anydev/anyway
0
0
Fork 0
Code Issues Pull Requests 3 Actions Packages Projects Releases Wiki Activity

Merge pull request 'better naming and MM' (#45) from backend/fix/better-match-making into main

Browse Source 
Reviewed-on: #45
This commit is contained in:
kscheidecker
2025-01-08 09:42:35 +00:00
parent e764393706 a0467e1e19
commit f297094c1a
4 changed files with 120 additions and 105 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
backend/src/parameters/landmark_parameters.yaml
View File

@@ -1,11 +1,11 @@
city_bbox_side: 7500 #m
radius_close_to: 50
church_coeff: 0.9
nature_coeff: 1.25
church_coeff: 0.65
nature_coeff: 1.35
overall_coeff: 10
tag_exponent: 1.15
image_bonus: 10
viewpoint_bonus: 15
viewpoint_bonus: 5
wikipedia_bonus: 4
name_bonus: 3
N_important: 40

11
backend/src/tests/test_main.py
View File

@@ -53,7 +53,7 @@ def test_bellecour(client, request) : # pylint: disable=redefined-outer-name
client:
request:
"""
duration_minutes = 30
duration_minutes = 120
response = client.post(
"/trip/new",
json={
@@ -72,10 +72,16 @@ def test_bellecour(client, request) : # pylint: disable=redefined-outer-name
# Add details to report
log_trip_details(request, landmarks, result['total_time'], duration_minutes)
for elem in landmarks :
print(elem)
print(elem.osm_id)
# checks :
assert response.status_code == 200 # check for successful planning
assert duration_minutes*0.8 < int(result['total_time']) < duration_minutes*1.2
assert 136200148 in osm_ids # check for Cathédrale St. Jean in trip
# assert response.status_code == 2000 # check for successful planning
def test_shopping(client, request) : # pylint: disable=redefined-outer-name
@@ -86,7 +92,7 @@ def test_shopping(client, request) : # pylint: disable=redefined-outer-name
client:
request:
"""
duration_minutes = 600
duration_minutes = 240
response = client.post(
"/trip/new",
json={
@@ -100,7 +106,6 @@ def test_shopping(client, request) : # pylint: disable=redefined-outer-name
)
result = response.json()
landmarks = load_trip_landmarks(client, result['first_landmark_uuid'])
# osm_ids = landmarks_to_osmid(landmarks)
# Add details to report
log_trip_details(request, landmarks, result['total_time'], duration_minutes)

175
backend/src/utils/cluster_processing.py → backend/src/utils/cluster_manager.py
View File

