from jobs import utils
import pdb
test_file = 'd:/jobs/dctree/degree/test.csv'
labels = utils.get_labels(test_file, 0)
pdb.set_trace()
plabel = utils.read_rst('result')
i = 0
rst = 0
for label in labels:
if plabel[i] == label:
rst += 1
i += 1
print rst
from jobs import utils
import pdb
#pdb.set_trace()
test_file = 'd:/jobs/dctree/random/test.csv'
labels = utils.get_labels(test_file, 7)
plabel = []
# for j in range(10):
# plabel.append(utils.read_rst('result03' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result04' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result08' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result09' +str(j)+ '.txt'))
rst = [0 for j in xrange(1)]
plabel.append(utils.read_rst('finalrut'))
i = 0
for label in labels:
for j in range(1):
if plabel[j][i] == label:
rst[j] += 1
i += 1
print rst
try:
conn = MySQLdb.connect(host='localhost', user='******', passwd='123456', db='jobs', use_unicode=True, charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
file = open('d:/jobs/dctree/position/train.csv', 'w+')
sql = 'select userid, age, bstart_year, gender, shortmar from jobs_uinfo'
cur.execute(sql)
userdlst = cur.fetchall()
sqlze = 'select wk.salary, wk.industry, wk.position_name from work_size as wk'
position_dct = get_position_meta()
major_dct = read_rst('sharemajor')
cur.execute(sqlze)
sizelst = cur.fetchall()
i = 0
for userd in userdlst:
sizes = sizelst[i:i+3]
i += 3
if not position_dct.has_key(sizes[1][2]):
continue
# print userd
userid = []
userd = list(userd)
if int(userd[1]) <= 20:
userd[1] = '18'
elif int(userd[1]) >= 60:
userd[1] = '60'
passwd='123456',
db='jobs',
use_unicode=True,
charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select name, degreer0, degreer1, degreer2 from major ;'
cur.execute(sql)
majorlst = cur.fetchall()
majordct = {}
pdb.set_trace()
treelst = utils.read_rst('result.txt')
resultlst = []
for major in majorlst:
majordct[major[0]] = [major[1], major[2], major[3]]
sq = 'select userid, major from jobs_uinfotest'
cur.execute(sq)
usertst = cur.fetchall()
i = 0
for user in usertst:
if majordct.has_key(user[1]):
majorat = majordct[user[1]]
mnu = max(majorat)
index = majorat.index(mnu)
resultlst.append(index)
else:
resultlst.append(0)
from jobs import utils
import pdb
pdb.set_trace()
test_file = 'd:/jobs/dctree/size/test.csv'
labels = utils.get_labels(test_file, 5)
plabel = []
# for j in range(10):
# plabel.append(utils.read_rst('result03' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result04' +str(j)+ '.txt'))
for j in range(10):
plabel.append(utils.read_rst('result05' +str(j)+ '.txt'))
for j in range(6):
plabel.append(utils.read_rst('result06' +str(j)+ '.txt'))
rst = [0 for j in xrange(16)]
i = 0
for label in labels:
for j in range(16):
if plabel[j][i] == label:
rst[j] += 1
i += 1
print rst
from jobs import utils
import pdb
# pdb.set_trace()
test_file = 'd:/jobs/dctree/size/test.csv'
labels = utils.get_labels(test_file, 5)
plabel = utils.read_rst('sizeresult.txt')
i = 0
rst = 0
for label in labels:
if plabel[i] == label:
rst += 1
i += 1
print rst
from jobs import utils
import pdb
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
# pdb.set_trace()
plabel1 = utils.read_rst('sizeresult.txt')
plabel2 = utils.read_rst('sizeresult1.txt')
i = 0
rst = 0
for label in plabel1:
if plabel2[i] == label:
rst += 1
i += 1
print rst
postdct = get_position.get_pos()
try:
conn = MySQLdb.connect(host="localhost", user="******", passwd="123456", db="jobs", use_unicode=True, charset="utf8")
cur = conn.cursor()
cur.execute("set character_set_client=utf8")
cur.execute("set character_set_connection=utf8")
cur.execute("set character_set_database=utf8")
cur.execute("set character_set_results=utf8")
cur.execute("set character_set_server=utf8")
# sql = 'select userid from jobs_uinfotest'
# pdb.set_trace()
position_dct = {}
industryr = utils.read_rst("industryr.txt")
