def parse_input_data(cache_file, url, cookie_file):
"""
Calls the utility function to load either a cached data file or pull it from the AoC server, then splits it
on newlines, strips the lines and extracts the password rules.
:param cache_file: The file in which the input data is stored
:param url: The URL to the AoC page
:param cookie_file: The path to the cookie file (only used if the cache file isn't found
:return: The input data as a list of rows, where each row is (min_val, max_val, required_letter, password)
"""
raw_data = utils.load_input_data(cache_file, url, cookie_file)
def extract_info(line):
m = re.match(r'^(\d+)-(\d+)\s(\w+):\s([^\s]+)\s*$', line)
if m is None:
return None
min_val = int(m.group(1))
max_val = int(m.group(2))
req_letter = m.group(3)
password = m.group(4)
return (min_val, max_val, req_letter, password)
data = [extract_info(line) for line in raw_data.split("\n")]
return [d for d in data if d is not None]
def parse_input_data(cache_file, url, cookie_file):
"""
Calls the utility function to load either a cached data file or pull it from the AoC server, then splits it
on newlines, strips the lines and extracts the integer values.
:param cache_file: The file in which the input data is stored
:param url: The URL to the AoC page
:param cookie_file: The path to the cookie file (only used if the cache file isn't found
:return: The input data as a list of ints
"""
raw_data = utils.load_input_data(cache_file, url, cookie_file)
return [int(r.strip()) for r in raw_data.split("\n") if r.strip() != ""]
def get_x_y(features, label, data=None, reduce_classes=True, path=None):
config = load_global_config()
if data is None:
data = load_input_data(path)
try:
combined_label, combined_from = config['combined_label'], config[
'combined_from']
except KeyError:
X, y = extract_features_and_label(
data,
label,
features,
)
else:
if label == combined_label:
X, _ = extract_features_and_label(
data,
features=features,
)
class1 = data[combined_from[0]]
class2 = data[combined_from[1]]
y = lvddf_to_1_class(class1).combine(
class2, lambda val1, val2: val1 or val2)
else:
X, y = extract_features_and_label(
data,
label,
features,
)
try:
reduce_class_label = config['reduce_class_label']
except KeyError:
pass
else:
if label == reduce_class_label and reduce_classes:
y = lvddf_to_1_class(y)
return X, y
data = list(utils.parse_int_data(raw_data))
freqs = get_difference_frequency(data)
prod_1x3 = freqs[1] * freqs[3]
print("Joltage differences: ", freqs)
print(f"Product of 1 and 3 differences: {prod_1x3}")
return prod_1x3
def task_2(raw_data):
data = list(utils.parse_int_data(raw_data))
tgt = max(data) + 3
data += [tgt]
data.sort()
scores = defaultdict(int)
scores[0] = 1
for d in data:
scores[d] = scores[d - 3] + scores[d - 2] + scores[d - 1]
print(f"Total number of valid arrangements is {scores[data[-1]]}")
if __name__ == "__main__":
input_data = utils.load_input_data("cached_input.txt", INPUT_URL, '../session_cookie.txt')
print("Task 1:")
task_1(input_data)
print()
print("Task 2:")
task_2(input_data)
:param pattern_len: 패턴 매트릭스의 크기
:param text_len: 텍스트 매트릭스의 크기
:param patterns: 패턴 2차원 배열
:param text:: 텍스트 2차원 배열
:return: 텍스트 내 매칭되는 모든 매트릭스의 postion을 튜플 형식으로 반환
'''
ret = []
pattern_arr, text_arr = [list(v)
for v in patterns], [list(v) for v in text]
for i in range(text_len - pattern_len + 1):
for j in range(text_len - pattern_len + 1):
mask = [e[j:j + pattern_len]
for e in text_arr[i:i + pattern_len]] == pattern_arr
if mask:
ret.append((i + pattern_len - 1, j + pattern_len - 1))
return ret
if __name__ == "__main__":
data = load_input_data(args.input)
outputs = load_output_data(args.output)
check_result = checker_naive(*data)
print(check_result)
if check_result == outputs:
save_check_data(args.check, "yes")
else:
save_check_data(args.check, "no")
)
def format_columns(data: DataFrame) -> DataFrame:
data_new = data.copy()
data_new.columns = [column.upper() for column in data.columns]
return data_new
def select_features(data: DataFrame, features: Iterable[str]) -> DataFrame:
return data[[feature.upper() for feature in features]]
def impute_missing(data: DataFrame) -> DataFrame:
return DataFrame(SimpleImputer().fit_transform(data), columns=data.columns, index=data.index)
config = load_global_config()
data_all = load_input_data()
data_without_cardiac_events = data_all[data_all['HCAR2'] != 1]
datasets_all = get_datasets(data_all)
datasets_without_cardiac_events = get_datasets(data_without_cardiac_events)
datasets_to_report = [
(
'All—correlated >0.8 removed', 'all_without_correlation',
datasets_all['clustering_correlated_removed']
),
]
from utils import one_hot_encoder, load_input_data, \
build_classifier
# ----------------------
# Load input data
# ----------------------
# Class labels:
# antigen binder = 1, non-binder = 0
# Load non-binding sequences
ab_neg_files = [
'mHER_H3_1_Ab.txt', 'mHER_H3_1_AgN.txt', 'mHER_H3_2_Ab.txt',
'mHER_H3_2_AgN.txt', 'mHER_H3_3_Ab.txt', 'mHER_H3_3_AgN.txt'
]
mHER_H3_AgNeg = load_input_data(ab_neg_files, Ag_class=0)
# Load binding sequences
ab_pos_files = [
'mHER_H3_1_2Ag647.txt', 'mHER_H3_1_2Ag488.txt', 'mHER_H3_2_2Ag647.txt',
'mHER_H3_2_2Ag488.txt', 'mHER_H3_3_2Ag647.txt', 'mHER_H3_3_2Ag488.txt'
]
mHER_H3_AgPos = load_input_data(ab_pos_files, Ag_class=1)
# Fuse Ag positive and negative sequences
Ag_combined = pd.concat([mHER_H3_AgPos, mHER_H3_AgNeg])
Ag_combined = Ag_combined.drop_duplicates(subset='AASeq')
Ag_combined = Ag_combined.sample(frac=1).reset_index(drop=True)
# Save sequences and labels
X = Ag_combined.AASeq