def evaluate_clustering():
"""
Evaluate the clustering algorithm.
"""
logger.info('Constructing expected clusters and articles...')
expected_clusters = {}
articles = []
all_files = []
# Collect all appropriate files.
for dir, subdir, files in os.walk('manage/evaluate/organized_articles'):
for file in files:
filepath = os.path.join(dir, file)
name, ext = os.path.splitext(filepath)
if ext == '.txt':
all_files.append((dir, name, filepath))
# Create articles for appropriate files.
for dir, name, filepath in all_files:
category = dir.split('/')[-1]
f = open(filepath, 'r')
article = Article(
text=f.read(),
title=name.split('/')[-1]
)
expected_clusters.setdefault(category, []).append(article)
articles.append(article)
progress_bar(len(articles)/len(all_files) * 100)
print('\n')
logger.info('Will cluster {0} articles.'.format(len(articles)))
logger.info('Expecting {0} clusters.'.format(len(expected_clusters.keys())))
logger.info('Clustering...')
p = cProfile.Profile()
clusters = p.runcall(Event.cluster, articles, threshold=0.04, debug=True)
logger.info('Created {0} clusters.'.format(len(clusters)))
logger.info('Cluster composition is as follows...')
for c in clusters:
logger.info([m.title for m in c.members])
logger.info('Profiling statistics from the clustering...')
ps = pstats.Stats(p)
ps.sort_stats('time').print_stats(10)
def evaluate_clustering():
"""
Evaluate the clustering algorithm.
"""
logger.info('Constructing expected clusters and articles...')
expected_clusters = {}
articles = []
all_files = []
# Collect all appropriate files.
for dir, subdir, files in os.walk('manage/evaluate/organized_articles'):
for file in files:
filepath = os.path.join(dir, file)
name, ext = os.path.splitext(filepath)
if ext == '.txt':
all_files.append((dir, name, filepath))
# Create articles for appropriate files.
for dir, name, filepath in all_files:
category = dir.split('/')[-1]
f = open(filepath, 'r')
article = Article(text=f.read(), title=name.split('/')[-1])
expected_clusters.setdefault(category, []).append(article)
articles.append(article)
progress_bar(len(articles) / len(all_files) * 100)
print('\n')
logger.info('Will cluster {0} articles.'.format(len(articles)))
logger.info('Expecting {0} clusters.'.format(len(
expected_clusters.keys())))
logger.info('Clustering...')
p = cProfile.Profile()
clusters = p.runcall(Event.cluster, articles, threshold=0.04, debug=True)
logger.info('Created {0} clusters.'.format(len(clusters)))
logger.info('Cluster composition is as follows...')
for c in clusters:
logger.info([m.title for m in c.members])
logger.info('Profiling statistics from the clustering...')
ps = pstats.Stats(p)
ps.sort_stats('time').print_stats(10)
def generate(keywords, num=5000):
this_dir = path.dirname(__file__)
articles = load_articles()
# Filter down to articles from the specified sources.
results = []
articles = articles[:num]
for idx, article in enumerate(articles):
article_words = tokenize(article['text'])
if set(article_words).issuperset(set(keywords)):
results.append(article)
progress_bar(idx / len(articles) * 100)
# Store articles into separate text files.
for article in results:
#print(json.dumps(article, sort_keys=True, indent=4))
article_path = 'unorganized_articles/{0}.txt'.format(article['title'])
f = open(path.join(this_dir, article_path), 'wb')
f.write(article['text'].encode('utf-8'))
def generate(keywords, num=5000):
this_dir = path.dirname(__file__)
articles = load_articles()
# Filter down to articles from the specified sources.
results = []
articles = articles[:num]
for idx, article in enumerate(articles):
article_words = tokenize(article['text'])
if set(article_words).issuperset(set(keywords)):
results.append(article)
progress_bar(idx/len(articles) * 100)
