read_csv没有在此文件上正确读取列名？

Question

我有一个csv文件如下:

0 5
1 10
2 15
3 20
4 25

我想将它保存为一个数据框,其中x,y轴作为名称,然后绘制它.然而,当我分配时x,y我得到一个混乱的DataFrame,发生了什么？

column_names = ['x','y']
x = pd.read_csv('csv-file.csv', header = None, names = column_names)
print(x)

          x   y
0   0 5 NaN
1  1 10 NaN
2  2 15 NaN
3  3 20 NaN
4  4 25 NaN

我试过不指定None的header,都无济于事.

Answer 1

添加参数sep="\s+"或delim_whitespace=True到read_csv:

import pandas as pd

temp=u"""0 5
1 10
2 15
3 20
4 25"""
#after testing replace io.StringIO(temp) to filename
column_names = ['x','y']
df = pd.read_csv(pd.compat.StringIO(temp), sep="\s+", header = None, names = column_names)

print (df)
   x   y
0  0   5
1  1  10
2  2  15
3  3  20
4  4  25

要么:

column_names = ['x','y']
df = pd.read_csv(pd.compat.StringIO(temp),
                 delim_whitespace=True, 
                 header = None, 
                 names = column_names)

print (df)
   x   y
0  0   5
1  1  10
2  2  15
3  3  20
4  4  25

Answer 2

你可以试试这个:

import pandas as pd    

column_names = ['x','y']
df = pd.read_csv('csv-file.csv',header=None)
df.columns = column_names