Question

我有一个列表，我想迭代并在pandas数据框中创建元组，以模仿大小为4的滑动窗口。我想要做的是：

tuples = pd.DataFrame()
for index, row in expertsDF.iterrows():
      newlst = row['name'] 
      counter = 0
      for x in newlst:
        if counter < len(newlst) - 3:
            tuples['A'] = x
            tuples['B'] =newlst(counter+1)
            tuples['C'] =newlst(counter+2)
            tuples['D'] =newlst(counter+3)
            counter = counter + 1

newlst看起来像这样：

list (var1, var2, var3....)

我的DataFrame应该是这样的：

     A       B      C       D
1   var1   var2   var3    var4
2   var2   var3   var4    var5
3   var3   var4   var5    var6

有没有办法在python中做到这一点？

Answer 1

这是否接近？

import pandas as pd
tuples = pd.DataFrame(columns=['A', 'B', 'C', 'D'])
newlst = "abcdefg"
i = 0
for x in newlst:
    if i < len(newlst) - 3:
        t = pd.DataFrame([[x, newlst[i + 1], newlst[i + 2], newlst[i + 3]]],
                         columns=['A', 'B', 'C', 'D'])
    tuples = tuples.append(t, ignore_index=True)
    i += 1
print tuples

打印：

   A  B  C  D
0  a  b  c  d
1  b  c  d  e
2  c  d  e  f
3  d  e  f  g

Answer 2

如何从Series对象创建DataFrame？

import pandas as pd

data_list = list(map(lambda x: 'var{}'.format(x),range(0,100)))
df = pd.Series(data_list[0:97]).to_frame(name='A')
df['B'] = pd.Series(data_list[1:98])
df['C'] = pd.Series(data_list[2:99])
df['D'] = pd.Series(data_list[3:100])

df.head()

# output
#       A     B     C     D
# 0  var0  var1  var2  var3
# 1  var1  var2  var3  var4
# 2  var2  var3  var4  var5
# 3  var3  var4  var5  var6
# 4  var4  var5  var6  var7

创建panda数据帧并在for循环中追加值

2 个答案: