def buildLocationVector(raws_user_staypointcenter, mindist=80):
'''
建立分类算法时需要用到的用户总的位置向量列表，提取用户停留位置中心，删除重复的停留位置中心-v1.1--------------算法还存在漏洞，在最后写毕设的时候再来排查
:param raws_user_staypointcenter: dict{'user':list[Location]}，原始用户的位置数据
:param mindist: 位置归类，最小距离为20m
:return: 所有用户的停留位置中心向量列表final_vector类型为list[Location]
'''
final_vector = []
for key_fisrt in raws_user_staypointcenter.keys():
content = raws_user_staypointcenter.get(key_fisrt) # 获得某个用户的所有中心位置点
skip = False
for item in content: # 开始循环
for skip_item in final_vector: #查找某个位置是否已经添加至最终位置列表中
if skip_item.lat == item.lat and skip_item.longti == item.longti: #由于每次添加位置信息都是新生产一个Location类型实例，所以必须要使用数值相等判断是否相等而不能使用地址相等判断
# print('------skipping [%f,%f] ==> [%f,%f]------' % (skip_item.lat, skip_item.longti, item.lat, item.longti))
skip = True
break
if skip:
skip = False
continue
nerbor = []
for key_second in raws_user_staypointcenter.keys(
): #从所有用户数据中依次寻找符合条件的同类型中心点
temp = raws_user_staypointcenter.get(key_second)
for line in temp:
if item == line:
continue
dist = calDistance(item, line)
if dist <= mindist: #符合条件的中心点，依次添加至临时向量中，由于这里是直接添加的中心点的地址，因此在后续修改数值时，可以直接影响原始结果
nerbor.append(line)
if len(nerbor) != 0:
nerbor.append(item)
location = Location()
location.lat, location.longti = calAvgPoint(nerbor)
isnew = False #用来判断是否产生新的中心点
for ci in nerbor:
for cm in final_vector: #判断这个新的中心点是否是从已经计算过的中心点中再次产生的
if cm.lat == ci.lat and cm.longti == ci.longti: #表明新产生的中心点是从已经计算过的中心点中产生的，因此只需修改已计算过的中心点数值，然后将其他中心点的值改为最新值即可
isnew = True
cm.lat, cm.longti = location.lat, location.longti
break
# print('******changing [%f,%f] ==> [%f,%f]******' % (ci.lat, ci.longti, location.lat, location.longti))
ci.lat, ci.longti = location.lat, location.longti
if not isnew: #表明这是一个完成新产生的中心点
final_vector.append(location)
else:
final_vector.append(item)
return final_vector
def generateRandomDistributedClan(_count, x, y, dist):
_geesearray = []
for count in range(_count):
_location = Location(x + random.uniform(-dist, dist) * 2,
y + random.uniform(-dist, dist))
_geesearray.append(entity.generateRandomClanGoose(_location))
return _geesearray
def __init__(self, entity_id=-1, server=None):
super(MissileEntity, self).__init__(entity_id)
self.player_id = -1
self.server = server
self.toward_vector = [0, 0, 0]
self.init_location = Location()
def buildUserTFIDFVector(user_location_vector, user_final_location_vector):
'''
建立某个用户的TFIDF向量列表-v1.1
:param user_location_vector: 某用户的停留位置中心向量列表，类型为list[location]
:param user_final_location_vector: 所有用户总的停留位置向量列表，类型为list[location]
:return: 返回值为list[location]
'''
result = []
for raws_locations in user_final_location_vector:
location = Location()
location.lat, location.longti = raws_locations.lat, raws_locations.longti
location.tfidf = 0
for item in user_location_vector:
if location.lat == item.lat and location.longti == item.longti:
location.tfidf = item.tfidf
break
result.append(location)
return result
def loadClassifyData(dir=r'saveresults', filename=r'clusterresult.pkl'):
'''
从二进制文件中加载停留位置数据信息，为分类算法提供输入数据-v1.1
:param dir: 保存结果目录名
:param filename: 保存数据文件名
:return: 返回所有用户的中心位置，类型为 dict{'user':list[location]}
'''
data = {}
for (thisDir, dirsHere, filesHere) in os.walk(dir):
for files in filesHere:
user = files.split('&')[0]
content = loadStayPointPklResults(dirs=dir, fname=files)
locations = []
for line in content:
location = Location()
location.lat, location.longti = line.centerlat, line.centerlon
location.tfidf = 0
locations.append(location)
data[user] = locations
return data
def loadRawsData(dirs=r'rawsdata', fname=r'5477.txt'):
'''
从文本文件中加载原始地理位置数据信息，DBScan挖掘用户停留区域的数据源，包括纬度(浮点类型)、经度(浮点类型)、时间点(日期类型)-v1.1
:param dirs: 目录名
:param fname:是读取存储处理好的数据文件名
:return: list[Location] 用户所有地理位置数量number
'''
fname = combileFileName(dirs, fname)
result = []
number = 0
for line in open(fname):
content = line.rstrip().split(',')
location = Location()
location.lat = float(content[0])
location.longti = float(content[1])
# location.timer = float(content[2])
location.timer = datetime.strptime(content[2], str_formate)
result.append(location)
number += 1
return result, number
def generateRandomClan(_count, x, y):
_geesearray = []
_location = Location(x, y)
for count in range(_count):
_geesearray.append(entity.generateRandomClanGoose(_location))
return _geesearray