def Modeling_Home_Prices():
print '>>Modeling Home Prices'
import zillow
if os.path.exists('housedata.txt'):
f=open('housedata.txt','r')
lines=f.readlines()
housedata=[]
for line in lines:
fields=line.split('\t')
l1=[fields[0],fields[1],fields[2],fields[3],fields[4],fields[5],fields[6]]
housedata.append(l1)
f.close();
else:
housedata=zillow.getpricelist( )
f=open('housedata.txt','w')
for l in housedata:
if l is None:
continue
print l
for k in l:
f.write('%s\t' % (k))
f.write('\n')
f.close
reload(treepredict)
housetree=treepredict.buildtree(housedata,scoref=treepredict.variance)
treepredict.drawtree(housetree,'housetree.jpg')
def main():
from treepredict import buildtree, entropy, drawtree
# house_data = getpricelist()
# print house_data
print 'build tree'
t = buildtree(house_data, scoref=entropy)
print 'draw tree'
drawtree(t, 'house_price_tree.jpeg')
def main(rows):
# fruits with their colors and size
tree = treepredict.buildtree(rows)
# print(treepredict.classify([2, 'red'], tree))
# print(treepredict.classify([5, 'red'], tree))
# print(treepredict.classify([1, 'green'], tree))
# 决策树
treepredict.printtree(tree)
treepredict.drawtree(tree, jpeg='treeview.jpg')
if code != '0':
return None
# extract the info about this property
try:
zipcode = doc.getElementsByTagName('zipcode')[0].firstChild.data
use = doc.getElementsByTagName('useCode')[0].firstChild.data
year = doc.getElementsByTagName('yearBuilt')[0].firstChild.data
bath = doc.getElementsByTagName('bathrooms')[0].firstChild.data
bed = doc.getElementsByTagName('bedrooms')[0].firstChild.data
rooms = doc.getElementsByTagName('totalRooms')[0].firstChild.data
price = doc.getElementsByTagName('amount')[0].firstChild.data
except:
return None
return zipcode, use, int(year), float(bath), int(bed), int(rooms), price
def getpricelist():
l1 = []
for line in file('addresslist.txt'):
data = getaddressdata(line.strip(), 'Cambridge+MA')
l1.append(data)
return l1
if __name__ == "__main__":
housedata = getpricelist()
housedata = [data for data in housedata if data != None]
housetree = treepredict.buildtree(housedata, treepredict.variance)
treepredict.drawtree(housetree, 'housetree.jpg')
treepredict.printtree(tree)
trainConfMat, crTrain = treepredict.testTree(train_data, tree)
print 'Training set confusion matrix (Classification rate:', crTrain,'):'
for row in trainConfMat:
print '\t'.join(map(lambda x:str(x), row))
print ''
testConfMat, crTest = treepredict.testTree(test_data, tree)
print 'Test set confusion matrix (Classification rate:', crTest,'):'
for row in testConfMat:
print '\t'.join(map(lambda x:str(x), row))
print ''
# Produce a png of the tree
print '\nPrinting tree image...'
treepredict.drawtree(tree,jpeg="sample_tree.jpg")
