def locations(rack_locations_path=RACKS_LOCATION_CSV):
with open(rack_locations_path, 'r') as file:
csv_file = DictReader(file,
["latitude", "longitude", "icon", "desc", "racks_count", "parking_places"])
acc = []
csv_file.__next__() # Skip the header
for attributes in csv_file:
acc.append(RacksLocation(attributes))
return acc
def __next__(self):
obj = DictReader.__next__(self)
if self.schema:
data, errors = self.schema.load(obj)
if errors:
raise ValidationError(errors, data=data)
obj = data
return obj
def parse_csv(csv_path: str) -> List[BfoUser]:
csv_file = None
try:
csv_file = open(csv_path, "r", newline='')
except Exception:
print(f"Could not open csv file at path {csv_path}")
exit(5)
users: List[BfoUser] = []
reader = DictReader(csv_file, list(BfoUser.__slots__))
reader.__next__() # Throw out header
for user_dict in reader:
users += [
BfoUser(user_dict['name'], user_dict['description'],
user_dict['underneath'], user_dict['url'])
]
csv_file.close()
return users
def test_research_study_title(study, transformed_dir):
specimens = transformed_dir / 'specimen.tsv'
assert specimens.exists(), "Specimen file exists?"
with specimens.open() as f:
reader = DictReader(f, delimiter='\t')
# The way the transformation works, all of the specimen in a given file will have
# the same study name
line = reader.__next__()
assert line['STUDY|NAME'] == study.title, "Is the title correct?"
def f(path):
label_path = path + 'validation.csv'
predict_path = path + 'submission.csv'
label_reader = DictReader(open(label_path))
predict_reader = DictReader(open(predict_path))
y_true = []
y_scores = []
for t, row in enumerate(label_reader):
predict = predict_reader.__next__()
actual = float(row['label'])
predicted = float(predict['prob'])
y_true.append(actual)
y_scores.append(predicted)
# Compute ROC curve and ROC area
fpr, tpr, threshold = roc_curve(y_true, y_scores, pos_label=1)
roc_auc = auc(fpr, tpr)
return fpr, tpr, roc_auc
def load_authenication(data_path):
"""Loads a Couchbase username and password
Args:
data_path: String that stores the full path to folder that contains
the Austin Tx Unified 311 open dataset
Returns:
authenticator: Dictionary that stores a Couchbase username and
password"""
password_file = os.path.join(data_path, '.password', 'password.txt')
with open(password_file, 'rt') as h_file:
keys = h_file.readline().rstrip().split(',')
readerobj = DictReader(h_file, keys)
authentication = readerobj.__next__()
return authentication
def test_three_does_not_stomp(self):
logging.getLogger().setLevel(logging.CRITICAL)
self.manager._migrate_one()
self.manager._migrate_two()
f = PathManager.open_input_file('input.csv', 'w+')
f.write('foo\nspecial')
f.close()
self.manager._migrate_three()
f = PathManager.open_input_file('input.csv.bak', 'r')
contents = f.read()
self.assertEqual('foo\nspecial', contents)
f.close()
f = PathManager.open_input_file('input.csv', 'r')
dict_reader = DictReader(f)
first_row = dict_reader.__next__()
self.assertTrue('foo' in first_row.keys())
self.assertEqual(first_row['foo'], 'special')
f.close()
def read_log(filename):
"""
Read a logfile from jiminy. This function supports both text (csv)
and binary log.
Parameters:
- filename: Name of the file to load.
Retunrs:
- A dictionnary containing the logged values, and a dictionnary
containing the constants.
"""
if is_log_binary(filename):
# Read binary file using C++ parser.
data_dict, constants_dict = Engine.read_log_binary(filename)
else:
# Read text csv file.
constants_dict = {}
with open(filename, 'r') as log:
constants_str = next(log).split(', ')
for c in constants_str:
c_split = c.split('=')
# Remove line end for last constant.
constants_dict[c_split[0]] = c_split[1].strip('\n')
# Read data from the log file, skipping the first line (the constants).
data = {}
reader = DictReader(log)
for key, value in reader.__next__().items():
data[key] = [value]
for row in reader:
for key, value in row.items():
data[key].append(value)
for key, value in data.items():
data[key] = np.array(value, dtype=np.float64)
# Convert every element to array to provide same API as the C++ parser,
# removing spaces present before the keys.
data_dict = {k.strip(): np.array(v) for k, v in data.items()}
return data_dict, constants_dict
def __next__(self): # For Python 3
row = DictReader.__next__(self)
for (att, func) in self._casts.items():
row[att] = func(row[att])
return row
data_path = sys.argv[1]
result_path = sys.argv[2]
label_path = data_path + 'validation.csv'
predict_path = result_path + 'submission.csv'
label_reader = DictReader(open(label_path))
predict_reader = DictReader(open(predict_path))
count = 0
y_true = []
y_pred = []
y_scores = []
for t, row in enumerate(label_reader):
predict = predict_reader.__next__()
actual = float(row['Label'])
predicted = float(predict['Predicted'])
y_true.append(actual)
y_scores.append(predicted)
# 大于阈值的即视为点击
if (predicted >= 0.5):
y_pred.append(1)
else:
y_pred.append(0)
count += 1
# 计算性能指标
auc = roc_auc_score(y_true, y_scores)
threshold = 0.5
label_path = path + 'validation.csv'
predict_path = path + 'submission.csv'
label_reader = DictReader(open(label_path))
predict_reader = DictReader(open(predict_path))
count = 0
y_true = []
y_pred = []
y_scores = []
for t, row in enumerate(label_reader):
predict = predict_reader.__next__()
actual = float(row['label'])
predicted = float(predict['prob'])
y_true.append(actual)
y_scores.append(predicted)
# 大于阈值的即视为点击
if predicted >= threshold:
y_pred.append(1)
else:
y_pred.append(0)
count += 1
# 计算性能指标
auc = roc_auc_score(y_true, y_scores)
logloss = log_loss(y_true, y_scores)
