Merge pull request 'fixed cluster names' (#57) from backend/fix/missing-cluster-names into main

Reviewed-on: #57

backend/src/main.py

@@ -130,11 +130,12 @@ def new_trip(preferences: Preferences,
     logger.debug(f'First stage optimization\t: {round(t_first_stage,3)} seconds')
     logger.debug(f'Second stage optimization\t: {round(t_second_stage,3)} seconds')
     logger.info(f'Total computation time\t: {round(t_first_stage + t_second_stage,3)} seconds')
-
     linked_tour = LinkedLandmarks(refined_tour)
+
     # upon creation of the trip, persistence of both the trip and its landmarks is ensured.
     trip = Trip.from_linked_landmarks(linked_tour, cache_client)
     logger.info(f'Generated a trip of {trip.total_time} minutes with {len(refined_tour)} landmarks in {round(t_generate_landmarks + t_first_stage + t_second_stage,3)} seconds.')
+    logger.debug('Detailed trip :\n\t' + '\n\t'.join(f'{landmark}' for landmark in refined_tour))
 
     background_tasks.add_task(fill_cache)
 

backend/src/optimization/optimizer.py

@@ -594,7 +594,7 @@ class Optimizer:
         status = pl.LpStatus[prob.status]
         solution = [pl.value(var) for var in x]  # The values of the decision variables (will be 0 or 1)
 
-        self.logger.debug("First results are out. Looking out for circles and correcting.")
+        self.logger.debug("First results are out. Looking out for circles and correcting...")
 
         # Raise error if no solution is found. FIXME: for now this throws the internal server error
         if status != 'Optimal' :

backend/src/overpass/overpass.py

@@ -52,7 +52,7 @@ class Overpass :
         # Retrieve cached data and identify missing cache entries
         cached_responses, non_cached_cells = self._retrieve_cached_data(overlapping_cells, osm_types, selector, conditions, out)
 
-        self.logger.info(f'Cache hit for {len(overlapping_cells)-len(non_cached_cells)}/{len(overlapping_cells)} quadrants.')
+        self.logger.debug(f'Cache hit for {len(overlapping_cells)-len(non_cached_cells)}/{len(overlapping_cells)} quadrants.')
 
         # If there is no missing data, return the cached responses after filtering.
         if not non_cached_cells :
@@ -61,6 +61,7 @@ class Overpass :
         # If there is no cached data, fetch all from Overpass.
         elif not cached_responses :
             query_str = Overpass.build_query(bbox, osm_types, selector, conditions, out)
+            self.logger.debug(f'Query string: {query_str}')
             return self.fetch_data_from_api(query_str)
 
         # Hybrid cache: some data from Overpass, some data from cache.
@@ -68,6 +69,7 @@ class Overpass :
             # Resize the bbox for smaller search area and build new query string.
             non_cached_bbox = Overpass._get_non_cached_bbox(non_cached_cells, bbox)
             query_str = Overpass.build_query(non_cached_bbox, osm_types, selector, conditions, out)
+            self.logger.debug(f'Query string: {query_str}')
             non_cached_responses = self.fetch_data_from_api(query_str)
             return Overpass._filter_landmarks(cached_responses, bbox) + non_cached_responses
 

backend/src/parameters/landmark_parameters.yaml

@@ -1,12 +1,11 @@
 max_bbox_side: 4000   #m
 radius_close_to: 50
-church_coeff: 0.55
-nature_coeff: 1.4
+church_coeff: 0.75
+nature_coeff: 1.6
 overall_coeff: 10
 tag_exponent: 1.15
 image_bonus: 1.1
-viewpoint_bonus: 5
+viewpoint_bonus: 10
 wikipedia_bonus: 1.25
-name_bonus: 3
 N_important: 60
 pay_bonus: -1

backend/src/parameters/optimizer_parameters.yaml

@@ -5,5 +5,5 @@ max_landmarks: 10
 max_landmarks_refiner: 20
 overshoot: 0.0016
 time_limit: 1
-gap_rel: 0.05
+gap_rel: 0.025
 max_iter: 40

backend/src/tests/test_main.py

@@ -31,9 +31,9 @@ def test_turckheim(client, request):    # pylint: disable=redefined-outer-name
             "shopping": {"type": "shopping", "score": 0},
             "max_time_minute": duration_minutes,
             "detour_tolerance_minute": 0},
-            # "start": [48.084588, 7.280405]
+            "start": [48.084588, 7.280405]
             # "start": [45.74445023349939, 4.8222687890538865]
-            "start": [45.75156398104873, 4.827154464827647]
+            # "start": [45.75156398104873, 4.827154464827647]
             }
         )
     result = response.json()