@@ -9,12 +9,12 @@ from OSMPythonTools.cachingStrategy import CachingStrategy, JSON
from ..structs.landmark import Landmark
from ..utils.get_time_separation import get_distance
from ..constants import AMENITY_SELECTORS_PATH, LANDMARK_PARAMETERS_PATH, OPTIMIZER_PARAMETERS_PATH, OSM_CACHE_DIR
from ..constants import OSM_CACHE_DIR
class ShoppingLocation(BaseModel):
class Cluster(BaseModel):
""""
A classe representing an interesting area for shopping.
A class representing an interesting area for shopping or sightseeing.
It can represent either a general area or a specifc route with start and end point.
The importance represents the number of shops found in this cluster.
@@ -33,7 +33,7 @@ class ShoppingLocation(BaseModel):
# end: Optional[list] = None
class ShoppingManager:
class ClusterManager:
logger = logging.getLogger(__name__)
@@ -42,12 +42,21 @@ class ShoppingManager:
all_points: list
cluster_points: list
cluster_labels: list
shopping_locations: list[ShoppingLocation]
cluster_type: Literal['sightseeing', 'shopping']
def __init__(self, bbox: tuple) -> None:
def __init__(self, bbox: tuple, cluster_type: Literal['sightseeing', 'shopping']) -> None:
"""
Upon intialization, generate the point cloud used for cluster detection.
The points represent bag/clothes shops and general boutiques.
If the first step is successful, it applies the DBSCAN clustering algorithm with different
parameters depending on the size of the city (number of points).
It filters out noise points and keeps only the largest clusters.
A successful initialization updates:
- `self.cluster_points`: The points belonging to clusters.
- `self.cluster_labels`: The labels for the points in clusters.
The method also calls `filter_clusters()` to retain only the largest clusters.
Args:
bbox: The bounding box coordinates (around:radius, center_lat, center_lon).
@@ -57,13 +66,23 @@ class ShoppingManager:
self.overpass = Overpass()
CachingStrategy.use(JSON, cacheDir=OSM_CACHE_DIR)
self.cluster_type = cluster_type
if cluster_type == 'shopping' :
elem_type = ['node']
sel = ['"shop"~"^(bag|boutique|clothes)$"']
out = 'skel'
else :
elem_type = ['way']
sel = ['"historic"="building"']
out = 'center'
# Initialize the points for cluster detection
query = overpassQueryBuilder(
bbox = bbox,
elementType = ['node'],
selector = ['"shop"~"^(bag|boutique|clothes)$"'],
elementType = elem_type,
selector = sel,
includeCenter = True,
out = 'skel'
out = out
)
try:
@@ -77,87 +96,50 @@ class ShoppingManager:
else :
points = []
for elem in result.elements() :
points.append(tuple((elem.lat(), elem.lon())))
coords = tuple((elem.lat(), elem.lon()))
if coords[0] is None :
coords = tuple((elem.centerLat(), elem.centerLon()))
points.append(coords)
self.all_points = np.array(points)
self.valid = True
self.valid = True
# Apply DBSCAN to find clusters. Choose different settings for different cities.
if self.cluster_type == 'shopping' and len(self.all_points) > 200 :
dbscan = DBSCAN(eps=0.00118, min_samples=15, algorithm='kd_tree') # for large cities
elif self.cluster_type == 'sightseeing' :
dbscan = DBSCAN(eps=0.0025, min_samples=15, algorithm='kd_tree') # for historic neighborhoods
else :
dbscan = DBSCAN(eps=0.00075, min_samples=10, algorithm='kd_tree') # for small cities
labels = dbscan.fit_predict(self.all_points)
# Separate clustered points and noise points
self.cluster_points = self.all_points[labels != -1]
self.cluster_labels = labels[labels != -1]
# filter the clusters to keep only the largest ones
self.filter_clusters()
def generate_shopping_landmarks(self) -> list[Landmark]:
def generate_clusters(self) -> list[Landmark]:
"""
Generate shopping landmarks based on clustered locations.
This method first generates clusters of locations and then extracts shopping-related
locations from these clusters. It transforms each shopping location into a `Landmark` object.
Returns:
list[Landmark]: A list of `Landmark` objects representing shopping locations.
Returns an empty list if no clusters are found.
"""
self.generate_clusters()
if len(set(self.cluster_labels)) == 0 :
return [] # Return empty list if no clusters were found
# Then generate the shopping locations
self.generate_shopping_locations()
# Transform the locations in landmarks and return the list
shopping_landmarks = []
for location in self.shopping_locations :
shopping_landmarks.append(self.create_landmark(location))
return shopping_landmarks
def generate_clusters(self) :
"""
Generate clusters of points using DBSCAN.
This method applies the DBSCAN clustering algorithm with different
parameters depending on the size of the city (number of points).
It filters out noise points and keeps only the largest clusters.
The method updates:
- `self.cluster_points`: The points belonging to clusters.
- `self.cluster_labels`: The labels for the points in clusters.
The method also calls `filter_clusters()` to retain only the largest clusters.
"""
# Apply DBSCAN to find clusters. Choose different settings for different cities.
if len(self.all_points) > 200 :
dbscan = DBSCAN(eps=0.00118, min_samples=15, algorithm='kd_tree') # for large cities
else :
dbscan = DBSCAN(eps=0.00075, min_samples=10, algorithm='kd_tree') # for small cities
labels = dbscan.fit_predict(self.all_points)
# Separate clustered points and noise points
self.cluster_points = self.all_points[labels != -1]
self.cluster_labels = labels[labels != -1]
# filter the clusters to keep only the largest ones
self.filter_clusters()
def generate_shopping_locations(self) :
"""
Generate shopping locations based on clustered points.
Generate a list of landmarks based on identified clusters.
This method iterates over the different clusters, calculates the centroid
(as the mean of the points within each cluster), and assigns an importance
based on the size of the cluster.
The generated shopping locations are stored in `self.shopping_locations`
as a list of `ShoppingLocation` objects, each with:
The generated shopping locations are stored in `self.clusters`
as a list of `Cluster` objects, each with:
- `type`: Set to 'area'.
- `centroid`: The calculated centroid of the cluster.
- `importance`: The number of points in the cluster.
"""
if not self.valid :
return [] # Return empty list if no clusters were found
locations = []
# loop through the different clusters
@@ -169,16 +151,25 @@ class ShoppingManager:
# Calculate the centroid as the mean of the points
centroid = np.mean(current_cluster, axis=0)
locations.append(ShoppingLocation(
if self.cluster_type == 'shopping' :
score = len(current_cluster)*2
else :
score = len(current_cluster)*8
locations.append(Cluster(
type='area',
centroid=centroid,
importance = len(current_cluster)
importance = score
))
self.shopping_locations = locations
# Transform the locations in landmarks and return the list
cluster_landmarks = []
for cluster in locations :
cluster_landmarks.append(self.create_landmark(cluster))
return cluster_landmarks
def create_landmark(self, shopping_location: ShoppingLocation) -> Landmark:
def create_landmark(self, cluster: Cluster) -> Landmark:
"""
Create a Landmark object based on the given shopping location.
@@ -187,7 +178,7 @@ class ShoppingManager:
result and creates a landmark with the associated details such as name, type, and OSM ID.
Parameters:
shopping_location (ShoppingLocation): A ShoppingLocation object containing
shopping_location (Cluster): A Cluster object containing
the centroid and importance of the area.
Returns:
@@ -196,14 +187,21 @@ class ShoppingManager:
"""
# Define the bounding box for a given radius around the coordinates
lat, lon = shopping_location.centroid
lat, lon = cluster.centroid
bbox = ("around:1000", str(lat), str(lon))
# Query neighborhoods and shopping malls
selectors = ['"place"~"^(suburb|neighborhood|neighbourhood|quarter|city_block)$"', '"shop"="mall"']
selectors = ['"place"~"^(suburb|neighborhood|neighbourhood|quarter|city_block)$"']
if self.cluster_type == 'shopping' :
selectors.append('"shop"="mall"')
new_name = 'Shopping Area'
t = 40
else :
new_name = 'Neighborhood'
t = 15
min_dist = float('inf')
new_name = 'Shopping Area'
new_name_en = None
osm_id = 0
osm_type = 'node'
@@ -231,7 +229,7 @@ class ShoppingManager:
if location[0] is None :
continue
d = get_distance(shopping_location.centroid, location)
d = get_distance(cluster.centroid, location)
if d < min_dist :
min_dist = d
new_name = elem.tag('name')
@@ -246,13 +244,14 @@ class ShoppingManager:
return Landmark(
name=new_name,
type='shopping',
location=shopping_location.centroid, # TODO: use the fact the we can also recognize streets.
attractiveness=shopping_location.importance,
type=self.cluster_type,
location=cluster.centroid, # TODO: use the fact the we can also recognize streets.
attractiveness=cluster.importance,
n_tags=0,
osm_id=osm_id,
osm_type=osm_type,
name_en=new_name_en
name_en=new_name_en,
duration=t
)