# Store articles into separate text files.
for article in results:
#print(json.dumps(article, sort_keys=True, indent=4))
article_path = 'unorganized_articles/{0}.txt'.format(article['title'])
f = open(path.join(this_dir, article_path), 'wb')
f.write(article['text'].encode('utf-8'))
def seed(debug=False):
this_dir = os.path.dirname(__file__)
seeds = open(os.path.join(this_dir, 'seed.json'), 'r')
sources = open(os.path.join(this_dir, 'seed_sources.json'), 'r')
sample_images = [
'https://upload.wikimedia.org/wikipedia/commons/d/d5/Michael_Rogers_-_Herbiers_2004.jpg',
'https://upload.wikimedia.org/wikipedia/commons/6/6e/Brandenburger_Tor_2004.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/a/ad/ChicagoAntiGaddafiHopeless.jpg/576px-ChicagoAntiGaddafiHopeless.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/e/e6/Evo_morales_2_year_bolivia_Joel_Alvarez.jpg/640px-Evo_morales_2_year_bolivia_Joel_Alvarez.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/9/9f/2010_wet_season_cloud_over_colombia.jpg/640px-2010_wet_season_cloud_over_colombia.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/2/27/Barack_Obama_at_Las_Vegas_Presidential_Forum.jpg/640px-Barack_Obama_at_Las_Vegas_Presidential_Forum.jpg'
'https://upload.wikimedia.org/wikipedia/commons/thumb/a/a9/2010-10-23-Demo-Stuttgart21-Befuerworter.png/640px-2010-10-23-Demo-Stuttgart21-Befuerworter.png',
'https://upload.wikimedia.org/wikipedia/commons/thumb/f/fc/51%C2%BA_Congresso_da_UNE_%28Conune%29.jpg/640px-51%C2%BA_Congresso_da_UNE_%28Conune%29.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Tough_return%2C_365.35.jpg/640px-Tough_return%2C_365.35.jpg'
]
print('Resetting the database...')
db.drop_all()
db.create_all()
# Create sources
print('Creating sources...')
for url in json.load(sources):
s = Source(ext_url=url, name='The Times') # fake name
db.session.add(s)
db.session.commit()
num_sources = Source.query.count()
print('Created {0} sources.'.format(num_sources))
# Create articles
entries = json.load(seeds)
print('Seeding {0} articles...'.format(len(entries)))
articles = []
for entry in entries:
if debug:
print(json.dumps(entry, sort_keys=True, indent=4))
source = Source.query.filter_by(ext_url=entry['source']).first()
a = Article(
ext_url=entry['url'],
source=source,
html=entry['html'],
text=entry['text'],
tags=entry['tags'],
title=entry['title'],
created_at = parse(entry['published']),
updated_at = parse(entry['updated']),
image=random.choice(sample_images) # fake image
)
articles.append(a)
db.session.add(a)
progress_bar(len(articles) / len(entries) * 100)
db.session.commit()
num_articles = Article.query.count()
num_entities = Entity.query.count()
print('Seeded {0} articles.'.format(num_articles))
print('Found {0} entities.'.format(num_entities))
print('Clustering articles into events...')
Event.cluster(articles, threshold=0.02, debug=True)
num_events = Event.query.count()
print('Created {0} event clusters.'.format(num_events))
print('Clustering events into stories...')
events = Event.query.all()
Story.cluster(events, threshold=0.02, debug=True)
num_stories = Story.query.count()
print('Created {0} story clusters.'.format(num_stories))
print('\n\n==============================================')
print('From {0} sources, seeded {1} articles, found {2} entities, created {3} events and {4} stories.'.format(num_sources, num_articles, num_entities, num_events, num_stories))
print('==============================================\n\n')
def seed(debug=False):
