def main():
f = file("comment.classifier", "rb")
classifier = pickle.load(f)
f.close()
db = util.makeDbConnection()
classifyComments(db, classifier)
db.close()
def root():
query = flask.request.args.get("q", "")
logging.debug("query=%s" % query)
postResults = []
userResults = []
if query:
logging.debug("connecting to the database...")
db = util.makeDbConnection(database)
logging.debug("querying the topic model...")
postResults = topicModel.queryResults(db, query, resultCutoff)
logging.debug("%d results returned..." % len(postResults))
if postResults is not None:
logging.debug("scoring users...")
userResults = scoring.scoreUsers(db,
query,
postResults,
topicModel,
cutoffPercentile=percentileCutoff,
resultCutoff=resultCutoff)
logging.debug("star-scoring users...")
userResults = [
userResult.starScore(cutoffPercentile=percentileCutoff,
nStars=5) for userResult in userResults
]
db.close()
return flask.render_template("experts.html",
query=query,
users=userResults,
posts=postResults)
def main():
f = file("comment.classifier", "rb")
classifier = pickle.load(f)
f.close()
db = util.makeDbConnection()
classifyComments(db, classifier)
db.close()
def testIterate():
try:
db = util.makeDbConnection("stackoverflow")
dictionary = gensim.corpora.dictionary.Dictionary()
for doc in StackOverflowCorpus(db, dictionary):
print "doc=", doc, "dictionary=", dictionary
raw_input()
finally:
db.close()
def testIterate():
try:
db=util.makeDbConnection("stackoverflow")
dictionary = gensim.corpora.dictionary.Dictionary()
for doc in StackOverflowCorpus(db, dictionary):
print "doc=", doc, "dictionary=",dictionary
raw_input()
finally:
db.close()
def root():
query = flask.request.args.get("q", "")
logging.debug("query=%s" % query)
postResults = []
userResults = []
if query:
logging.debug("connecting to the database...")
db = util.makeDbConnection(database)
logging.debug("querying the topic model...")
postResults = topicModel.queryResults(db, query, resultCutoff)
logging.debug("%d results returned..." % len(postResults))
if postResults is not None:
logging.debug("scoring users..." )
userResults = scoring.scoreUsers(db, query, postResults, topicModel, cutoffPercentile=percentileCutoff, resultCutoff=resultCutoff)
logging.debug("star-scoring users...")
userResults = [userResult.starScore(cutoffPercentile=percentileCutoff, nStars=5) for userResult in userResults]
db.close()
return flask.render_template("experts.html", query=query, users=userResults, posts=postResults)
def makeStackOverflowCorpus(fileName, topic=None, usePostList=False, useTags=[]):
db=util.makeDbConnection(Config.myDb)
if usePostList:
print "Using post list"
if not useTags:
useTags = util.topTags(db, 200)
postList = util.tagPosts(db, useTags)
print "Need to import", len(postList), "posts"
else:
postList = None
dictionary = gensim.corpora.dictionary.Dictionary()
soCorpus = StackOverflowCorpus(db, dictionary, topic, postList)
try:
gensim.corpora.MmCorpus.serialize(fileName + ".mm", soCorpus)
finally:
dictionary.save(fileName + ".dict")
soCorpus.saveCorpusToPost(fileName + ".c2p")
db.close()
return len(postList)
def makeStackOverflowCorpus(fileName,
topic=None,
usePostList=False,
useTags=[]):
db = util.makeDbConnection(Config.myDb)
if usePostList:
print "Using post list"
if not useTags:
useTags = util.topTags(db, 200)
postList = util.tagPosts(db, useTags)
print "Need to import", len(postList), "posts"
else:
postList = None
dictionary = gensim.corpora.dictionary.Dictionary()
soCorpus = StackOverflowCorpus(db, dictionary, topic, postList)
try:
gensim.corpora.MmCorpus.serialize(fileName + ".mm", soCorpus)
finally:
dictionary.save(fileName + ".dict")
soCorpus.saveCorpusToPost(fileName + ".c2p")
db.close()
return len(postList)
def main():
""" create precalculated scoring table """
db = util.makeDbConnection()
createPrescoringTables(db)
db.close()