33
backend/src/utils/landmarks_manager.py
View File

@@ -5,7 +5,7 @@ from OSMPythonTools.cachingStrategy import CachingStrategy, JSON
from ..structs.preferences import Preferences
from ..structs.landmark import Landmark
from .take_most_important import take_most_important
from .cluster_processing import ShoppingManager
from .cluster_manager import ClusterManager
from ..constants import AMENITY_SELECTORS_PATH, LANDMARK_PARAMETERS_PATH, OPTIMIZER_PARAMETERS_PATH, OSM_CACHE_DIR
@@ -86,6 +86,11 @@ class LandmarkManager:
current_landmarks = self.fetch_landmarks(bbox, self.amenity_selectors['sightseeing'], preferences.sightseeing.type, score_function)
all_landmarks.update(current_landmarks)
# special pipeline for historic neighborhoods
neighborhood_manager = ClusterManager(bbox, 'sightseeing')
historic_clusters = neighborhood_manager.generate_clusters()
all_landmarks.update(historic_clusters)
# list for nature
if preferences.nature.score != 0:
score_function = lambda score: score * 10 * self.nature_coeff * preferences.nature.score / 5
@@ -102,11 +107,9 @@ class LandmarkManager:
all_landmarks.update(current_landmarks)
# special pipeline for shopping malls
shopping_manager = ShoppingManager(bbox)
if shopping_manager.valid :
shopping_clusters = shopping_manager.generate_shopping_landmarks()
for landmark in shopping_clusters : landmark.duration = 45
all_landmarks.update(shopping_clusters)
shopping_manager = ClusterManager(bbox, 'shopping')
shopping_clusters = shopping_manager.generate_clusters()
all_landmarks.update(shopping_clusters)
@@ -277,6 +280,11 @@ class LandmarkManager:
skip = True
break
if "building:" in tag_key:
# do not count the building description as being particularly useful
n_tags -= 1
if "boundary" in tag_key:
# skip "areas" like administrative boundaries and stuff
skip = True
@@ -327,13 +335,16 @@ class LandmarkManager:
continue
score = score_function(score)
if "place_of_worship" in elem.tags().values():
score = score * self.church_coeff
duration = 10
if "place_of_worship" in elem.tags().values() :
if "cathedral" not in elem.tags().values() :
score = score * self.church_coeff
duration = 5
else :
duration = 10
if 'viewpoint' in elem.tags().values() :
elif 'viewpoint' in elem.tags().values() :
# viewpoints must count more
score += self.viewpoint_bonus
score = score * self.viewpoint_bonus
duration = 10
elif "museum" in elem.tags().values() or "aquarium" in elem.tags().values() or "planetarium" in elem.tags().values():