def Main_Window(self):
self.display = pygame.display.set_mode((1280, 720))
manager = ResManager()
pygame.display.set_icon(manager.get_image('icon.png'))
pygame.display.set_caption("Map Editor")
self.display.fill((220, 220, 250))
pygame.display.flip()
i = 0
def Main_Window(self):
self.display = pygame.display.set_mode((1280,720))
manager = ResManager()
pygame.display.set_icon(manager.get_image('icon.png'))
pygame.display.set_caption("War for cookies")
self.map_type = self.core.load_file(self.map_name,self.file)
self.display.fill((220,220,250))
pygame.display.flip()
i = 0
def __init__(self,
width = 1280,
height = 720,
color = (255,255,255),
fps = 40,
scene = None,
manager = ResManager()):
pygame.init()
self.set_display(width, height)
self.fps = fps
self.__manager = manager
self.scene = scene
# self.__display.fill(color)
pygame.display.flip()
def __init__(self):
self.manager = ResManager()
def main():
#************************** PARAMETERS *****************************
fileCollectionSimilarities = "./CollectionSimilarities/dbToDb.gov2.formattedExp"
fileCollectionRanks = "./CollectionRanks/ranks_gov2_crcs_clustfuse_ideal_50_50_0.0_701-850"
fileCollectionRanksUpd = "./CollectionRanks/ranks_gov2_crcs_clustfuse_ideal_50_50_0.0_701-850_upd"
clusteringType = "online"
fileCollectionRanksColumns = 3
clusterSize = 3
collectionRerankCutoff = 10
collectionSearchCutoff = 5
#*******************************************************************
# compute clusters
clustering = Clustering()
clustering._readSimilarities(fileCollectionSimilarities)
if clusteringType == OFFLINE_CLASTERING:
clustering.computeOfflineClusters(fileCollectionSimilarities,
clusterSize)
#clustering.printClusters()
# read collection scores
resManager = ResManager()
resManager.clean()
resManager.readResults(fileCollectionRanks, fileCollectionRanksColumns)
# open output stream
otp = open(fileCollectionRanksUpd, 'w')
# compute scores
for query in resManager.getQueries():
if clusteringType == ONLINE_CLASTERING:
initialCollectionSet = resManager.getDocs(query,
collectionRerankCutoff)
clustering.computeOnlineClusters(initialCollectionSet, clusterSize)
print query, clustering.clusters.keys()
print query, clustering.clusters
scoreFunction = GeometricScoreFunction()
clusterScores = scoreFunction.scoreClusters(clustering, resManager,
query)
# prepare result collection ranking
boost = 100
rerankedCollections = scoreFunction.scoreCollections(
clustering, resManager, clusterScores, query, boost)
print query, rerankedCollections
# check if number of re-ranked collections satisfies collection cutoff
# if it does not, then append more collections to the end of the list
if len(rerankedCollections) < collectionSearchCutoff:
initialCollectionIds = resManager.getDocs(query)
initialCollectionScores = resManager.getScores(query)
pos = 0
while pos < len(initialCollectionIds) and len(
rerankedCollections) < collectionSearchCutoff:
collectionId = initialCollectionIds[pos]
if not rerankedCollections.has_key(collectionId):
rerankedCollections[
collectionId] = initialCollectionScores[pos]
pos += 1
# save results to a file
for collection in sorted(rerankedCollections.iteritems(),
key=itemgetter(1),
reverse=True):
otp.write(" ".join([query, collection[0],
str(collection[1])]) + '\n')
otp.close()
print "ready"
