def Run(setLink):
global dayTimeStart
dayTimeStart = Tools.InitTimeSet()
for i in range(1, 5):
GenMergeIC(setLink + DLSet.link_Feature_I % i,
setLink + DLSet.link_dfUIC_Label % i,
setLink + DLSet.link_Feature_IC % i)
def Main():
global dayTimeStart
dayTimeStart = Tools.InitTimeSet()
print('start')
str = 'OrgSet'
CalCRTNum(DLSet.link_ChoiceSet % str + DLSet.link_orgData)
"""
def CutData(dataLink, storeLink):
global dayTimeStart
batch = 0
for df in pd.read_csv(open(dataLink, 'r'),
header=None,
names=DLSet.orgDataHead,
dtype=DLSet.orgDataType,
chunksize=100000):
try:
# 初始化数据集合
daySet = {}
for i in range(9):
daySet[i] = []
# 划分至不同日期集合里
for index, row in df.iterrows():
idx = Tools.UpperBound(dayTimeStart, row['timeStamp']) - 1
if idx < 0 or idx > 8:
continue
aList = []
for i in range(5):
aList.append(row[i])
daySet[idx].append(aList)
# 存储至各个文件之中
for i in range(9):
with open(storeLink % (i + 1), 'a') as f:
for each in daySet[i]:
for k in range(4):
f.write(str(each[k]) + ',')
f.write(str(each[4]) + '\n')
batch += 1
print('chunk %d done.' % batch)
# break
except StopIteration:
print("finish data process")
break
def Run(setLink):
global dayTimeStart
dayTimeStart = Tools.InitTimeSet()
for i in range(1, 5):
GenMergeC(setLink + DLSet.link_dayX2Y,
setLink + DLSet.link_Feature_C % i, i)
def Run(setLink):
global dayTimeStart
dayTimeStart = Tools.InitTimeSet()
CutData(setLink + DLSet.link_orgData, setLink + DLSet.link_dayN)