def run(seed):
ans = sol_fn(seed, inp, log)
process(inp,
ans,
seed,
sc_fn,
args.testcase,
ignore=args.ignore,
force=args.force)
def run(seed):
ans = get_ans(seed)
process(inp,
ans,
seed,
sc_fn,
args.testcase,
ignore=args.ignore,
force=args.force)
def get_discography(self, artist, song_name):
"""
Return the list of albums and their track names.
Invalid albums (as decided by `is_value_invalid()`) are not included.
Song names are preprocessed using `process()`.
The result is a dictionary indexed by album name and sorted by release
date.
"""
query = self.sp.search(f'artist:{artist} track:{song_name}',
type='track')
artist_id = query['tracks']['items'][0]['artists'][0]['id']
query = self.sp.artist_albums(artist_id, album_type='album')
artist_albums = {}
while query:
for album in query['items']:
name = process(album['name'], key='album')
name = name.lower()
if is_value_invalid(name, key='album'):
continue
_set_release_date(album)
elem = dict(id=album['id'], release_date=album['release_date'])
artist_albums[name] = elem
query = self.sp.next(query)
sort = sorted(artist_albums.items(),
key=lambda x: x[1]['release_date'])
artist_albums = dict(sort)
album_ids = [album['id'] for album in artist_albums.values()]
for albums, ids in chunks(artist_albums.values(), album_ids, 20):
query = self.sp.albums(ids)
for album, response in zip(albums, query['albums']):
tracks = []
while response:
if 'tracks' in response:
response = response['tracks']
tracks.extend(
process(t['name'], key='name').lower()
for t in response['items'])
response = self.sp.next(response)
tracks = dict.fromkeys(tracks)
tracks.pop('Unknown', None)
album['tracks'] = list(tracks)
return {
k: v
for k, v in artist_albums.items() if v.get('tracks', None)
}
def _addrow(self, row, tablename):
c = self.conn.cursor()
(transcript_si, transcript, decode_si,
decode, conf, decode_time, callsrepath, acoustic_model,
date, time, milliseconds, grammarlevel, firstname, lastname,
oration_id, chain, store) = process(row)
# There might a better way. Breaking out of a function if the transcript column is empty.
# This files without transcription are skipped
if transcript in ['', '""', ' ', None, 'NULL', '(NULL)']:
print 'No transcription for %s'
print callsrepath
return False
# Insert row into database
# Check if row exists. If yes don't add. If not add
datarow_exist_status = self._check_if_row_exists(
time, milliseconds, firstname, lastname, grammarlevel, tablename)
if datarow_exist_status:
print 'The data row you are trying to add might already exist'
else:
print 'Adding data row to database'
sql_statement = '''INSERT INTO %s VALUES ("%s", "%s", "%s", "%s", %d, %d, "%s", "%s", %s, %s, %s, "%s", "%s", "%s", "%s", "%s", "%s", "%s")''' % (
tablename, transcript_si, transcript, decode_si,
decode, int(conf), int(
decode_time), callsrepath, acoustic_model,
date, time, milliseconds, grammarlevel, firstname, lastname,
oration_id, chain, store, row)
try:
c.execute(sql_statement)
except:
print 'Could not enter row to database'
def test_1():
from util import read_file_to_list
from util import process
data = read_file_to_list("testinput.txt")
terminate, acc = process(data)
assert terminate == False
assert acc == 5
def capture(self, image_name):
currentDir = os.getcwd()
img_path = "%s/%s.jpg" % (currentDir, image_name)
print('image path:' + img_path)
self.camera.capture(img_path, use_video_port=True)
image = util.process(img_path, assemble.get_height_each())
self.session_images.append(image)
def capture(self, image_name):
currentDir = os.getcwd()
img_path = "%s/%s.jpg" % (currentDir, image_name)
print("image path:" + img_path)
self.camera.capture(img_path, use_video_port=True)
image = util.process(img_path, assemble.get_height_each())
self.session_images.append(image)
def generate_victim_offender():
victims = []
offenders = []
final = []
file_path = util.find_all("people.json")
count = 0
with open(file_path, "r") as people_json:
people = json.load(people_json)
for person in people:
if person['Victim']:
victims.append(person['SSN'])
else:
offenders.append(person['SSN'])
count += 1
util.process(count)
