retrieve of size chosen

AirbnbMapVisualiser.py

@@ -44,7 +44,11 @@ class AirbnbMapVisualiser:
         try:
             self.neighborhoods = self.get_all_neighborhoods()
             self.cached_listings = {}
-            self.cached_listings["Southern"] = self.get_neighborhood_listings("Southern")
+            # Initialize a nested dictionary to store listings by neighborhood and limit
+            self.cached_listings = {}
+            # Pre-cache Southern neighborhood with default limit of 10
+            self.cached_listings["Southern"] = {}
+            self.cached_listings["Southern"][10] = self.get_neighborhood_listings("Southern", 10)
             self.cached_embeddings = {}  # Cache for listing embeddings
         except Exception as e:
             print(f"Initialization error: {str(e)}")
@@ -52,7 +56,7 @@ class AirbnbMapVisualiser:
             self.cached_listings = {}
             self.cached_embeddings = {}
 
-    def find_nearest_traffic_spot(self, airbnb_lat, airbnb_lng, max_distance_km=1.0):
+    def find_nearest_traffic_spot(self, airbnb_lat, airbnb_lng, max_distance_km=0.7):
         """Find the nearest traffic spot within max_distance_km kilometers of an Airbnb listing"""
         nearest_spot = None
         min_distance = float('inf')
@@ -98,9 +102,28 @@ class AirbnbMapVisualiser:
         finally:
             self.pool.release(connection)
 
-    def get_neighborhood_listings(self, neighborhood):
-        if neighborhood in self.cached_listings:
-            return self.cached_listings[neighborhood]
+    def get_neighborhood_listings(self, neighborhood, limit=10):
+        """
+        Get listings for a neighborhood with a specified limit.
+
+        Args:
+            neighborhood: The neighborhood to get listings for
+            limit: Maximum number of listings to return (10, 20, 30, 40, or 50)
+
+        Returns:
+            List of listings data
+        """
+        # Ensure limit is one of the allowed values
+        if limit not in [10, 20, 30, 40, 50]:
+            limit = 10  # Default to 10 if invalid limit provided
+
+        # Check if we already have this neighborhood and limit cached
+        if neighborhood in self.cached_listings and limit in self.cached_listings[neighborhood]:
+            return self.cached_listings[neighborhood][limit]
+
+        # Initialize neighborhood in cache if needed
+        if neighborhood not in self.cached_listings:
+            self.cached_listings[neighborhood] = {}
 
         connection = self.pool.acquire()
         try:
@@ -121,11 +144,11 @@ class AirbnbMapVisualiser:
                          m.LATITUDE, m.LONGITUDE, m.ROOM_TYPE, m.PRICE,
                          m.REVIEWS_PER_MONTH, m.MINIMUM_NIGHTS, m.AVAILABILITY_365
                 ORDER BY COUNT(r.LISTING_ID) DESC, m.PRICE ASC
-                FETCH FIRST 100 ROWS ONLY
-            """, neighborhood=neighborhood)
+                FETCH FIRST :limit ROWS ONLY
+            """, neighborhood=neighborhood, limit=limit)
 
             listings = cursor.fetchall()
-            self.cached_listings[neighborhood] = listings
+            self.cached_listings[neighborhood][limit] = listings
             return listings
         except Exception as e:
             print(f"Database error: {str(e)}")
@@ -345,8 +368,10 @@ class AirbnbMapVisualiser:
                     review_similarity = self.compute_similarity(query_embedding, review_embedding)
 