# For group homework, modify "buildtree" function so that it stops
# when a threshold value on entropy is no longer satisfied. It should
# accept a minimum gain parameter and stop dividing the branch if
# this condition is not met. Pruning the tree will not be used in
# this cas.
rooms=doc.getElementsByTagName('totalRooms')[0].firstChild.data
price=doc.getElementsByTagName('amount')[0].firstChild.data
except:
return None
print zipcode,use,year,bath,bed,rooms,price
return (zipcode,use,int(year),float(bath),int(bed),int(rooms),price)
def getpricelist():
ll=[]
for line in file('addresslist.txt'):
data=getaddressdata(line.strip(),'Cambrige,MA')
#这里需要加个判断语句，因为list会蛋疼地将None加入，使求方差时出错
if data!=None:
ll.append(data)
return ll
#
#
#
#
#全局测试代码一
if 1:
housedata=getpricelist()
import treepredict
housetree=treepredict.buildtree(housedata,scoref=treepredict.variance)
treepredict.drawtree(housetree)
def tree_view():
from treepredict import buildtree, drawtree
my_data = [map(float, line.split(',')) for line in open('data/agesonly.csv')]
tree = buildtree(my_data)
drawtree(tree, 'treeview.png')
def decision_tree1(self, evt):
import treepredict
reload(treepredict)
full_price = 1130
flights = self.fl_lines[('PEK','PVG')].set_of_flights
data = []
for deptid in flights.keys():
flight = flights[deptid]
ftype = flight['ftype']
deptdate = flight['date']
deptdate = deptdate.split('/')
# 获得周几
weekday = datetime.datetime(int(deptdate[0]),int(deptdate[1]),int(deptdate[2])).weekday()
weekday = int(weekday)
#print weekday
time = flight['time']
# 若起飞时间非常早或者非常晚，取为1，否则为0
if int(time[0:2]) < 9 or int(time[0:2]) > 20:
time = 1
else:
time = 0
# 处理价格
dd = flight['date']
# str-->date
dd = dd.split('/')
deptdate = datetime.date(int(dd[0]), int(dd[1]), int(dd[2]))
price = flight['price']
points = []
for ftdate in price.keys():
ff = ftdate.split('/')
fetchdate = datetime.date(int(ff[0]), int(ff[1]), int(ff[2]))
days = (deptdate-fetchdate).days
points.append((days, price[ftdate]))
points = self.pre(points)
p = []
for i in points.keys():
if i >= 6:
p.append(points[i])
#print p
if len(p) <= 1:
continue
result_price = p[0]
p.pop(0)
avg_price = sum(p)/len(p)
result_price = int(float(result_price)/float(full_price)*10)
avg_price = int(float(avg_price)/float(full_price)*10)
data.append((weekday, time, avg_price, result_price))
## fout = open('task.txt', 'w')
## lines = ['%s %s %s %s\n' %v for v in data]
## fout.writelines(lines)
## fout.close()
flighttree = treepredict.buildtree(data, scoref = treepredict.giniimpurity)
treepredict.drawtree(flighttree,'flighttree_entropy.jpg')
dataset = ads_df_final[['Adjective', 'Adverb', 'Noun', 'Verb', 'Sentiment']]
############# Splitting the Dataset into Testing and Training Sets ##############
final_acc = 0.0
for i in range(no_of_trials):
splitRatio = 0.7
trainingSet, testSet = splitDataset(dataset, splitRatio)
#print(trainingSet)
# print(type(trainingSet))
print('Split {0} rows into train = {1} and test = {2} rows'.format(
len(dataset), len(trainingSet), len(testSet)))
############# Model Building ##############
b = dt.buildtree(trainingSet)
dt.drawtree(b, jpeg='treeview.jpg')
#print("original_testset=",testSet)
############# Preparing Testing DataSet ##############
testlabels = []
for i in range(len(testSet)):
label = testSet[i].pop(-1)
testlabels.append(label)
#print("testSet=",testSet)
#print("testlabels=",testlabels)
############# Classification of Test Records ##############
number = 0
for i in range(len(testSet)):
#print("\ntest_data",testSet[i])
a = dt.classify(testSet[i], b)
gain_threshold=0,
instance_minimum=1)
trainConfMat, crTrain = treepredict.testTree(train_data, tree)
print 'Training set confusion matrix (Classification rate:', crTrain, '):'
for row in trainConfMat:
print '\t'.join(map(lambda x: str(x), row))
print ''
testConfMat, crTest = treepredict.testTree(test_data, tree)
print 'Test set confusion matrix (Classification rate:', crTest, '):'
for row in testConfMat:
print '\t'.join(map(lambda x: str(x), row))
print ''
# Let's see what it looks like...
#print "\nFinal tree...\n"
#treepredict.printtree(tree)
# Produce a png of the tree
print '\nPrinting tree image...'
