Update app.py

app.py

@@ -5,127 +5,159 @@ import os
 from sentence_transformers import SentenceTransformer
 from sklearn.metrics.pairwise import cosine_similarity
 
-# --- Loading Assets (Exact logic from Colab) ---
+# --- 1. SETTINGS & MODEL ---
+# Using the exact SBERT model from our pipeline
 model = SentenceTransformer('all-MiniLM-L6-v2')
 
-# Database and Embeddings
-main_df = pd.read_csv('bitewise_clean_dataset.csv')
-dish_embeddings = np.load('BiteWise_Dish_Embeddings.npy')
+# File configurations
+MAIN_DB = 'bitewise_clean_dataset.csv'
+DISH_VECTORS = 'BiteWise_Dish_Embeddings.npy'
+USER_DB_PATH = 'BiteWise_User_Database.csv'
+REVIEWS_DB_PATH = 'BiteWise_User_Reviews.csv'
 
-# User Reviews Database
-USER_REVIEWS_FILE = "BiteWise_User_Reviews.csv"
+# Pre-loading static assets
+df_main = pd.read_csv(MAIN_DB)
+dish_embeddings = np.load(DISH_VECTORS)
 
-# --- Styling (Light Brown / Beige) ---
+# --- 2. THE STYLING (Bespoke Light Brown/Beige Theme) ---
 custom_css = """
-.gradio-container { background-color: #F5F5DC; }
-.action-btn {
-    background-color: #8B4513 !important;
-    color: white !important;
-    border-radius: 8px !important;
-}
-.main-header {
-    color: #5D4037;
-    text-align: center;
-    font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+@import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;600&display=swap');
+* { font-family: 'Inter', sans-serif !important; }
+.gradio-container { background-color: #F8F5F2 !important; }
+.main-header { color: #3D2B1F; text-align: center; font-weight: 600; font-size: 2.5em; margin-bottom: 5px; }
+.action-btn { 
+    background: #5D4037 !important; 
+    color: white !important; 
+    border: none !important; 
+    border-radius: 8px !important; 
+    font-weight: 600 !important;
+    padding: 10px 20px !important;
 }
+.action-btn:hover { background: #3D2B1F !important; }
+input, textarea, .dropdown, .slider { border-radius: 8px !important; border: 1px solid #D7CCC8 !important; }
 """
 
-# --- Step 1: User Profile Context ---
-def create_user_profile(user_name, user_preferences):
-    if not user_name:
-        return "Please enter your name."
-    # exact phrasing from Part 4
-    return f"User Profile Created: Welcome {user_name}! We'll tailor your experience based on your preferences: '{user_preferences}'"
+# --- 3. CORE LOGIC (Part 4 Pipeline) ---
+
+def create_user_persona(name, age, cuisine, spice, adventure, diet, fav_dish, drink, frequency):
+    """Step 1: Building the user database record."""
+    if not name:
+        return "⚠️ Error: Full Name is required."
+    
+    user_data = {
+        'full_name': name, 'age_group': age, 'fav_cuisine': cuisine,
+        'spice_level': spice, 'adventure_level': adventure, 'dietary_restrictions': diet,
+        'favorite_dish': fav_dish, 'preferred_drink': drink, 'dining_frequency': frequency
+    }
+    
+    # Save to persistent user storage
+    u_df = pd.DataFrame([user_data])
+    if os.path.exists(USER_DB_PATH):
+        pd.concat([pd.read_csv(USER_DB_PATH), u_df], ignore_index=True).to_csv(USER_DB_PATH, index=False)
+    else:
+        u_df.to_csv(USER_DB_PATH, index=False)
+    
+    # Return exactly as phrased in Colab
+    return f"""### 👤 Profile Created Successfully!
+**Welcome, {name}!**
+
+**Your Culinary Context:**
+* **Preferences:** {cuisine} | **Spice:** {spice}/5 | **Adventure:** {adventure}/5
+* **Details:** {diet} | **Favorite:** {fav_dish} | **Vibe:** {drink}
+
+We have updated our records. You can now use the search engine to find dishes tailored to your taste."""
 
