answer", lines=1)
-                    random_metadata = gr.Textbox(label="ℹ️ Metadata", lines=1)
+                img_random = gr.Image(label="Image", height=400)
+                with gr.Column():
+                    q_random = gr.Textbox(label="❓ Question", lines=3)
+                    a_random = gr.Textbox(label="✅ Answer", lines=2)
+                    m_random = gr.Textbox(label="ℹ️ Info", lines=1)
             
-            random_btn.click(
-                fn=get_random_sample,
-                outputs=[random_image, random_question, random_answer, random_metadata]
+            btn_random.click(
+                show_random,
+                outputs=[img_random, q_random, a_random, m_random]
             )
         
-        # Tab 2: Search by Question
-        with gr.Tab("🔍 Search Questions"):
-            gr.Markdown("### Search for questions containing specific keywords")
+        # Question Search Tab
+        with gr.TabItem("🔍 Search Questions"):
+            gr.Markdown("### Find questions containing keywords")
             
             with gr.Row():
-                question_query = gr.Textbox(
-                    label="Search Query",
-                    placeholder="e.g., 'color', 'many', 'wearing', 'holding'",
-                    lines=1
+                query_q = gr.Textbox(
+                    label="Search",
+                    placeholder="e.g., color, wearing, many, holding",
+                    scale=4
                 )
-                question_search_btn = gr.Button("🔎 Search", variant="primary")
+                btn_q = gr.Button("🔎 Search", variant="primary", scale=1)
             
             with gr.Row():
-                with gr.Column(scale=1):
-                    question_image = gr.Image(label="Image", type="pil")
-                
-                with gr.Column(scale=1):
-                    question_text = gr.Textbox(label="❓ Question", lines=2)
-                    question_answer = gr.Textbox(label="✅ Answer", lines=1)
-                    question_metadata = gr.Textbox(label="ℹ️ Metadata", lines=1)
+                img_q = gr.Image(label="Image", height=400)
+                with gr.Column():
+                    q_q = gr.Textbox(label="❓ Question", lines=3)
+                    a_q = gr.Textbox(label="✅ Answer", lines=2)
+                    m_q = gr.Textbox(label="ℹ️ Info", lines=1)
             
-            question_search_btn.click(
-                fn=search_by_question,
-                inputs=[question_query],
-                outputs=[question_image, question_text, question_answer, question_metadata]
-            )
+            btn_q.click(search_question, inputs=[query_q], outputs=[img_q, q_q, a_q, m_q])
+            query_q.submit(search_question, inputs=[query_q], outputs=[img_q, q_q, a_q, m_q])
         
-        # Tab 3: Search by Answer
-        with gr.Tab("🎯 Search Answers"):
+        # Answer Search Tab
+        with gr.TabItem("🎯 Search Answers"):
             gr.Markdown("### Find examples with specific answers")
             
             with gr.Row():
-                answer_query = gr.Textbox(
-                    label="Answer to Search",
-                    placeholder="e.g., 'red', 'cat', '2', 'eating'",
-                    lines=1
+                query_a = gr.Textbox(
+                    label="Answer",
+                    placeholder="e.g., red, cat, pizza, 2",
+                    scale=4
                 )
-                answer_search_btn = gr.Button("🔎 Search", variant="primary")
+                btn_a = gr.Button("🔎 Search", variant="primary", scale=1)
             
-            gr.Markdown("**Popular answers**: white, black, blue, red, 2, 3, brown, green, pizza, dog")
+            gr.Markdown("**Popular:** white, black, blue, red, 2, 3, dog, cat, pizza")
             
             with gr.Row():
-                with gr.Column(scale=1):
-                    answer_image = gr.Image(label="Image", type="pil")
-                
-                with gr.Column(scale=1):
-                    answer_question = gr.Textbox(label="❓ Question", lines=2)
-                    answer_text = gr.Textbox(label="✅ Answer", lines=1)
-                    answer_metadata = gr.Textbox(label="ℹ️ Metadata", lines=1)
+                img_a = gr.Image(label="Image", height=400)
+                with gr.Column():
+                    q_a = gr.Textbox(label="❓ Question", lines=3)
+                    a_a = gr.Textbox(label="✅ Answer", lines=2)
+                    m_a = gr.Textbox(label="ℹ️ Info", lines=1)
             
-            answer_search_btn.click(
-                fn=search_by_answer,
-                inputs=[answer_query],
-                outputs=[answer_image, answer_question, answer_text, answer_metadata]
-            )
+            btn_a.click(search_answer, inputs=[query_a], outputs=[img_a, q_a, a_a, m_a])
+            query_a.submit(search_answer, inputs=[query_a], outputs=[img_a, q_a, a_a, m_a])
         
-        # Tab 4: Statistics
-        with gr.Tab("📊 Statistics"):
-            gr.Markdown("### Dataset Statistics and Analysis")
-            
-            stats_btn = gr.Button("📈 Show Statistics", variant="primary")
-            stats_output = gr.Markdown()
+        # Statistics Tab
+        with gr.TabItem("📊 Stats"):
+            gr.Markdown("### Dataset overview and analysis")
+            btn_stats = gr.Button("📈 Load Statistics", variant="primary")
+            stats_md = gr.Markdown()
             
-            stats_btn.click(
-                fn=get_statistics,
-                outputs=[stats_output]
-            )
+            btn_stats.click(get_stats, outputs=[stats_md])
     
     gr.Markdown("""
     ---
     
-    ## About This Dataset
-    
-    **Generative VQA-V2-Curated** is a cleaned and balanced version of VQA v2:
+    ### About
     
-    - ✅ Removed yes/no questions
-    - ✅ Balanced answer distribution (max 600 per answer)
+    This dataset is a curated version of VQA v2 with:
+    - ✅ No yes/no questions
+    - ✅ Balanced answer distribution
     - ✅ Filtered ambiguous questions
-    - ✅ 135,268 high-quality QA pairs
-    - ✅ 1,251 unique answer classes
-    
-    **License**: CC BY 4.0 (COCO + VQA v2)
     
-    **Citation**:
-    ```bibtex
-    @misc{devarajan_genvqa_2026,
-      author = {Devarajan},
-      title = {Generative-VQA-V2-Curated},
-      year = {2026},
-      publisher = {Hugging Face},
-    }
-    ```
+    **Dataset:** [Deva8/Generative-VQA-V2-Curated](https://huggingface.co/datasets/Deva8/Generative-VQA-V2-Curated)  
+    **License:** CC BY 4.0
     """)
 
-# Launch the app
 if __name__ == "__main__":
     demo.launch()