178 lines
6.3 KiB
Python
178 lines
6.3 KiB
Python
from __future__ import unicode_literals
|
|
#import imapclient
|
|
from config import Config
|
|
import sys
|
|
#from email.header import decode_header
|
|
#import email
|
|
import codecs
|
|
|
|
#import sys
|
|
#import bs4
|
|
import yaml
|
|
#sys.stdout = codecs.getwriter('utf8')(sys.stdout)
|
|
from storage.fetch_mail import fetch_mail
|
|
from storage.fetch_mail import fetch_threads, flatten_threads
|
|
from storage import Mail, MailThread, db_session
|
|
#import yaml
|
|
#import email
|
|
from classifier import get_training_threads, print_answers,get_pipe, test_pipe, train_single_thread # , pipe2, pipe2b
|
|
from classifier import predict_threads
|
|
maintopic_values=["studium", "information","ausleihen","jobausschreibung", "umfragen"]
|
|
|
|
def predict_thread(p,l,t):
|
|
pre=p.predict([t])
|
|
print "Status is answered is estimated to be: " + str(l.inverse_transform(pre)[0])
|
|
return pre
|
|
|
|
|
|
#print "arg1:"+sys.argv[1]
|
|
if len(sys.argv)>1:
|
|
|
|
if sys.argv[1] == "fetch_threads":
|
|
print flatten_threads(fetch_threads())
|
|
if sys.argv[1] == "predict_threads":
|
|
predict_threads()
|
|
if sys.argv[1]=="stats":
|
|
for topic in maintopic_values:
|
|
print topic
|
|
n_answ=db_session.query(MailThread).filter(MailThread.maintopic==topic).filter(MailThread.answered.op("IS")(True)).count()
|
|
n_nansw=db_session.query(MailThread).filter(MailThread.maintopic==topic).filter(MailThread.answered.op("IS NOT")(True)).count()
|
|
n_ges=db_session.query(MailThread).filter(MailThread.maintopic==topic).count()
|
|
print "%d answered and %d not answered of %d(%d) that are %d percent answerd" % (n_answ,n_nansw, n_ges,n_answ+n_nansw, float(n_answ)/float(n_ges)*100.0)
|
|
|
|
|
|
|
|
|
|
if sys.argv[1] == "run_server":
|
|
from flaskapp import app
|
|
app.run(port=3000,debug=True)
|
|
|
|
|
|
if sys.argv[1] == "train_thrd2":
|
|
p, le=get_pipe("pipe2", "maintopic",["db"])
|
|
pb, lb =get_pipe("pipe2b", "maintopic",["db"])
|
|
|
|
train_single_thread(int(sys.argv[2]),p,le,b"maintopic")
|
|
|
|
if sys.argv[1] == "train_thrd3":
|
|
# p, le=get_pipe("pipe2", "maintopic")
|
|
pb, lb =get_pipe("pipe2b", "lang")
|
|
|
|
train_single_thread(int(sys.argv[2]),pb,lb,b"lang")
|
|
|
|
|
|
if sys.argv[1] == "train_all2":
|
|
p, labelencoder=train_fit_pipe2()
|
|
pb, lb=train_fit_pipe2b()
|
|
mth=db_session.query(MailThread).all()
|
|
print mth
|
|
for t in mth:
|
|
if not in_training(t.firstmail,"maintopic"):
|
|
print "---------------------------------------------------"
|
|
print "---------------------------------------------------"
|
|
print t.firstmail
|
|
print t.text()
|
|
predict_thread(pb,lb,t)
|
|
train_single_thread(t.firstmail, p, labelencoder, b"maintopic")
|
|
if sys.argv[1] == "benchpipe3":
|
|
test_pipe(["pipe2d","pipe2e","pipe2e1","pipe2f","pipe2g"],"maintopic",["db","de"])
|
|
|
|
if sys.argv[1] == "benchpipe2":
|