treepredict.drawtree(tree, jpeg="sample_tree.jpg")
# For group homework, modify "buildtree" function so that it stops
# when a threshold value on entropy is no longer satisfied. It should
# accept a minimum gain parameter and stop dividing the branch if
# this condition is not met. Pruning the tree will not be used in
# this cas.
url += "&method=MeetMe.getProfile&emid=%s&get_keywords=true" % emid
# 得到所有关于此人的详细信息
try:
rating = int(float(rating) + 0.5)
doc2 = xml.dom.minidom.parseString(
urllib.request.urlopen(url).read())
gender = doc2.getElementsByTagName('gender')[0].firstChild.data
age = doc2.getElementsByTagName('age')[0].firstChild.data
loc = doc2.getElementsByTagName('location')[0].firstChild.data[0:2]
# 将州转换为地区
for r, s in stateregions.items():
if loc in s: region = r
if region != None:
result.append((gender, int(age), region, rating))
except:
pass
return result
if __name__ == '__main__': #只有在执行当前模块时才会运行此函数
l1 = getrandomratings(500)
print(len(l1))
pdata = getpeopledata(l1)
print(pdata)
tree = treepredict.buildtree(pdata, scoref=treepredict.variance) #创建决策树
treepredict.prune(tree, 0.5) #剪支
treepredict.drawtree(tree, 'hot.jpg')
def Graphical_Display(): print '>>Graphical Display' treepredict.drawtree(tree,jpeg='treeview.jpg')
try:
zipcode=doc.getElementsByTagName('zipcode')[0].firstChild.data
use=doc.getElementsByTagName('useCode')[0].firstChild.data
year=doc.getElementsByTagName('yearBuilt')[0].firstChild.data
sqft=doc.getElementsByTagName('finishedSqFt')[0].firstChild.data
bath=doc.getElementsByTagName('bathrooms')[0].firstChild.data
bed=doc.getElementsByTagName('bedrooms')[0].firstChild.data
rooms=1 #doc.getElementsByTagName('totalRooms')[0].firstChild.data
price=doc.getElementsByTagName('amount')[0].firstChild.data
except:
return None
return (zipcode,use,int(year),float(bath),int(bed),int(rooms),price)
# 读取文件构造数据集
def getpricelist():
l1=[]
for line in open('addresslist.txt'):
data=getaddressdata(line.strip(),'Cambridge,MA')
print(data)
l1.append(data)
return l1
if __name__=='__main__': #只有在执行当前模块时才会运行此函数
housedata = getpricelist()
print(housedata)
tree = treepredict.buildtree(housedata,scoref=treepredict.variance) #创建决策树
treepredict.drawtree(tree,'house.jpg')
code = doc.getElementsByTagName('code')[0].firstChild.data
if code != '0':
return None
try:
zipcode = doc.getElementsByTagName('zipcode')[0].firstChild.data
use = doc.getElementsByTagName('useCode')[0].firstChild.data
year = doc.getElementsByTagName('yearBuilt')[0].firstChild.data
bath = doc.getElementsByTagName('bathrooms')[0].firstChild.data
bed = doc.getElementsByTagName('bedrooms')[0].firstChild.data
rooms = doc.getElementsByTagName('totalRooms')[0].firstChild.data
price = doc.getElementsByTagName('amount')[0].firstChild.data
except:
return None
return (zipcode, use, int(year), float(bath), int(bed), int(rooms), price)
def getpricelist():
l1 = []
for line in file('addresslist.txt'):
data = getaddressdata(line.strip(), 'Cambridge,MA')
if data is not None:
l1.append(data)
return l1
if __name__ == "__main__":
housedata = getpricelist()
housetree = treepredict.buildtree(housedata, scoref=treepredict.variance)
treepredict.drawtree(housetree, "housetree.jpg")
gender = doc2.getElementsByTagName('gender')[0].firstChild.data
age = doc2.getElementsByTagName('age')[0].firstChild.data
loc = doc2.getElementsByTagName('location')[0].firstChild.data[0:2]
# 将州转换成地区
for r, s in stateregions.items():
if loc in s: region = r
if region != None:
result.append((gender, int(age), region, rating))
except:
pass
return result
l1 = getrandomratings(500)
print len(l1)
pdata = getpeopledata(l1)
print pdata[0]
import treepredict
hottree = treepredict.buildtree(pdata, scoref=treepredict.variance)
treepredict.prune(hottree, 0.5)
treepredict.drawtree(hottree, 'hottree.jpg')