# Patch out saving images to S3.
patcher = patch('argos.util.storage.save_from_url', autospec=True, return_value='https://i.imgur.com/Zf9mXlj.jpg')
patcher.start()
seeds = open('manage/core/data/seed.json', 'r')
sample_images = [
'https://upload.wikimedia.org/wikipedia/commons/d/d5/Michael_Rogers_-_Herbiers_2004.jpg',
'https://upload.wikimedia.org/wikipedia/commons/6/6e/Brandenburger_Tor_2004.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/a/ad/ChicagoAntiGaddafiHopeless.jpg/576px-ChicagoAntiGaddafiHopeless.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/e/e6/Evo_morales_2_year_bolivia_Joel_Alvarez.jpg/640px-Evo_morales_2_year_bolivia_Joel_Alvarez.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/9/9f/2010_wet_season_cloud_over_colombia.jpg/640px-2010_wet_season_cloud_over_colombia.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/2/27/Barack_Obama_at_Las_Vegas_Presidential_Forum.jpg/640px-Barack_Obama_at_Las_Vegas_Presidential_Forum.jpg'
'https://upload.wikimedia.org/wikipedia/commons/thumb/a/a9/2010-10-23-Demo-Stuttgart21-Befuerworter.png/640px-2010-10-23-Demo-Stuttgart21-Befuerworter.png',
'https://upload.wikimedia.org/wikipedia/commons/thumb/f/fc/51%C2%BA_Congresso_da_UNE_%28Conune%29.jpg/640px-51%C2%BA_Congresso_da_UNE_%28Conune%29.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/f/f7/Tough_return%2C_365.35.jpg/640px-Tough_return%2C_365.35.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/8/83/Saint_Isaac%27s_Cathedral_in_SPB.jpeg/800px-Saint_Isaac%27s_Cathedral_in_SPB.jpeg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/a/a5/Ponta_de_S%C3%A3o_Louren%C3%A7o_north_north_east.jpg/800px-Ponta_de_S%C3%A3o_Louren%C3%A7o_north_north_east.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/e/e9/TU_Bibl_01_DSC1099w.jpg/644px-TU_Bibl_01_DSC1099w.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/f/fe/%D0%9C%D0%B0%D0%BA%D0%B5%D0%B4%D0%BE%D0%BD%D0%B8%D1%83%D0%BC_-_%D0%9A%D1%80%D1%83%D1%88%D0%B5%D0%B2%D0%BE.jpg/800px-%D0%9C%D0%B0%D0%BA%D0%B5%D0%B4%D0%BE%D0%BD%D0%B8%D1%83%D0%BC_-_%D0%9A%D1%80%D1%83%D1%88%D0%B5%D0%B2%D0%BE.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/6/64/Puente_Mong%2C_Ciudad_Ho_Chi_Minh%2C_Vietnam%2C_2013-08-14%2C_DD_01.JPG/800px-Puente_Mong%2C_Ciudad_Ho_Chi_Minh%2C_Vietnam%2C_2013-08-14%2C_DD_01.JPG',
'https://upload.wikimedia.org/wikipedia/commons/thumb/c/c3/Autignac%2C_H%C3%A9rault_01.jpg/800px-Autignac%2C_H%C3%A9rault_01.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/6/6a/Caesio_teres_in_Fiji_by_Nick_Hobgood.jpg/800px-Caesio_teres_in_Fiji_by_Nick_Hobgood.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/9/9a/Ash_in_Yogyakarta_during_the_2014_eruption_of_Kelud_01.jpg/800px-Ash_in_Yogyakarta_during_the_2014_eruption_of_Kelud_01.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/b/bd/12-07-12-wikimania-wdc-by-RalfR-010.jpg/800px-12-07-12-wikimania-wdc-by-RalfR-010.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/7/71/Mortagne-sur-Gironde_Civellier_Mayflowers_2013.jpg/800px-Mortagne-sur-Gironde_Civellier_Mayflowers_2013.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/5/5a/British_Museum_Great_Court%2C_London%2C_UK_-_Diliff.jpg/611px-British_Museum_Great_Court%2C_London%2C_UK_-_Diliff.jpg',
'https://upload.wikimedia.org/wikipedia/commons/thumb/2/26/Mercedes-Benz_Museum_201312_08_blue_hour.jpg/800px-Mercedes-Benz_Museum_201312_08_blue_hour.jpg'
]
print('Resetting the database...')