with codecs.open("position_meta.txt") as file:
lines = file.readlines()
for linet in lines:
line = linet[:-2]
uline = unicode(line)
position_dct[uline] = "1"
sql = "select position_name, industry from work_sizetest"
cur.execute(sql)
positionlst = cur.fetchall()
positions = []
industrys = []
result = []
flag = False
i = 0
for position in positionlst:
conn = MySQLdb.connect(host='localhost',
user='******',
passwd='123456',
db='jobs',
use_unicode=True,
charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select industry, position_name from work_sizetest'
cur.execute(sql)
workprobdct = utils.read_rst('workprobdct')
worklst = cur.fetchall()
i = 0
result = []
for j in xrange(20000):
tworks = worklst[i:i + 2]
i += 2
position_prob = {}
for key in position_dct.keys():
# pdb.set_trace()
position_prob[key] = get_position_prob(key, workprobdct, tworks)
# pdb.set_trace()
sortedprob = sorted(position_prob.iteritems(),
key=lambda jj: jj[1],
reverse=True)
#coding:utf8
import sys
from jobs import utils
reload(sys)
import pdb
sys.setdefaultencoding('utf8')
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
# pdb.set_trace()
plabel1 = utils.read_rst('position.txt')
plabel2 = utils.read_rst('positionlet')
i = 0
rst = 0
for label in plabel1:
if plabel2[i] == label:
rst += 1
i += 1
print rst
conn = MySQLdb.connect(host='localhost',
user='******',
passwd='123456',
db='jobs',
use_unicode=True,
charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select industry, salary from work_sizetest'
cur.execute(sql)
salaryprobdct = utils.read_rst('salaryprobdct')
worklst = cur.fetchall()
i = 0
result = []
for j in xrange(20000):
salarys = worklst[i:i + 2]
i += 2
salary_prob = {}
for key in range(7):
# pdb.set_trace()
salary_prob[key] = get_salary_prob(key, salaryprobdct, salarys)
# pdb.set_trace()
sortedprob = sorted(salary_prob.iteritems(),
key=lambda jj: jj[1],
reverse=True)
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
#sql = 'select userid from jobs_uinfotest'
position_dct = {}
with codecs.open('position_meta.txt') as file:
lines = file.readlines()
for linet in lines:
line = linet[:-2]
uline = unicode(line)
lintlst = uline.split(',')
position_dct[lintlst[0]] = lintlst[1]
file = open('d:/jobs/xgboost/data.csv', 'w+')
pdb.set_trace()
keyshare = utils.read_rst('keyshare')
sql = 'select position_name, industry from work_size'
cur.execute(sql)
datalst = []
positionlst = cur.fetchall()
i = 0
for j in xrange(70000):
worklst = positionlst[i:i+3]
i += 1
if not position_dct.has_key(worklst[1][0]):
continue
else:
rst = []
if keyshare.has_key(worklst[0][0]):
rst.append(keyshare[worklst[0][0]])
from jobs import utils
import pdb
test_file = 'd:/jobs/dctree/degree/test.csv'
labels = utils.get_labels(test_file, 0)
pdb.set_trace()
plabel = utils.read_rst('result')
i = 0
rst = 0
for label in labels:
if plabel[i] == label:
rst += 1
i += 1
print rst
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select jb.userid,jb.degree,jb.age,jb.start_age,jb.bstart_year,jb.gender,jb.start_salary,wk.size \
from jobs_uinfotest as jb left join workexperiencetest as wk on \
jb.userid = wk.userid and wk.num = 1'
cur.execute(sql)
file = open('d:/jobs/dctree/salary/salary.csv', 'w+')
useridlst = cur.fetchall()
# rsultlabel = utils.read_rst('result.txt')
# pdb.set_trace()
#wsresult = utils.read_rst('wsresult.txt')
# sizeresult = utils.read_rst('sizeresult.txt')
salaryresult = utils.read_rst('salary')
# posresult = utils.read_rst('position13')
#degreelst = utils.read_rst('degree.txt')
# degreelst = utils.read_rst('result1.txt')
file.write('id,degree,size,salary,position_name\n')
i = 0
# pdb.set_trace()
j = 0
for userid in useridlst:
result = []
result.append(userid[0])
print i
# result.append(degreelst[i])
# result.append(sizeresult[i])
result.append(8)
result.append(8)
#coding:utf8
import sys
from jobs import utils
reload(sys)
import pdb
sys.setdefaultencoding('utf8')
test_file = 'd:/jobs/dctree/random/test.csv'