|
test_pipe(["pipe2","pipe2b","pipe2c","pipe2d"],"maintopic",["db","de"])
|
|
# print "testing with db training data:"
|
|
# test_pipe(["pipe2b"],"maintopic",["db"])
|
|
# test_pipe(["pipe2b"],"maintopic",["db"])
|
|
# print "testing only with german data"
|
|
# test_pipe(["pipe2b"],"maintopic",["db","de"])
|
|
|
|
if sys.argv[1] == "testpipe2":
|
|
from classifier import ThreadSubjectExtractor, ThreadTextExtractor
|
|
pipe2,le=train_fit_pipe2()
|
|
|
|
if len(sys.argv)>2:
|
|
t=db_session.query(MailThread).filter(MailThread.firstmail==sys.argv[2]).first()
|
|
print t.to_text()
|
|
print le.inverse_transform(pipe2.predict([t]))
|
|
|
|
|
|
|
|
if sys.argv[1] == "train_thrd":
|
|
pipe1, labelencoder=train_fit_pipe()
|
|
train_single_thread(int(sys.argv[2]),pipe1,labelencoder)
|
|
|
|
if sys.argv[1] == "train_all":
|
|
pipe1, labelencoder=train_fit_pipe()
|
|
mth=db_session.query(MailThread).all()
|
|
print mth
|
|
for t in mth:
|
|
if not in_training(t.firstmail):
|
|
print "---------------------------------------------------"
|
|
print "---------------------------------------------------"
|
|
print t.firstmail
|
|
train_single_thread(t.firstmail,pipe1,labelencoder)
|
|
|
|
if sys.argv[1] == "print_thread":
|
|
mth=db_session.query(MailThread).filter(MailThread.firstmail==int(sys.argv[2])).first()
|
|
print mth.mail_dicts()
|
|
print mth.mail_flat_dict()
|
|
|
|
if sys.argv[1] == "store_threads":
|
|
thrds=flatten_threads(fetch_threads())
|
|
for t in thrds:
|
|
if type(t[0]) is int:
|
|
th=db_session.query(MailThread).filter(MailThread.firstmail==t[0]).first()
|
|
if th == None:
|
|
th=MailThread()
|
|
th.firstmail=t[0]
|
|
if not th.body == yaml.dump(t):
|
|
th.body=yaml.dump(t)
|
|
th.islabeled=False
|
|
th.opened=True
|
|
th.compile()
|
|
db_session.add(th)
|
|
db_session.commit()
|
|
print thrds
|
|
|
|
if sys.argv[1] == "print_raw_mail":
|
|
mm=db_session.query(Mail).filter(Mail.id==int(sys.argv[2])).first()
|
|
print yaml.load(mm.envelope)
|
|
|
|
if sys.argv[1] == "print_mail":
|
|
mm=db_session.query(Mail).filter(Mail.id==int(sys.argv[2])).first()
|
|
mm.compile_text()
|
|
mm.compile_envelope()
|
|
print mm.subject
|
|
print "----------"
|
|
print mm.text
|
|
|
|
if sys.argv[1] == "mail_dict_test":
|
|
mm=db_session.query(Mail).filter(Mail.id==int(sys.argv[2])).first()
|
|
mm.compile_envelope()
|
|
print mm.dict_envelope()
|
|
|
|
|
|
if sys.argv[1] == "load_mail":
|
|
mm=db_session.query(Mail).filter(Mail.id==int(sys.argv[2])).first()
|
|
mm.compile_text()
|
|
print mm.text
|
|
env=yaml.load(mm.envelope)
|
|
print env.subject
|
|
print env
|
|
|
|
|
|
if sys.argv[1] == "store_mail":
|
|
m=fetch_mail(int(sys.argv[2]))
|
|
mm=Mail()
|
|
mm.envelope=yaml.dump(m['ENVELOPE'])
|
|
mm.body=yaml.dump(m['RFC822'])
|
|
mm.id=m['id']
|
|
db_session.add(mm)
|
|
db_session.commit()
|
|
|
|
|
|
|
|
|
|
if sys.argv[1] == "initdb":
|
|
from storage import init_db
|
|
init_db()
|