backend/src/utils/cluster_manager.py

@@ -1,6 +1,6 @@
 """Find clusters of interest to add more general areas of visit to the tour."""
 import logging
-from typing import Literal
+from typing import Literal, Tuple
 
 import numpy as np
 from sklearn.cluster import DBSCAN
@@ -33,7 +33,7 @@ class Cluster(BaseModel):
     """
     type: Literal['street', 'area']
     importance: int
-    centroid: tuple
+    centroid: Tuple[float, float]
     # start: Optional[list] = None      # for later use if we want to have streets as well
     # end: Optional[list] = None
 
@@ -178,11 +178,12 @@ class ClusterManager:
 
             # Calculate the centroid as the mean of the points
             centroid = np.mean(current_cluster, axis=0)
+            centroid = tuple((round(centroid[0], 7), round(centroid[1], 7)))
 
             if self.cluster_type == 'shopping' :
-                score = len(current_cluster)*2
+                score = len(current_cluster)*3
             else :
-                score = len(current_cluster)*8
+                score = len(current_cluster)*15
             locations.append(Cluster(
                 type='area',
                 centroid=centroid,
@@ -215,7 +216,7 @@ class ClusterManager:
         """
 
         # Define the bounding box for a given radius around the coordinates
-        bbox = create_bbox(cluster.centroid, 1000)
+        bbox = create_bbox(cluster.centroid, 300)
         
         # Query neighborhoods and shopping malls
         selectors = ['"place"~"^(suburb|neighborhood|neighbourhood|quarter|city_block)$"']
@@ -223,10 +224,10 @@ class ClusterManager:
         if self.cluster_type == 'shopping' :
             selectors.append('"shop"="mall"')
             new_name = 'Shopping Area'
-            t = 40
+            t = 30
         else :
             new_name = 'Neighborhood'
-            t = 15
+            t = 20
 
         min_dist = float('inf')
         osm_id = 0
@@ -238,7 +239,7 @@ class ClusterManager:
                 result = self.overpass.send_query(bbox = bbox,
                                                   osm_types = osm_types,
                                                   selector = sel,
-                                                  out = 'ids center'
+                                                  out = 'ids center tags'
                                                   )
             except Exception as e:
                 self.logger.error(f"Error fetching clusters: {e}")
@@ -259,9 +260,9 @@ class ClusterManager:
                 d = get_distance(cluster.centroid, coords)
                 if  d < min_dist :
                     min_dist = d
-                    new_name = name
-                    osm_type = osm_type     # Add type: 'way' or 'relation'
-                    osm_id = id             # Add OSM id
+                    new_name = name         # add name
+                    osm_type = osm_type     # add type: 'way' or 'relation'
+                    osm_id = id             # add OSM id
 
         return Landmark(
             name=new_name,

backend/src/utils/landmarks_manager.py

@@ -39,7 +39,6 @@ class LandmarkManager:
             self.overall_coeff = parameters['overall_coeff']
             self.tag_exponent = parameters['tag_exponent']
             self.image_bonus = parameters['image_bonus']
-            self.name_bonus = parameters['name_bonus']
             self.wikipedia_bonus = parameters['wikipedia_bonus']
             self.viewpoint_bonus = parameters['viewpoint_bonus']
             self.pay_bonus = parameters['pay_bonus']
@@ -147,6 +146,8 @@ class LandmarkManager:
             score *= self.wikipedia_bonus
         if landmark.is_place_of_worship :
             score *= self.church_coeff
+        if landmark.is_viewpoint :
+            score *= self.viewpoint_bonus
         if landmarktype == 'nature' :
             score *= self.nature_coeff
 
@@ -201,7 +202,7 @@ class LandmarkManager:
 
             return_list += self._to_landmarks(result, landmarktype, preference_level)
 
-        self.logger.debug(f"Fetched {len(return_list)} landmarks of type {landmarktype} in {bbox}")
+        # self.logger.debug(f"Fetched {len(return_list)} landmarks of type {landmarktype} in {bbox}")
 
         return return_list
 
@@ -267,7 +268,7 @@ class LandmarkManager:
                     landmark.image_url = value
                 if key == 'website' :
                     landmark.website_url = value
-                if key == 'place_of_worship' :
+                if value == 'place_of_worship' :
                     landmark.is_place_of_worship = True
                 if key == 'wikipedia' :
                     landmark.wiki_url = value