labels = utils.get_labels(test_file, 9)
# pdb.set_trace()
plabel1 = utils.read_rst('result.txt')
plabel2 = utils.read_rst('finalrut')
i = 0
rst = 0
for label in labels:
if plabel2[i] == label:
rst += 1
i += 1
print rst
db='jobs',
use_unicode=True,
charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
file = open('d:/jobs/dctree/position/train.csv', 'w+')
sql = 'select userid, age, bstart_year, gender, shortmar from jobs_uinfo'
cur.execute(sql)
userdlst = cur.fetchall()
sqlze = 'select wk.salary, wk.industry, wk.position_name from work_size as wk'
position_dct = get_position_meta()
major_dct = read_rst('sharemajor')
cur.execute(sqlze)
sizelst = cur.fetchall()
i = 0
for userd in userdlst:
sizes = sizelst[i:i + 3]
i += 3
if not position_dct.has_key(sizes[1][2]):
continue
# print userd
userid = []
userd = list(userd)
if int(userd[1]) <= 20:
userd[1] = '18'
elif int(userd[1]) >= 60:
userd[1] = '60'
from jobs import utils
import pdb
#pdb.set_trace()
test_file = 'd:/jobs/dctree/random/test.csv'
labels = utils.get_labels(test_file, 7)
plabel = []
# for j in range(10):
# plabel.append(utils.read_rst('result03' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result04' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result08' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result09' +str(j)+ '.txt'))
rst = [0 for j in xrange(1)]
plabel.append(utils.read_rst('finalrut'))
i = 0
for label in labels:
for j in range(1):
if plabel[j][i] == label:
rst[j] += 1
i += 1
print rst
from jobs import utils
import pdb
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
# pdb.set_trace()
plabel1 = utils.read_rst('sizeresult.txt')
plabel2 = utils.read_rst('sizeresult1.txt')
i = 0
rst = 0
for label in plabel1:
if plabel2[i] == label:
rst += 1
i += 1
print rst
#coding:utf8
from jobs import utils
import jieba
import jieba.posseg as pseg
import os
import pdb
import sys
from sklearn import feature_extraction
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import CountVectorizer
reload(sys)
sys.setdefaultencoding('utf8')
letter_dct = utils.read_rst('letterdct')
tfidf_dct = {}
# pdb.set_trace()
corpus = []
keys = []
for item in letter_dct.items():
keys.append(item[0])
corpus.append(' '.join(item[1][0]))
# corpus = [[' '.join(item[1][0])] for item in letter_dct.items()]
vectorizer = CountVectorizer() #该类会将文本中的词语转换为词频矩阵，矩阵元素a[i][j] 表示j词在i类文本下的词频
transformer = TfidfTransformer() #该类会统计每个词语的tf-idf权值
tfidf = transformer.fit_transform(vectorizer.fit_transform(
corpus)) #第一个fit_transform是计算tf-idf，第二个fit_transform是将文本转为词频矩阵
word = vectorizer.get_feature_names() #获取词袋模型中的所有词语
weight = tfidf.toarray() #将tf-idf矩阵抽取出来，元素a[i][j]表示j词在i类文本中的tf-idf权重
total = total*(pos1prob + pos2prob + industry1prob + industry2prob)
return total
try:
conn = MySQLdb.connect(host='localhost', user='******', passwd='123456', db='jobs', use_unicode=True, charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select industry, position_name from work_sizetest'
cur.execute(sql)
workprobdct = utils.read_rst('workprobdct')
worklst = cur.fetchall()
i = 0
result = []
for j in xrange(20000):
tworks = worklst[i:i+2]
i += 2
position_prob = {}
for key in position_dct.keys():
# pdb.set_trace()
position_prob[key] = get_position_prob(key, workprobdct, tworks)
# pdb.set_trace()
sortedprob = sorted(position_prob.iteritems(), key=lambda jj:jj[1], reverse=True)
# for prob in sortedprob:
# print prob[0] + str(prob[1])
try:
conn = MySQLdb.connect(host="localhost", user="******", passwd="123456", db="jobs", use_unicode=True, charset="utf8")
cur = conn.cursor()
cur.execute("set character_set_client=utf8")
cur.execute("set character_set_connection=utf8")
cur.execute("set character_set_database=utf8")
cur.execute("set character_set_results=utf8")
cur.execute("set character_set_server=utf8")
file = open("d:/jobs/dctree/position/test.csv", "w+")