# now lets make the connection between the two arrays.
print(f"length of victims: {len(victims)}")
print(f"length of offenders: {len(offenders)}")
# length of victims is always larger than offenders (percentages)
# therefore, we have a temp_offenders array that stores all deleted offenders
# in case we run out of offenders to pair with victimes, hence the if stmt
temp_offenders = []
for victim in victims:
length = len(offenders) - 1
if length < 0:
l = len(temp_offenders) - 1
num = random.randint(0, l)
offender = temp_offenders[num]
temp_offenders.remove(temp_offenders[num])
temp = [victim, offender]
final.append(temp)
else:
num = random.randint(0, length)
offender = offenders[num]
temp_offenders.append(offenders[num])
offenders.remove(offenders[num])
temp = [victim, offender]
final.append(temp)
util.output_SQL("VICTIM_OFFENDER", final,
"/mysql/insertVictimOffender.sql")
util.output_json(final, "victim_offender.json")
def test_2():
from util import read_file_to_list
from util import process
from util import fix_command
data = read_file_to_list("testinput.txt")
data1 = fix_command(data, 7)
terminate, acc = process(data1)
assert terminate == True
assert acc == 8
def __iter__(self):
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
tknzr = TweetTokenizer(preserve_case=False);
stop_words = set(stopwords.words("english"))
for paragraph in self.paragraph_iterator():
raw_sentences = tokenizer.tokenize(paragraph.decode('utf8').strip())
for raw_sentence in raw_sentences:
# process the sentence
sentence = util.process(raw_sentence, tknzr, self.process_option, stop_words)
yield sentence
def readCorpus(self,corpusName): print "Loading corpus" with open(corpusName,'r') as f: for tweet in f.readlines(): tweet = tweet.strip() #print tweet self.corpus.append(tweet) #print process(tweet) self.bagOfWords.append(process(tweet)) return self.bagOfWords
def generate_weapons():
file_name = util.find_all("weapons.csv")
with open(file_name) as csv_file:
csv_reader = csv.reader(csv_file, delimiter=',')
line_count = 0
weapons = []
for row in csv_reader:
if line_count == 0: # some glitch (weird UTF-8 symbol)
weapon = row[0][1:]
else:
weapon = row[0]
fatalities = row[1]
temp = {weapon: fatalities}
weapons.append(temp)
line_count += 1
util.process(line_count)
return weapons
def __iter__(self):
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
tknzr = TweetTokenizer(preserve_case=False)
stop_words = set(stopwords.words("english"))
for paragraph in self.paragraph_iterator():
raw_sentences = tokenizer.tokenize(
paragraph.decode('utf8').strip())
for raw_sentence in raw_sentences:
# process the sentence
sentence = util.process(raw_sentence, tknzr,
self.process_option, stop_words)
yield sentence
def fetch_album(self, song):
"""
Get the name of the album for a song from spotify.
"""
artist, title = song.artist, song.title
self.fetch_discography(song)
if not self.discography_cache.get(artist, ''):
return 'Unknown'
title = process(title, key='name')
for album_name, info in self.discography_cache[artist].items():
if title.lower() in map(str.lower, info['tracks']):
return album_name
return 'Unknown'
def run(): camera.warmup(width, height) images = [] for i in range(takes): pic_paths = camera.take(1, width, height) start = time.time() image1 = util.process(pic_paths[0], assemble.get_height_each()) end = time.time() diff = end - start images.append(image1) print "took in: " + str(diff) + "ms" if(diff < PAUSE_MAX and i < takes - 1): sleep_time = PAUSE_MAX - diff print "sleeping for another: " + str(sleep_time) + "ms" time.sleep(diff) assemble.assemble(images)
def run():
camera.warmup(width, height)
images = []
for i in range(takes):
pic_paths = camera.take(1, width, height)
start = time.time()
image1 = util.process(pic_paths[0], assemble.get_height_each())
end = time.time()
diff = end - start
images.append(image1)
print "took in: " + str(diff) + "ms"
if (diff < PAUSE_MAX and i < takes - 1):
sleep_time = PAUSE_MAX - diff
print "sleeping for another: " + str(sleep_time) + "ms"
time.sleep(diff)
assemble.assemble(images)
def get_album_tracks(self, song):
"""
Get the list of tracks of the album this song belongs to.
