def get_standard_data_for_cloumn(filename, header):
f = FileProcess()
d = f.file_get_data(filename, header)
try:
d = deal_with_NA(d)
d = list(map(int, d))
except:
d = list(map(float, d))
max_d = max(d)
min_d = min(d)
if max_d == min(d):
return d
#if header not in label_list:
# return d
if header == 'Creditability':
return d
for i in range(len(d)):
#if d[i] == -1:
#d[i] = 99
x = len(str(d[i]))
d[i] = d[i] / (math.pow(10, x))
d[i] = round(d[i], 1)
#d[i] = int((d[i] - min_d)*(9/(max_d - min_d)))
return d
def data_get():
f = FileProcess()
data_0 = []
data_1 = []
headers = []
headers, data = f.file_get_data_row(filename)
for line in data:
if line[CLASS] == '0':
data_0.append(line[1:])
else:
data_1.append(line[1:])
## class 0 to class 1 ratio
k = len(data_0)/len(data_1)
print(k)
return k, data_0, data_1,headers
def data_stard_write():
data_all = []
f = FileProcess()
headers = f.file_get_headers(filename)
for header in headers:
d = get_standard_data_for_cloumn(filename, header)
data_all.append(d)
data_all = data_transpose(data_all)
data_all = data_sort(data_all)
data_train, data_val, data_test = get_train_val_test(data_all)
f.write_csv('data_train.csv', headers, data_train)
f.write_csv('data_val.csv', headers, data_val)
f.write_csv('data_test.csv', headers, data_test)
def get_standard_data_for_cloumn(filename, header):
f = FileProcess()
d = f.file_get_data(filename, header)
d = list(map(int, d))
max_d = max(d)
min_d = min(d)
#print(d)
if max_d == min_d:
return d
if header == 'class':
return d
for i in range(len(d)):
d[i] = int((d[i] - min_d) * (9 / (max_d - min_d)))
#print(d)
#print(max_d, min_d)
return d
def get_standard_data_for_cloumn(filename, header):
f = FileProcess()
d = f.file_get_data(filename, header)
try:
d = deal_with_NA(d)
d = list(map(int, d))
except:
d = list(map(float, d))
max_d = max(d)
min_d = min(d)
#print(d)
if max_d == min_d:
return d
if header == 'SeriousDlqin2yrs':
return d
for i in range(len(d)):
d[i] = int((d[i] - min_d) * (99 / (max_d - min_d)))
return d
def get_standard_data_for_cloumn(filename, header):
f = FileProcess()
d = f.file_get_data(filename, header)
try:
d = deal_with_NA(d)
d = list(map(int, d)) ##list to int
except:
d = list(map(float, d)) ##list to int
max_d = max(d)
min_d = min(d)
if max_d == min(d):
return d
if header == 'acc_now_delinq':
return d
for i in range(len(d)):
#if d[i] == -1:
#d[i] = 99
d[i] = int((d[i] - min_d)*(9/(max_d - min_d)))
return d
print(data)
s=Smote(data,N=100)
s = (s.over_sampling())
s = s.tolist()
smote = []
for line in s:
l = [0] + line
for i in range(1, len(line)+1):
l[i] = int(line[i-1] + 0.5)
print(l)
smote.append(l)
f = FileProcess()
headers, d = f.file_get_data_row(filename)
smote_data = smote + d
f = FileProcess()
f.write_csv('smote_data.csv', headers, smote_data)
'''
a = []
for d in data_tmp:
a.append(list(map(int,d)))