sql = "select userid, age, bstart_year, gender, shortmar from jobs_uinfo limit 50000, 20000"
cur.execute(sql)
userdlst = cur.fetchall()
sqlze = "select wk.salary, wk.industry, wk.position_name from work_size as wk limit 150000, 60000"
position_dct = get_position_meta()
major_dct = read_rst("sharemajor")
cur.execute(sqlze)
sizelst = cur.fetchall()
i = 0
for userd in userdlst:
sizes = sizelst[i : i + 3]
i += 3
if not position_dct.has_key(sizes[1][2]):
continue
print userd
userid = []
userd = list(userd)
if int(userd[1]) <= 20:
userd[1] = "18"
elif int(userd[1]) >= 60:
userd[1] = "60"
total = total*(pos1prob + pos2prob + industry1prob + industry2prob)
return total
try:
conn = MySQLdb.connect(host='localhost', user='******', passwd='123456', db='jobs', use_unicode=True, charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select industry, position_name from work_sizetest'
cur.execute(sql)
workprobdct = utils.read_rst('workletterprobdct')
worklst = cur.fetchall()
i = 0
result = []
for j in xrange(20000):
tworks = worklst[i:i+2]
i += 2
position_prob = {}
pdb.set_trace()
for key in position_dct.keys():
# pdb.set_trace()
position_prob[key] = get_position_prob(key, workprobdct, tworks)
# pdb.set_trace()
sortedprob = sorted(position_prob.iteritems(), key=lambda jj:jj[1], reverse=True)
# for prob in sortedprob:
return total
try:
conn = MySQLdb.connect(host="localhost", user="******", passwd="123456", db="jobs", use_unicode=True, charset="utf8")
cur = conn.cursor()
cur.execute("set character_set_client=utf8")
cur.execute("set character_set_connection=utf8")
cur.execute("set character_set_database=utf8")
cur.execute("set character_set_results=utf8")
cur.execute("set character_set_server=utf8")
sql = "select position_name from work_sizetest"
cur.execute(sql)
wordprobdct = utils.read_rst("position_word")
wordlst = cur.fetchall()
i = 0
result = []
# pdb.set_trace()
for j in xrange(20000):
tworks = wordlst[i : i + 2]
i += 2
position_prob = {}
for key in position_dct.keys():
position_prob[key] = get_position_prob(key, wordprobdct, tworks)
sortedprob = sorted(position_prob.iteritems(), key=lambda jj: jj[1], reverse=True)
# for prob in sortedprob:
# print prob[0] + str(prob[1])
result.append(sortedprob[0][0])
from jobs import utils
import pdb
pdb.set_trace()
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
plabel = []
# for j in range(10):
# plabel.append(utils.read_rst('result03' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result04' +str(j)+ '.txt'))
for j in range(10):
plabel.append(utils.read_rst('result1' + str(j) + '.txt'))
for j in range(10):
plabel.append(utils.read_rst('result2' + str(j) + '.txt'))
rst = [0 for j in xrange(20)]
i = 0
for label in labels:
for j in range(20):
if plabel[j][i] == label:
rst[j] += 1
i += 1
print rst
user='******',
passwd='123456',
db='jobs',
use_unicode=True,
charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
#sql = 'select userid from jobs_uinfotest'
# pdb.set_trace()
position_dct = {}
industryr = utils.read_rst('industryr.txt')
with codecs.open('position_meta.txt') as file:
lines = file.readlines()
for linet in lines:
line = linet[:-2]
uline = unicode(line)
position_dct[uline] = '1'
sql = 'select position_name, industry from work_sizetest'
cur.execute(sql)
positionlst = cur.fetchall()
positions = []
industrys = []
result = []
flag = False
i = 0
for position in positionlst:
from jobs import utils
import pdb
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
pdb.set_trace()
plabel = utils.read_rst('salary')
i = 0
rst = 0
for label in labels:
if plabel[i] == label:
rst += 1
i += 1
print rst
#coding:utf8
import sys
from jobs import utils
reload(sys)
import pdb
sys.setdefaultencoding('utf8')
test_file = 'd:/jobs/dctree/random/test.csv'
labels = utils.get_labels(test_file, 9)
# pdb.set_trace()
plabel1 = utils.read_rst('result.txt')
plabel2 = utils.read_rst('finalrut')
i = 0
rst = 0
for label in labels:
if plabel2[i] == label:
rst += 1
i += 1
print rst
conn = MySQLdb.connect(host='localhost',
user='******',
passwd='123456',
db='jobs',
use_unicode=True,
charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select position_name from work_sizetest'
cur.execute(sql)
wordprobdct = utils.read_rst('position_word')
wordlst = cur.fetchall()
i = 0
result = []
# pdb.set_trace()
for j in xrange(20000):
tworks = wordlst[i:i + 2]
i += 2
position_prob = {}
for key in position_dct.keys():
position_prob[key] = get_position_prob(key, wordprobdct, tworks)
sortedprob = sorted(position_prob.iteritems(),
key=lambda jj: jj[1],
reverse=True)
# for prob in sortedprob:
# print prob[0] + str(prob[1])
from jobs import utils
import pdb
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
pdb.set_trace()
plabel1 = utils.read_rst('salaryresult.txt')
plabel2 = utils.read_rst('salaryresult1.txt')
i = 0
rst = 0
for label in plabel1:
if plabel2[i] == label:
rst += 1
i += 1
print rst
postdct = get_position.get_pos()
try:
conn = MySQLdb.connect(host='localhost', user='******', passwd='123456', db='jobs', use_unicode=True, charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
#sql = 'select userid from jobs_uinfotest'
import pdb
pdb.set_trace()
position_dct = {}
industryr = utils.read_rst('industryr.txt');
with codecs.open('position_meta.txt') as file:
lines = file.readlines()
for linet in lines:
line = linet[:-2]
uline = unicode(line)
position_dct[uline] = '1'
sql = 'select position_name, industry from work_sizetest'
cur.execute(sql)
positionlst = cur.fetchall()
positions = []
industrys = []
result = []
flag = False
i = 0
for position in positionlst:
#coding:utf8
from jobs import utils
import jieba
import jieba.posseg as pseg
import os
import pdb
import sys
from sklearn import feature_extraction
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import CountVectorizer
reload(sys)
sys.setdefaultencoding('utf8')
letter_dct = utils.read_rst('letterdct')
tfidf_dct = {}
# pdb.set_trace()
corpus = []
keys = []
for item in letter_dct.items():
keys.append(item[0])
corpus.append(' '.join(item[1][0]))
# corpus = [[' '.join(item[1][0])] for item in letter_dct.items()]
vectorizer=CountVectorizer()#该类会将文本中的词语转换为词频矩阵，矩阵元素a[i][j] 表示j词在i类文本下的词频
transformer=TfidfTransformer()#该类会统计每个词语的tf-idf权值
tfidf=transformer.fit_transform(vectorizer.fit_transform(corpus))#第一个fit_transform是计算tf-idf，第二个fit_transform是将文本转为词频矩阵
word=vectorizer.get_feature_names()#获取词袋模型中的所有词语
weight=tfidf.toarray()#将tf-idf矩阵抽取出来，元素a[i][j]表示j词在i类文本中的tf-idf权重
total = total*salary1prob*salary2prob
# pdb.set_trace()
return total
try:
conn = MySQLdb.connect(host='localhost', user='******', passwd='123456', db='jobs', use_unicode=True, charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select industry, salary from work_sizetest'
cur.execute(sql)
salaryprobdct = utils.read_rst('salaryprobdct')
worklst = cur.fetchall()
i = 0
result = []
for j in xrange(20000):
salarys = worklst[i:i+2]
i += 2
salary_prob = {}
for key in range(7):
# pdb.set_trace()
salary_prob[key] = get_salary_prob(key, salaryprobdct, salarys)
# pdb.set_trace()
sortedprob = sorted(salary_prob.iteritems(), key=lambda jj:jj[1], reverse=True)
# for prob in sortedprob:
# print prob[0] + str(prob[1])
try:
conn = MySQLdb.connect(host='localhost', user='******', passwd='123456', db='jobs', use_unicode=True, charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select jb.userid,jb.degree,jb.age,jb.start_age,jb.bstart_year,jb.gender,jb.start_salary,wk.size \
from jobs_uinfotest as jb left join workexperiencetest as wk on \
jb.userid = wk.userid and wk.num = 1'
cur.execute(sql)
file = open('d:/jobs/dctree/bresult.csv', 'w+')
useridlst = cur.fetchall()
rsultlabel = utils.read_rst('result.txt')
pdb.set_trace()
#wsresult = utils.read_rst('wsresult.txt')
sizeresult = utils.read_rst('sizeresult.txt')
salaryresult = utils.read_rst('salaryresult.txt')
posresult = utils.read_rst('position.txt')
