In [1]:
%pylab
from __future__ import print_function
%matplotlib inline
In [2]:
from __future__ import print_function
import csv
from src import main, utils
import sys
projects = main.load_projects()
interests = ['argouml v0.26.2 method', 'mucommander v0.8.5 method']
p = list()
for project in projects:
if ' '.join([project.name, project.version, project.level]) not in interests:
continue
p.append(project)
In [3]:
project = p[1]
print(project)
repos = main.load_repos(project)
goldsets = main.load_goldsets(project)
queries = main.create_queries(project)
snapshot = main.create_release_corpus(project, repos)
changesets = main.create_corpus(project, repos, main.ChangesetCorpus, use_level=False)
snapshot_lda, _ = main.create_lda_model(project, snapshot, None, "Release", use_level=True)
changeset_lda, _ = main.create_lda_model(project, changesets, None, "Changeset", use_level=False)
In [12]:
queries.metadata = True
qs = list(filter(lambda x: x[1][0] in ["37", "142"], queries))
for query, metadata in qs:
doc = sorted(query, key=lambda x: x[1], reverse=True)
words = [ ( queries.id2word[wid], freq ) for wid, freq in doc ]
print("{0}: {1}".format(metadata[0], words))
In [13]:
q = dict()
for query, topics in main.get_topics(snapshot_lda, queries, by_ids=["37", "142"], full=False):
topics = sorted(topics, key=lambda x: x[1], reverse=True)
q[query[0]] = topics
print(q)
for qid, topics in q.items():
print("{0}: {1}".format(qid, snapshot_lda.print_topic(topics[0][0])))
In [14]:
qc = dict()
for query, topics in main.get_topics(changeset_lda, queries, by_ids=["37", "142"], full=False):
topics = sorted(topics, key=lambda x: x[1], reverse=True)
qc[query[0]] = topics
print(qc)
for qid, topics in qc.items():
print("{0}: {1}".format(qid, changeset_lda.print_topic(topics[0][0])))
In [30]:
g37 = ["com.mucommander.ui.main.menu.MainMenuBar.MainMenuBar(MainFrame)",
"com.mucommander.ui.action.impl.MinimizeWindowAction.performAction()"]
g142 = ["com.mucommander.ui.main.DrivePopupButton.RefreshDriveNamesAndIcons.run()",
"com.mucommander.ui.main.StatusBar.VolumeSpaceLabel.paint(Graphics)"]
In [31]:
q37 = dict()
for query, topics in main.get_topics(snapshot_lda, snapshot, by_ids=[g37[0]], full=False):
topics = sorted(topics, key=lambda x: x[1], reverse=True)
q37[query[0]] = topics
#print(q37)
for qid, topics in q37.items() -
print("{0}: {1}".format(qid, snapshot_lda.print_topic(topics[0][0])))
In [32]:
qc37 = dict()
for query, topics in main.get_topics(changeset_lda, snapshot, by_ids=[g37[1]], full=False):
topics = sorted(topics, key=lambda x: x[1], reverse=True)
qc37[query[0]] = topics
#print(qc37)
for qid, topics in qc37.items():
print("{0}: {1}".format(qid, changeset_lda.print_topic(topics[0][0])))
In [33]:
q142 = dict()
for query, topics in main.get_topics(snapshot_lda, snapshot, by_ids=[g142[0]], full=False):
topics = sorted(topics, key=lambda x: x[1], reverse=True)
q142[query[0]] = topics
#print(q142)
for qid, topics in q142.items():
print("{0}: {1}".format(qid, snapshot_lda.print_topic(topics[0][0])))
In [34]:
qc142 = dict()
for query, topics in main.get_topics(changeset_lda, snapshot, by_ids=[g142[1]], full=False):
topics = sorted(topics, key=lambda x: x[1], reverse=True)
qc142[query[0]] = topics
#print(qc142)
for qid, topics in qc142.items():
print("{0}: {1}".format(qid, changeset_lda.print_topic(topics[0][0])))
In [ ]: