def download_image(url, dest):
"""
Downloads a single image into provided destination.
:param url:
:param dest:
"""
logging.info('Downloading {} into {}'.format(url, dest))
dest = um.join_paths(dest, url.split('/')[-1])
response = requests.get(url)
if um.is_image_response(response):
with open(dest, 'wb') as f:
f.write(response.content)
def download_image(url, dest):
"""
Downloads a single image into provided destination.
:param url:
:param dest:
"""
logging.info('Downloading {} into {}'.format(url, dest))
dest = um.join_paths(dest, url.split('/')[-1])
response = requests.get(url)
if um.is_image_response(response):
with open(dest, 'wb') as f:
f.write(response.content)
def download_image(url, dest):
"""
Downloads a single image into provided destination.
:param url:
:param dest:
"""
logging.info('Downloading {} into {}'.format(url, dest))
dest = um.join_paths(dest, url.split('/')[-1])
response = requests.get(url)
if um.is_image_response(response):
with open(dest, 'wb') as f:
f.write(response.content)
if __name__ == '__main__':
parser = um.get_cli_parser()
parser.add_option("-f", "--file", dest="filename", default=um.join_paths(um.get_script_path(), 'links'),
help="data file with urls")
parser.add_option("-d", "--dest", dest="dest", default=um.join_paths(um.get_script_path(), 'images'),
help="destination folder")
opts = um.parse_cli_options(parser)
um.set_logging()
# create destination path if it doesn't exist
um.make_path(opts.dest)
with cnc.ProcessPoolExecutor(max_workers=mp.cpu_count()) as executor:
with open(opts.filename) as f:
for line in f:
executor.submit(download_image, line.replace('\n', ''), opts.dest)
from sklearn.grid_search import GridSearchCV
__author__ = "Denys Sobchyshak"
__email__ = "*****@*****.**"
# setting up constants
outlier_share = 0.1
validation_share = 0.1
# operational settings
ut.set_logging()
parser = ut.get_cli_parser()
parser.add_option("-f",
"--file",
dest="filename",
default=ut.join_paths(ut.get_script_path(), 'data',
'day.csv'),
help="data file with urls")
opts = ut.parse_cli_options(parser)
data_path = opts.filename
is_hour_data = 'hour' in data_path
# loading data
data = pd.read_csv(data_path,
parse_dates=[1],
index_col=1,
date_parser=lambda x: pd.datetime.strptime(x, "%Y-%m-%d"))
logging.info('Cleaning and normalizing the feature space')
# we have no need in an index feature
features = data.drop(['instant'], axis=1)
fields_to_encode = ['season', 'yr', 'mnth', 'weekday', 'weathersit']
:param dest:
"""
logging.info('Downloading {} into {}'.format(url, dest))
dest = um.join_paths(dest, url.split('/')[-1])
response = requests.get(url)
if um.is_image_response(response):
with open(dest, 'wb') as f:
f.write(response.content)
if __name__ == '__main__':
parser = um.get_cli_parser()
parser.add_option("-f",
"--file",
dest="filename",
default=um.join_paths(um.get_script_path(), 'links'),
help="data file with urls")
parser.add_option("-d",
"--dest",
dest="dest",
default=um.join_paths(um.get_script_path(), 'images'),
help="destination folder")
opts = um.parse_cli_options(parser)
um.set_logging()
# create destination path if it doesn't exist
um.make_path(opts.dest)
with cnc.ProcessPoolExecutor(max_workers=mp.cpu_count()) as executor:
with open(opts.filename) as f:
for line in f:
from sklearn import linear_model
from sklearn.grid_search import GridSearchCV
__author__ = "Denys Sobchyshak"
__email__ = "*****@*****.**"
# setting up constants
outlier_share = 0.1
validation_share = 0.1
# operational settings
ut.set_logging()
parser = ut.get_cli_parser()
parser.add_option("-f", "--file", dest="filename", default=ut.join_paths(ut.get_script_path(), 'data', 'day.csv'),
help="data file with urls")
opts = ut.parse_cli_options(parser)
data_path = opts.filename
is_hour_data = 'hour' in data_path
# loading data
data = pd.read_csv(data_path, parse_dates=[1], index_col=1, date_parser=lambda x: pd.datetime.strptime(x, "%Y-%m-%d"))
logging.info('Cleaning and normalizing the feature space')
# we have no need in an index feature
features = data.drop(['instant'], axis=1)
fields_to_encode = ['season', 'yr', 'mnth', 'weekday', 'weathersit']
if is_hour_data:
fields_to_encode.append('hr')
mt.hot_encode(features, fields_to_encode)
