python - 使用 pandas to_csv 时删除标题中的前导逗号

Question

默认情况下 to_csv 写一个 CSV 像

,a,b,c
0,0.0,0.0,0.0
1,0.0,0.0,0.0
2,0.0,0.0,0.0

但我希望它这样写:

a,b,c
0,0.0,0.0,0.0
1,0.0,0.0,0.0
2,0.0,0.0,0.0

我如何实现这一目标？我无法设置 index=False因为我想保留索引。我只想删除前导逗号。

df = pd.DataFrame(np.zeros((3,3)), columns = ['a','b','c'])
df.to_csv("test.csv") # this results in the first example above.

Answer 1

在追加模式下，可以先只写入没有索引的列，然后再写入没有标题的数据:

df = pd.DataFrame(np.zeros((3,3)), columns = ['a','b','c'], index=list('XYZ'))

pd.DataFrame(columns=df.columns).to_csv("test.csv", index=False)
#alternative for empty df
#df.iloc[:0].to_csv("test.csv", index=False)
df.to_csv("test.csv", header=None, mode='a')

df = pd.read_csv("test.csv")
print (df)
     a    b    c
X  0.0  0.0  0.0
Y  0.0  0.0  0.0
Z  0.0  0.0  0.0