"""
if song.album and song.album != 'Unknown':
song.album = process(song.album, key='album', invalid=False)
self.fetch_discography(song)
else:
logger.debug('song has no embedded album. searching')
song.album = self.fetch_album(song)
logger.debug('got this album: %s', song.album)
try:
if not song.album or song.album == 'Unknown':
raise KeyError('Album not found')
return self.discography_cache[song.artist][song.album]['tracks']
except KeyError:
msg = 'Spotify could not find the list of tracks for %s'
logging.info(msg, song)
return []
if shuffle == 1:
shuffled_pairs = util.random_shuffle(training_pairs)
save_root = save_p + '/set%02d_%s' % (set_no, PHASE)
if HNM == 1:
save_root = save_root + '_hnm'
print 'saved to: ' + save_root
if os.path.exists(save_root):
shutil.rmtree(save_root)
os.makedirs(save_root)
X = np.zeros((2000, 6, 160, 80), dtype=np.float32)
map_size = X.nbytes * 1000
env = lmdb.open(save_root, map_size=map_size)
util.process(shuffled_pairs, M, N, transformer, env)
PHASE = 'test'
training_pairs = []
for setname in dataset_list:
if dataset_info_dict[setname]['Used'] == 1:
filename_part = dataset_info_dict[setname]['PartitionTest']
if not setname == 'MARKET':
main_dict = dataset_info_dict[setname]['Util'].collect_data(
dataset_info_dict[setname]['DatasetPath'])
this_list = dataset_info_dict[setname][
'Util'].partition_file_to_list(filename_part)
sub_dict = util.get_sub_dict(main_dict, this_list)
else:
main_dict = dataset_info_dict[setname]['Util'].collect_data(
dataset_info_dict[setname]['DatasetPath'] +
sp.load_data()
sp.train_df['X_train'] = list(sp.X_train)
sp.train_df['y_train'] = list(sp.y_train)
train_df = sp.train_df
n_fold = 4
train_df.sort_values('coverage_class', inplace=True)
train_df['fold'] = (list(range(n_fold)) *
train_df.shape[0])[:train_df.shape[0]]
subsets = [train_df[train_df['fold'] == i] for i in range(n_fold)]
for fold_idx in range(n_fold):
# get the train/val split
X_tr, X_val, y_tr, y_val = get_train_val(subsets, fold_idx)
# add depth information or triple the gray channel
X_tr, y_tr = process(X_tr, y_tr, use_depth=use_depth)
X_val, y_val = process(X_val, y_val, use_depth=use_depth)
y_tr = np.squeeze(y_tr)
y_val = np.squeeze(y_val)
# prepare PyTorch dataset and dataloader
train_ds = SaltDataset(X_tr, y_tr)
val_ds = SaltDataset(X_val, y_val, transform=None)
train_loader = torch.utils.data.DataLoader(train_ds,
batch_size=32,
shuffle=True)
val_loader = torch.utils.data.DataLoader(val_ds,
batch_size=32,
shuffle=True)
net = UNetResNet(34)
def run(solve_args):
solve_args_orig = dict(solve_args)
ans = get_ans(solve_args)
process(inp, ans, solve_args_orig, sc_fn)
def __iter__(self):
for document in self.field_iterator(setting.csv_option.review_name):
yield util.process(document)
def test_process():
from util import process
assert 73 == process('XXXXXXXXXXXXXXXXXXXXXXXXXXXXX1XXXX0X', 11)
def test_process():
from util import process
assert 0 == process([0, 3, 6], 10)
assert 1 == process([1, 3, 2], 2020)
def summarize(self):
for graph_id in self.dags.keys():
latency.process('%s/data/%s/dag' % (self._local_log_dir, graph_id),
graph_id, '%s/summary' % (self._local_log_dir))
util.process('%s/util' % (self._local_log_dir),
'%s/summary' % (self._local_log_dir))
def test_process(name, expect):
assert process(name, key='name') == expect
def run(solve_args):
solve_args_orig = dict(solve_args)
ans = get_ans(solve_args) # run the solver function and get the answer
process(inp, ans, solve_args_orig,
sc_fn) # add to best submission or not