-# --- Step 2: Search Engine (The Engine) ---
-def search_engine(query):
+def run_search_engine(query):
+    """Step 2: Semantic search logic."""
     try:
-        # Step 2 logic: Encode and compare
         query_embedding = model.encode([query])
-        similarities = cosine_similarity(query_embedding, dish_embeddings).flatten()
+        sims = cosine_similarity(query_embedding, dish_embeddings).flatten()
         
-        main_df['match_score'] = similarities
-        top_results = main_df.sort_values(by='match_score', ascending=False).head(3)
+        # Temporary view for results
+        search_df = df_main.copy()
+        search_df['match_score'] = sims
+        top_3 = search_df.sort_values(by='match_score', ascending=False).head(3)
         
-        # Displaying the results in the exact format we built
-        results_text = "### Top 3 Recommended Dishes:\n"
-        for _, row in top_results.iterrows():
-            results_text += f"**Dish Name:** {row['dish_name']}\n"
-            results_text += f"**Restaurant:** {row['restaurant_name']}\n"
-            results_text += f"**Rating:** {row['rating']} ⭐\n"
-            results_text += f"**Taste Review:** {row['taste_review']}\n\n"
-            results_text += "---\n\n"
-        return results_text
+        output = "### 🍴 Top Matches Found:\n\n"
+        for _, row in top_3.iterrows():
+            output += f"**{row['dish_name']}** at **{row['restaurant_name']}**\n"
+            output += f"Rating: {row['rating']} ⭐ | Review: {row['taste_review']}\n\n"
+            output += "---\n"
+        return output
     except Exception as e:
         return f"Error: {e}"
 
-# --- Step 3: Upload Review ---
-def upload_review(dish_name, restaurant_name, rating, taste_review, visual_description, user_name):
-    if not dish_name or not restaurant_name:
-        return "⚠️ Dish and Restaurant names are required.", dish_name, restaurant_name, rating, taste_review, visual_description, user_name
-
-    new_review = {
-        'dish_name': dish_name,
-        'restaurant_name': restaurant_name,
-        'rating': rating,
-        'taste_review': taste_review,
-        'visual_description': visual_description,
-        'user_name': user_name
-    }
+def submit_review(dish, rest, rate, taste, visual, user):
+    """Step 3: Contributing a new review."""
+    if not dish or not rest:
+        return "⚠️ Dish and Restaurant names are required.", dish, rest, rate, taste, visual, user
     
-    new_df = pd.DataFrame([new_review])
+    review_data = {
+        'dish_name': dish, 'restaurant_name': rest, 'rating': rate,
+        'taste_review': taste, 'visual_description': visual, 'user_name': user
+    }
     
-    if os.path.exists(USER_REVIEWS_FILE):
-        pd.concat([pd.read_csv(USER_REVIEWS_FILE), new_df], ignore_index=True).to_csv(USER_REVIEWS_FILE, index=False)
+    r_df = pd.DataFrame([review_data])
+    if os.path.exists(REVIEWS_DB_PATH):
+        pd.concat([pd.read_csv(REVIEWS_DB_PATH), r_df], ignore_index=True).to_csv(REVIEWS_DB_PATH, index=False)
     else:
-        new_df.to_csv(USER_REVIEWS_FILE, index=False)
+        r_df.to_csv(REVIEWS_DB_PATH, index=False)
+        
+    return f"✨ Thank you {user}! Your review for '{dish}' has been synced.", "", "", 4.5, "", "", ""
 
-    # Success message and Resetting fields
-    success_msg = f"Review for '{dish_name}' by {user_name} uploaded successfully!"
-    return success_msg, "", "", 4.5, "", "", ""
+# --- 4. USER INTERFACE (Gradio) ---
 
-# --- UI Layout (Exact sequence from Part 4) ---
-with gr.Blocks(css=custom_css, theme=gr.themes.Soft()) as demo:
-    gr.HTML("<h1 class='main-header'>BiteWise AI Engine</h1>")
+with gr.Blocks(css=custom_css) as demo:
+    gr.HTML("<h1 class='main-header'>BiteWise</h1>")
     