db.reflect()
db.drop_all()
db.create_all()
# Create sources
print('Creating sources...')
create_sources()
num_sources = Source.query.count()
print('Created {0} sources.'.format(num_sources))
# Create articles
entries = json.load(seeds)
print('Seeding {0} articles...'.format(len(entries)))
articles = []
for entry in entries:
if debug:
print(json.dumps(entry, sort_keys=True, indent=4))
feed = Feed.query.filter_by(ext_url=entry['source']).first()
a = Article(
ext_url=entry['url'],
feed=feed,
source=feed.source,
html=entry['html'],
text=entry['text'],
tags=entry['tags'],
title=entry['title'],
created_at = parse(entry['published']),
updated_at = parse(entry['updated']),
image=random.choice(sample_images), # fake image
score=random.random() * 100 # fake score
)
articles.append(a)
db.session.add(a)
progress_bar(len(articles) / len(entries) * 100)
print('Creating additional articles...')
# Collect all appropriate files.
all_files = []
for dir, subdir, files in os.walk('manage/core/data/organized_articles'):
for file in files:
filepath = os.path.join(dir, file)
name, ext = os.path.splitext(filepath)
if ext == '.txt':
all_files.append((dir, name, filepath))
# Create articles for appropriate files.
for dir, name, filepath in all_files:
category = dir.split('/')[-1]
f = open(filepath, 'r')
article = Article(
text=f.read(),
title=name.split('/')[-1],
ext_url='http://fauxurl/',
source = Source.query.get(1), # fake source
image=random.choice(sample_images), # fake image
score=random.random() * 100 # fake score
)
db.session.add(article)
articles.append(article)
progress_bar(len(articles)/len(all_files) * 100)
db.session.commit()
num_articles = Article.query.count()
num_concepts = Concept.query.count()
print('Seeded {0} articles.'.format(num_articles))
print('Found {0} concepts.'.format(num_concepts))
print('Clustering articles into events...')
cluster.cluster(articles)
num_events = Event.query.count()
print('Created {0} event clusters.'.format(num_events))
print('Clustering events into stories...')
# TO DO
num_stories = Story.query.count()
print('Created {0} story clusters.'.format(num_stories))
patcher.stop()
print('\n\n==============================================')
print('From {0} sources, seeded {1} articles, found {2} concepts, created {3} events and {4} stories.'.format(num_sources, num_articles, num_concepts, num_events, num_stories))
print('==============================================\n\n')
client = current_app.test_client()
ctx = current_app.test_request_context()
ctx.push()
register_user(email='[email protected]', password='******')
ctx.pop()
print('\n\n==============================================')
print('Created a test user, email is [email protected], password is password')
print('==============================================\n\n')
client = Client(
#client_id=gen_salt(40),
#client_secret=gen_salt(50),
client_id='test',
client_secret='test',
_redirect_uris='http://localhost:5000/authorized',
_default_scopes='userinfo',
_allowed_grant_types='authorization_code refresh_token password',
user_id=None,
is_confidential=True # make a confidential client.
)
db.session.add(client)
db.session.commit()
print('\n\n==============================================')
print('Created a test client:\nclient id: {0}\nclient secret: {1}'.format(client.client_id, client.client_secret))
print('==============================================\n\n')
def download(url, save_path, filename=None, progress=False):
"""
Downloads a file from the specified URL.
Will resume an existing download if the target
server supports it (responds with the "Accepts-Range" header).
