def extractQuestFeatures(c, DIR):
if not os.path.exists(DIR + 'posts'):
os.makedirs(DIR + 'posts')
accepted_answers = pd.DataFrame(query.acceptedAnswers(c), columns=header['accepted_answers'])
accepted_answers.to_csv(DIR + 'posts/accepted_answers.csv', index=False)
post_text_data = pd.DataFrame(query.postsText_Data(c), columns=header['post_text_data'])
post_text_data = post_text_data.drop_duplicates(['PostId'])
post_text_data.to_csv(DIR + 'posts/post_text_data.csv', encoding='utf-8', index=False)
quest_stats = pd.DataFrame(query.questStats(c), columns=header['quest_stats'])
quest_stats.to_csv(DIR + 'posts/quest_stats.csv', index=False)
def extractAnsTime(c, DIR, MODE, VERSION):
if not os.path.exists(DIR + 'posts'):
os.makedirs(DIR + 'posts')
if MODE == 'FIRST':
ans_time_first = query.firstAnsTime(c)
util.save_to_csv(ans_time_first, header['ans_time_first'], 'posts/ans_time_first.csv')
elif MODE == 'ACCEPTED':
ans_time_accepted = pd.DataFrame(query.acceptAnsTime(c), columns=header['ans_time_accepted'])
ans_time_accepted.to_csv(DIR + 'posts/ans_time_accepted.csv', index=False)
elif MODE == 'UPVOTED':
ans_time_upvoted = query.upvotedAnsTime(c)
util.save_to_csv(ans_time_upvoted, header['ans_time_upvoted'], 'ans_time_upvoted.csv')
if VERSION == 'EXTENDED':
ans_time_upvoted_ex = query.upvotedAnsTime1(c)
util.save_to_csv(ans_time_upvoted_ex, header['ans_time_upvoted_ex'], 'ans_time_upvoted_ex.csv')
def extractUserFeatures(c, DIR):
if not os.path.exists(DIR + 'users'):
os.makedirs(DIR + 'users')
quest_per_user = pd.DataFrame(query.questPerUser(c), columns=header['quest_per_user'])
quest_per_user.to_csv(DIR + 'users/quest_per_user.csv', index=False)
answer_per_user = pd.DataFrame(query.ansPerUser(c), columns=header['answer_per_user'])
answer_per_user.to_csv(DIR + 'users/answer_per_user.csv', index=False)
posts_per_user = pd.DataFrame(query.postsPerUser(c), columns=header['posts_per_user'])
posts_per_user.to_csv(DIR + 'users/posts_per_user.csv', index=False)
users_stats = pd.DataFrame(query.usersStats(c), columns=header['users_stats'])
users_stats.to_csv(DIR + 'users/users_stats.csv', encoding='utf-8', index=False)
users_av_ans_time = pd.DataFrame(query.usersAvAnsTime(c), columns=header['users_av_ans_time'])
users_av_ans_time.to_csv(DIR + 'users/users_av_ans_time.csv', index=False)
def extractUserActivityFeatures(c, DIR):
if not os.path.exists(DIR + 'users'):
os.makedirs(DIR + 'users')
users_activity = pd.DataFrame(query.usersActivity(c), columns=header['users_activity'])
users_activity.to_csv(DIR + 'posts/users_activity.csv', index=False)
posts = useractivity.makePosts(users_activity)
posts = posts[posts['UserId'] != -1]
return posts
def main(argv):
setPlotDesign()
data = qry.queryDB(
"select Date_format(First_Occurrence_Date, '%H') as 'Hour', Count(1) as 'Crime_Count' from DenverCrime where Is_Crime = 1 group by Date_format(First_Occurrence_Date, '%H');"
)
queryToPlot(data, "#6DA2BE", 0, "All Crimes")
data = qry.queryDB(
"select Date_format(First_Occurrence_Date, '%H') as 'Hour', Count(1) as 'Crime_Count' from DenverCrime where Is_Crime = 1 and Offense_Code = "
+ offense[0] + " group by Date_format(First_Occurrence_Date, '%H');")
#plt.subplot(112)
queryToPlot(data, "#C16F65", .3, offense[1])
plt.legend(loc="upper center", frameon=False, fontsize=12)
#plt.show()
plt.savefig("CrimeHourlyBarGraphs/AllCrimes-" + offense[2] +
"-percent.png",
bbox_inches="tight",
dpi=300)
def main(argv):
cursor = qry.queryDB(
"select TMax, Round(Sum(Crime_Count)/Count(1)) as CrimeCountAvg from DenverDailyCrimeWeather group by TMax"
)
tMaxList = []
crimeCountAvgList = []
for (tMax, crimeCountAvg) in cursor:
if tMax is not None:
tMaxList.append(float(tMax))
crimeCountAvgList.append(int(crimeCountAvg))
plot(tMaxList, crimeCountAvgList)
def main(argv):
cursor = qry.queryDB(
"select Geo_Lon, Geo_Lat from DenverCrime where Geo_Lat > " +
str(minGeoLat) + " and Geo_Lat < " + str(maxGeoLat) +
" and Geo_Lon < " + str(maxGeoLon) + " and Geo_Lon > " +
str(minGeoLon) + " and Offense_Code = " + crime[1])
locations = []
for (geoLon, geoLat) in cursor:
if geoLon is not None and geoLat is not None:
locations.append(Location(geoLon, geoLat))
#downloadMap()
kMeans(locations, k, iteration)
def constructMatrix(self, donations):
fullDict = q.getAllRecent()
donationDict = {}
donationDictReverse = {}
count = 0
checked = []
m = []
spectralm = []
for donation in donations:
self.data[donation[0]][1][0].pop()
pastGrantYear = self.data[donation[0]][1][0] + donation[1:6]
currentYear = list(fullDict[donation[0]]) + [COVER, COVER, COVER, COVER, COVER]
m.append(pastGrantYear)
m.append(currentYear)
spectralm.append(self.data[donation[0]][1][0])
spectralm.append(list(fullDict[donation[0]]))
donationDict[count] = donation[0]
donationDictReverse[count + 1] = donation[0]
donationDict[donation[0]] = (count, count + 1)
checked.append(donation[0])
count = count + 2
for key, value in fullDict.items():
if key not in set(checked):
m.append(list(value) + [COVER, COVER, COVER, COVER, COVER])
spectralm.append(list(value))
donationDict[key] = (count)
donationDictReverse[count] = key
count = count + 1
np.savetxt("test.csv", np.matrix(m), delimiter=",")
sio.savemat('data.mat', {'a_dict': np.matrix(m)})
with open('dict.json', 'w') as fp:
json.dump(donationDict, fp)
with open('dictReverse.json', 'w') as fp:
json.dump(donationDictReverse, fp)
return spectralm
def init():
try:
test = QueryDB.connection("bolt://localhost:7687", "neo4j", "test")
except:
print("Cannot connect to server")
sys.exit()
page = Tk()
cypher = StringVar()
output = StringVar()
page.title("NLP Input")
page.geometry("700x350")
app = Frame(page)
app.grid()
btn1 = Button(app,
text="Submit",
command=lambda: setOutput(e1.get(), cypher, output, test))
lbl1 = Label(app, text="Enter your search: ")
lbl2 = Label(app, text="Query in Cypher: ")
lbl3 = Label(app, textvariable=cypher)
lbl4 = Label(app, text="Query result: ")
lbl5 = Label(app, textvariable=output)
e1 = Entry(app)
lbl1.grid(row=0)
lbl2.grid(row=2)
e1.grid(row=0, column=1)
lbl3.grid(row=2, column=1)
btn1.grid(row=0, column=2, padx=10, pady=10)
lbl4.grid(row=3, column=0)
lbl5.grid(row=3, column=1)
page.mainloop()
for index, r in enumerate(computed_mat):
if index < 76:
if index % 2 != 0:
translated_mat.append((fd.findRow(str(index)),
map(decFunc, r)))
else:
translated_mat.append((fd.findRow(str(index)),r))
ranking = sorted(translated_mat, key=lambda s: judgement(s[1]))
ranking = map(lambda x: (x[0], judgement(x[1])), ranking)
return ranking[::-1]
def spectralClustering(self):
S = s.SpectralClustering(n_clusters=2, gamma=1.0, \
affinity='rbf', n_neighbors=10, assign_labels='kmeans')
mat = self.performRegression()
mat = np.asmatrix(mat)
#print np.size(mat[:, [35, 36]])
X, test = d.make_circles(10000)
print np.size(X)
S.fit(X)
print S.fit_predict(X)
if __name__ == "__main__":
R = Regression(q.getDonatedSchools())
# R.rankWinners(lambda x: x[4])
R.performRegression()
#R.spectralClustering()
import tornado.web import tornado.ioloop from pymongo import MongoClient import PostServer import QueryDB # Connects to the default host and port on mongodb client = MongoClient() db = client.db query_db = QueryDB.QueryHandler(db) # Maps root URL to MainHandler # and passes db reference app = tornado.web.Application([(r"/", PostServer.MainHandler, dict(db=db))]) loop = tornado.ioloop.IOLoop.instance()
db_host = '127.0.0.1'
db_user = '******'
db_pw = ''
db_name = 'infouser'
# dlf_project_id = 'fir-thesis-v1'
# dlf_session_id = 'a409931f1c2ea9842c238469417ae2ac490b36bd'
# mgDLF = mgdia.manageDLF(projectID=dlf_project_id,
# sessionID=dlf_session_id)
conDB = conn.connectDB(host=db_host,
username=db_user,
password=db_pw,
database=db_name)
qy = qr.queryDB(conDB.getConnection())
dis_data = [
'มีอารมณ์ซึมเศร้า', 'มีอารมณ์หงุดหงิด', 'มีอารมณ์ก้าวร้าว',
'ขาดความสนใจสิ่งรอบข้าง', 'ไม่ค่อยมีสมาธิเวลาทำสิ่งต่างๆ',
'รู้สึกอ่อนเพลีย', 'ทำอะไรก็เชื่องช้า', '่รับประทานอาหารมากขึ้น',
'รับประทานน้อยลง', 'นอนมากขึ้น', 'นอนน้อยลง',
'ตำหนิตัวเองเป็นอันดับแรกถ้ามีอะไรพลาด', 'พยายามฆ่าตัวตาย'
]
print(dis_data)
for dis in dis_data:
query = 'INSERT INTO fact_data(fact_name) VALUES (\'{}\')'.format(dis)
import parse
import QueryDB
myConnection = QueryDB.connection("bolt://localhost:7687", "neo4j", "test")
infile = open("testQueries.txt", "r")
translations = []
results = []
for line in infile:
translations.append(parse.init(line.strip()))
for line in translations:
results.append(myConnection.run_Return_Query(line).strip())
infile.close
infile = open("testResults.txt", "r")
answers = []
for line in infile:
answers.append(line.strip())
for i in range(0, len(answers)):
if (answers[i] == results[i]):
print("Querie ", i, " successful")
else:
print("Querie ", i, " failed!")
print("\tExpected:\t", answers[i])
print("\tActual:\t", results[i])
infile.close