     with gr.Tabs():
-        # PART 4 - STEP 1
-        with gr.Tab("1. Create User Profile"):
-            gr.Markdown("### User Profile Context")
+        # STEP 1: PERSONA
+        with gr.Tab("1. User Profile Setup"):
+            gr.Markdown("### 📋 Create your culinary DNA")
+            with gr.Row():
+                u_name = gr.Textbox(label="Full Name")
+                u_age = gr.Dropdown(choices=["Under 18", "18-25", "26-40", "41-60", "60+"], label="Age Group")
             with gr.Row():
-                u_name = gr.Textbox(label="User Name", placeholder="Enter your name")
-                u_pref = gr.Textbox(label="User Preferences", placeholder="e.g., Spicy, Vegan, Sweet")
-            btn_profile = gr.Button("Create Profile", elem_classes="action-btn")
-            out_profile = gr.Markdown()
-            btn_profile.click(create_user_profile, inputs=[u_name, u_pref], outputs=out_profile)
+                u_cuisine = gr.Textbox(label="Favorite Cuisine Type", placeholder="e.g., Asian, Italian")
+                u_diet = gr.Textbox(label="Dietary Restrictions", placeholder="None, Vegan, etc.")
+            with gr.Row():
+                u_dish = gr.Textbox(label="One dish you love", placeholder="e.g., Truffle Pasta")
+                u_drink = gr.Textbox(label="Preferred Drink", placeholder="e.g., Matcha, Wine")
+            with gr.Row():
+                u_freq = gr.Dropdown(choices=["Daily", "Weekly", "Rarely"], label="Dining Frequency")
+                u_spice = gr.Slider(1, 5, step=1, label="Spice Tolerance")
+                u_adv = gr.Slider(1, 5, step=1, label="Adventurousness")
+            
+            btn_p = gr.Button("Create Profile", elem_classes="action-btn")
+            out_p = gr.Markdown()
+            btn_p.click(create_user_persona, 
+                       inputs=[u_name, u_age, u_cuisine, u_spice, u_adv, u_diet, u_dish, u_drink, u_freq], 
+                       outputs=out_p)
 
-        # PART 4 - STEP 2
-        with gr.Tab("2. Search Engine"):
-            gr.Markdown("### Dish Search Engine")
-            s_query = gr.Textbox(label="Enter your craving", placeholder="e.g., I want a juicy burger")
-            btn_search = gr.Button("Find My Dish", variant="primary", elem_classes="action-btn")
-            out_search = gr.Markdown()
-            btn_search.click(search_engine, inputs=s_query, outputs=out_search)
+        # STEP 2: SEARCH
+        with gr.Tab("2. Dish Search"):
+            gr.Markdown("### 🔍 Find your next flavor")
+            s_q = gr.Textbox(label="Describe your craving", placeholder="e.g., Something spicy and crunchy")
+            btn_s = gr.Button("Search Engine", elem_classes="action-btn")
+            out_s = gr.Markdown()
+            btn_s.click(run_search_engine, inputs=s_q, outputs=out_s)
 
-        # PART 4 - STEP 3
+        # STEP 3: CONTRIBUTION
         with gr.Tab("3. Upload Review"):
-            gr.Markdown("### Contribute Your Review")
+            gr.Markdown("### ✍️ Share your experience")
             with gr.Row():
-                in_user = gr.Textbox(label="Reviewer Name")
-                in_dish = gr.Textbox(label="Dish Name")
-                in_rest = gr.Textbox(label="Restaurant Name")
+                in_u = gr.Textbox(label="Reviewer Name")
+                in_d = gr.Textbox(label="Dish Name")
+                in_r = gr.Textbox(label="Restaurant")
             with gr.Row():
                 in_rate = gr.Slider(1, 5, value=4.5, step=0.1, label="Rating")
-                in_taste = gr.Textbox(label="Taste Review")
-                in_visual = gr.Textbox(label="Visual Description")
-            
-            btn_upload = gr.Button("Upload Review", elem_classes="action-btn")
-            out_status = gr.Markdown()
+                in_t = gr.Textbox(label="Taste Notes")
+                in_v = gr.Textbox(label="Visual Description")
             
-            btn_upload.click(
-                upload_review,
-                inputs=[in_dish, in_rest, in_rate, in_taste, in_visual, in_user],
-                outputs=[out_status, in_dish, in_rest, in_rate, in_taste, in_visual, in_user]
-            )
+            btn_r = gr.Button("Upload Review", elem_classes="action-btn")
+            out_r = gr.Markdown()
+            btn_r.click(submit_review, 
+                       inputs=[in_d, in_r, in_rate, in_t, in_v, in_u], 
+                       outputs=[out_r, in_d, in_r, in_rate, in_t, in_v, in_u])
 
 demo.launch()