south = treepredict.mdclassify((None, None, 'south'), hottree)
midat = treepredict.mdclassify((None, None, 'Mid Atlantic'), hottree)
print south[10] / sum(south.values())
print midat[10] / sum(midat.values())
price = doc.getElementsByTagName('amount')[0].firstChild.data
except:
return None
print zipcode, use, year, bath, bed, rooms, price
return (zipcode, use, int(year), float(bath), int(bed), int(rooms), price)
def getpricelist():
ll = []
for line in file('addresslist.txt'):
data = getaddressdata(line.strip(), 'Cambrige,MA')
#这里需要加个判断语句，因为list会蛋疼地将None加入，使求方差时出错
if data != None:
ll.append(data)
return ll
#
#
#
#
#全局测试代码一
if 1:
housedata = getpricelist()
import treepredict
housetree = treepredict.buildtree(housedata, scoref=treepredict.variance)
treepredict.drawtree(housetree)
import treepredict import zillow #tree = treepredict.buildtree(treepredict.my_data) #treepredict.printtree(tree) #treepredict.drawtree(tree, jpeg = 'treeview.jpg') #print treepredict.classify(['(direct)', 'USA', 'yes', 5], tree) #treepredict.prune(tree, 1) #treepredict.printtree(tree) #print treepredict.mdclassify(['google', 'None', 'yes', None], tree) housedata = zillow.getpricelist() housetree = treepredict.buildtree(housedata,scoref = treepredict.variance) treepredict.drawtree(housetree, jpeg = 'housetree.jpg')
import treepredict as tp
if __name__ == '__main__':
print("ESERCIZIO SU IRIS DATASET\n")
train_data = []
test_data = []
mydata = tp.aprifile("iris.txt")
train_data, test_data = tp.createdataset2(mydata, 60, [])
print("TRAIN DATA : \n")
print(train_data, "\n")
print("TEST DATA: \n")
print(train_data)
iris_tree = tp.buildtree(train_data)
tp.drawtree(iris_tree, "iris_tree.jpeg")
tp.fperformance(mydata)
import treepredict as tr
agesonly = ad.loadmatch('agesonly.csv', allnum=True)
matchmaker = ad.loadmatch('matchmaker.csv')
# ad.plotagematches(agesonly)
age = []
for line in file('agesonly.csv'):
l = []
for w in line.split(','):
l.append(int(w))
age.append(l)
tree = tr.buildtree(age)
tr.printtree(tree)
tr.drawtree(tree)
print tr.classify(tree, [65, 63])
avgs = ad.lineartrain(agesonly)
print avgs
print ad.dpclassify([30, 25], avgs.values())
print ad.dpclassify([25, 40], avgs.values())
print ad.dpclassify([48, 20], avgs.values())
print tr.classify(tree, [30, 25])
print tr.classify(tree, [25, 40])
print tr.classify(tree, [48, 20])
numericalset = ad.loadnumerical()
# 状态码为0代表操作成功， 否则代表有错误发生
if code != '0': return None
# 提取有关该房产的信息
try:
zipcode = doc.getElementsByTagName('zipcode')[0].firstChild.data
use = doc.getElementsByTagName('useCode')[0].firstChild.data
year = doc.getElementsByTagName('yearBuilt')[0].firstChild.data
bath = doc.getElementsByTagName('bathrooms')[0].firstChild.data
bed = doc.getElementsByTagName('bedrooms')[0].firstChild.data
rooms = doc.getElementsByTagName('totalRooms')[0].firstChild.data
price = doc.getElementsByTagName('amount')[0].firstChild.data
except:
return None
return (zipcode, use, int(year), float(bath), int(bed), int(rooms), price)
'''读取addresslist.txt文件并构造一个数据列表'''
def getpricelist():
l1 = []
for line in file('../data/addresslist.txt'):
data = getaddressdata(line.strip(), 'Cambridge, MA')
l1.append(data)
return l1
import treepredict
housedata = getpricelist()
housetree = treepredict.buildtree(housedata, scoref=treepredict.variance)
treepredict.drawtree(housetree, 'housetree.jpg')
import treepredict as tp
if __name__ == '__main__':
print("ESERCIZIO SU MUSHROOMS DATASET\n")
train_data = []
test_data = []
print("ALL DATASET:\n")
mydata = tp.aprifile("mushrooms_final.txt")
train_data, test_data = tp.createdataset2(mydata, 3250, [])
print("TRAIN DATA : \n")
print(train_data, "\n")
print("TEST DATA: \n")
print(train_data)
mushrooms_tree = tp.buildtree(train_data)
tp.drawtree(mushrooms_tree, "mushrooms_tree.jpeg")
tp.fperformance(mydata)