Args:
| url (str) -- url of the file to download
| save_path (str) -- path to the directory to save the file
| progress (bool) -- output progress bar to stdout
"""
# Strip trailing slash, if there is one.
save_path = save_path.rstrip('\/')
if filename is None:
filename = url.split('/').pop()
file = '{0}/{1}'.format(save_path, filename)
existing_size = 0
# If file already exists,
# but there is not a newer file is on the server...
if os.path.exists(file) and not _expired(url, file):
# Append to existing file.
outfile = open(file, 'ab')
# Figure out how many bytes we've got.
existing_size = outfile.tell()
# Setup request for only the remaining bytes.
headers = {'Range': 'bytes={0}-'.format(existing_size)}
req = request.Request(url, headers=headers)
# Otherwise, create a new/overwrite existing file.
else:
# Create/overwrite file.
outfile = open(file, 'wb')
outfile.seek(0)
# Vanilla request.
req = request.Request(url)
try:
# Get response.
resp = request.urlopen(req)
# Get total size of content.
total_size = float(resp.headers['Content-Length'].strip())
# Check if the file has already been downloaded_size.
if total_size == existing_size:
logger.info('File already downloaded.')
return
# Check that the server accepts ranges.
# If it does not, the server will ignore the Range header,
# And we have to start all over again.
if existing_size > 0 and not resp.headers.get('Accept-Ranges', None):
logger.info('Server does not allow resuming of downloads.')
logger.info('Starting from the beginning! :D')
outfile = open(file, 'wb')
outfile.seek(0)
if progress:
progress_bar( (existing_size/total_size) * 100 )
# Pull out the chunks!
for chunk in iter(lambda: resp.read(CHUNK), b''):
# Write the chunk to the file.
outfile.write(chunk)
# Update existing size.
existing_size += len(chunk)
percent_complete = (existing_size/total_size) * 100
# Show progress.
if progress:
progress_bar(percent_complete)
if progress:
sys.stdout.write('\n')
# Return the download's filepath.
return file
except request.HTTPError as e:
logger.error('HTTP Error:', e.code, url)
except request.URLError as e:
logger.error('URL Error:', e.reason, url)
def run(self, dumppath, use_patch):
if use_patch:
print('Patching out saving images to S3...')
patcher = patch('argos.util.storage.save_from_url', autospec=True, return_value='https://i.imgur.com/Zf9mXlj.jpg')
patcher.start()
else:
patcher = None
print('Loading sources...')
sources_map = {}
with open(os.path.join(dumppath, 'sources.json'), 'r') as f:
sources = json.load(f)
for i, s in enumerate(sources):
source = Source.query.filter(Source.name == s['name']).first()
if not source:
source = Source(name=s['name'])
db.session.add(source)
id = s['_id']['$oid']
sources_map[id] = source
progress_bar(i/(len(sources) - 1) * 100)
db.session.commit()
print('\nLoading feeds...')
feeds_map = {}
with open(os.path.join(dumppath, 'feeds.json'), 'r') as f:
feeds = json.load(f)
for i, f in enumerate(feeds):
feed = Feed.query.filter(Feed.ext_url == f['ext_url']).first()
if not feed:
feed = Feed(ext_url=f['ext_url'])
db.session.add(feed)
feed.source = sources_map[f['source']['$oid']]
id = f['_id']['$oid']
feeds_map[id] = feed
progress_bar(i/(len(feeds) - 1) * 100)
db.session.commit()
print('\nLoading articles...')
with open(os.path.join(dumppath, 'articles.json'), 'r') as f:
articles = json.load(f)
for i, a in enumerate(articles):
authors = []
for author in a['authors']:
authors.append(Author.find_or_create(name=author))
existing = Article.query.filter(Article.ext_url == a['ext_url']).first()
if not existing:
feed = feeds_map[a['feed']['$oid']]
article = Article(
ext_url=a['ext_url'],
source=feed.source,
feed=feed,
html=None, # not saved by argos.corpora
text=fix_text_segment(a['text']),
authors=authors,
tags=[],
title=fix_text_segment(a['title']),
created_at=datetime.fromtimestamp(a['created_at']['$date']/1000),
updated_at=datetime.fromtimestamp(a['updated_at']['$date']/1000),
image=a['image'],
score=evaluator.score(a['ext_url'])
)
db.session.add(article)
progress_bar(i/(len(articles) - 1) * 100)
print('Loaded {0} sources, {1} feeds, and {2} articles.'.format(len(sources), len(feeds), len(articles)))
print('Done!')
if patcher is not None:
patcher.stop()