             # Determine which source matched better
-            if title_similarity > 0.2 and review_similarity > 0.2:
+            if title_similarity > 0.2 and review_similarity > 0:
                 return "Strong match in title and reviews"
+            elif title_similarity > 0.2 and review_similarity > 0.2:
+                return "Strong match in title and strong match in reviews"
             elif title_similarity > 0.2:
                 return "Strong match in listing title"
             elif review_similarity > 0.2:
@@ -362,8 +387,30 @@ class AirbnbMapVisualiser:
         return df.sort_values('relevance_score', ascending=False)
 
     def create_map_and_data(self, neighborhood="Sha Tin", show_traffic=True, center_lat=None, center_lng=None,
-                            selected_id=None, search_query=None, current_page=1, items_per_page=3):
-        listings = self.get_neighborhood_listings(neighborhood)
+                            selected_id=None, search_query=None, current_page=1, items_per_page=3, listings_limit=10):
+        """
+        Create a map and dataframe of listings for a neighborhood
+
+        Args:
+            neighborhood: Neighborhood to get listings for
+            show_traffic: Whether to show traffic spots
+            center_lat: Center latitude for the map
+            center_lng: Center longitude for the map
+            selected_id: ID of the selected listing
+            search_query: Search query for filtering listings
+            current_page: Current page of pagination
+            items_per_page: Number of items per page
+            listings_limit: Maximum number of listings to display (10, 20, 30, 40, or 50)
+
+        Returns:
+            Tuple of (map, dataframe)
+        """
+        # Ensure listings_limit is valid
+        if listings_limit not in [10, 20, 30, 40, 50]:
+            listings_limit = 10
+
+        # Get the listings with the specified limit
+        listings = self.get_neighborhood_listings(neighborhood, listings_limit)
 
         if not listings:
             return None, None

app.py

@@ -101,6 +101,8 @@ def main():
         st.session_state.show_traffic_explanation = False
     if 'show_search_explanation' not in st.session_state:
         st.session_state.show_search_explanation = False
+    if 'listings_limit' not in st.session_state:
+        st.session_state.listings_limit = 10  # Default to 10 listings
 
     # Initialize visualizer with loading message for tokenizer
     if 'visualizer' not in st.session_state:
@@ -189,6 +191,22 @@ def main():
             options=visualizer.neighborhoods,
             index=visualizer.neighborhoods.index("Kowloon City") if "Kowloon City" in visualizer.neighborhoods else 0
         )
+
+        # Add dropdown for selecting number of listings to show
+        listings_limit = st.selectbox(
+            "Number of listings to show",
+            options=[10, 20, 30, 40, 50],
+            index=0,  # Default to 10
+            help="Select how many listings to display for this neighborhood"
+        )
+
+        # Update session state if listings_limit has changed
+        if listings_limit != st.session_state.listings_limit:
+            st.session_state.listings_limit = listings_limit
+            # Reset to page 1 when changing the number of listings
+            st.session_state.current_page = 1
+            st.session_state.show_review_dialog = False
+
         show_traffic = st.checkbox("Show Traffic Cameras", value=True)
 
         st.markdown('<hr style="margin: 20px 0; border: none; border-top: 1px solid #e0e0e0;">', unsafe_allow_html=True)
@@ -213,19 +231,23 @@ def main():
             st.session_state.selected_id = None
             st.session_state.current_page = 1
             st.session_state.search_query = ""
+            st.session_state.listings_limit = 10  # Reset to default
             st.session_state.show_review_dialog = False
             st.session_state.show_traffic_explanation = False
             st.session_state.show_search_explanation = False
             st.rerun()
 
-    # Create map and get data - pass current page information
+    # Create map and get data - pass current page information and listings limit
     m, df = visualizer.create_map_and_data(
         neighborhood,
         show_traffic,
         st.session_state.center_lat,
         st.session_state.center_lng,
         st.session_state.selected_id,
-        st.session_state.search_query
+        st.session_state.search_query,
+        st.session_state.current_page,
+        st.session_state.items_per_page,
+        st.session_state.listings_limit
     )
 
     # Handle neighborhood change
@@ -251,6 +273,11 @@ def main():
         st.markdown('</div>', unsafe_allow_html=True)
 
     with col2:
+        # Display info about how many listings are being shown
+        st.markdown(
+            f"<p style='text-align:center; color:#4285f4;'>Showing {st.session_state.listings_limit} listings in {neighborhood}</p>",
+            unsafe_allow_html=True)
+
         total_items = len(df)
         total_pages = math.ceil(total_items / st.session_state.items_per_page)
         st.session_state.current_page = min(max(1, st.session_state.current_page), total_pages)