#degreelst = utils.read_rst('degree.txt')
degreelst = utils.read_rst('result.txt')
rsultlabel = map(str, rsultlabel)
file.write('id,degree,size,salary,position_name\n')
i = 0
print rsultlabel
pdb.set_trace()
j = 0
for userid in useridlst:
result = []
#coding:utf8
import sys
from jobs import utils
reload(sys)
import pdb
sys.setdefaultencoding('utf8')
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
# pdb.set_trace()
plabel1 = utils.read_rst('salaryresult1.txt')
plabel2 = utils.read_rst('salary')
i = 0
rst = 0
for label in plabel1:
if str(plabel2[i]) == label:
rst += 1
i += 1
print rst
# coding:utf8
import sys
from jobs import utils
reload(sys)
import pdb
sys.setdefaultencoding("utf8")
test_file = "d:/jobs/dctree/salary/test.csv"
labels = utils.get_labels(test_file, 6)
# pdb.set_trace()
plabel1 = utils.read_rst("position.txt")
plabel2 = utils.read_rst("positionlet")
i = 0
rst = 0
for label in plabel1:
if plabel2[i] == label:
rst += 1
i += 1
print rst
try:
conn = MySQLdb.connect(host="localhost", user="******", passwd="123456", db="jobs", use_unicode=True, charset="utf8")
cur = conn.cursor()
cur.execute("set character_set_client=utf8")
cur.execute("set character_set_connection=utf8")
cur.execute("set character_set_database=utf8")
cur.execute("set character_set_results=utf8")
cur.execute("set character_set_server=utf8")
sql = "select jb.userid,jb.degree,jb.age,jb.start_age,jb.bstart_year,jb.gender,jb.start_salary,wk.size \
from jobs_uinfotest as jb left join workexperiencetest as wk on \
jb.userid = wk.userid and wk.num = 1"
cur.execute(sql)
file = open("d:/jobs/baysian/position.csv", "w+")
useridlst = cur.fetchall()
rsultlabel = utils.read_rst("result.txt")
pdb.set_trace()
# wsresult = utils.read_rst('wsresult.txt')
sizeresult = utils.read_rst("sizeresult.txt")
salaryresult = utils.read_rst("salaryresult.txt")
posresult = utils.read_rst("positionlet")
# degreelst = utils.read_rst('degree.txt')
degreelst = utils.read_rst("result1.txt")
rsultlabel = map(str, rsultlabel)
file.write("id,degree,size,salary,position_name\n")
i = 0
print rsultlabel
# pdb.set_trace()
j = 0
for userid in useridlst:
result = []
start = time.clock()
try:
conn = MySQLdb.connect(host='localhost', user='******', passwd='123456', db='jobs', use_unicode=True, charset='utf8')
cur = conn.cursor()
cur.execute('set character_set_client=utf8')
cur.execute('set character_set_connection=utf8')
cur.execute('set character_set_database=utf8')
cur.execute('set character_set_results=utf8')
cur.execute('set character_set_server=utf8')
sql = 'select name, degreer0, degreer1, degreer2 from major ;'
cur.execute(sql)
majorlst = cur.fetchall()
majordct = {}
pdb.set_trace()
treelst = utils.read_rst('result.txt')
resultlst = []
for major in majorlst:
majordct[major[0]] = [major[1], major[2], major[3]]
sq = 'select userid, major from jobs_uinfotest'
cur.execute(sq)
usertst = cur.fetchall()
i = 0
for user in usertst:
if majordct.has_key(user[1]):
majorat = majordct[user[1]]
mnu = max(majorat)
index = majorat.index(mnu)
resultlst.append(index)
else:
resultlst.append(0)
from jobs import utils
import pdb
pdb.set_trace()
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
plabel = []
# for j in range(10):
# plabel.append(utils.read_rst('result03' +str(j)+ '.txt'))
# for j in range(10):
# plabel.append(utils.read_rst('result04' +str(j)+ '.txt'))
for j in range(10):
plabel.append(utils.read_rst('result1' +str(j)+ '.txt'))
for j in range(10):
plabel.append(utils.read_rst('result2' +str(j)+ '.txt'))
rst = [0 for j in xrange(20)]
i = 0
for label in labels:
for j in range(20):
if plabel[j][i] == label:
rst[j] += 1
i += 1
print rst
from jobs import utils
import pdb
test_file = 'd:/jobs/dctree/salary/test.csv'
labels = utils.get_labels(test_file, 6)
pdb.set_trace()
plabel = utils.read_rst('salary')
i = 0
rst = 0
for label in labels:
if plabel[i] == label:
rst += 1
i += 1
print rst
