dataframe中的数据类型及转化

2021-09-04 23:50:00

1 float与str的互化

import pandas as pd
import numpy as np

df = pd.DataFrame({\'a\':[1.22, 4.33],
                   \'b\':[3.44, 5.66]})
# 将float类型转为str
# 法一
df[\'a\'] = df[\'a\'].apply(lambda x: str(x))
print(type(df[\'a\'].values[1]))
# 法二
df[\'b\'] = df[\'b\'].astype(str)
print(type(df[\'b\'].values[1]))
# str转float同上
df[\'a\'] = df[\'a\'].apply(lambda x: float(x))
print(type(df[\'a\'].values[1]))
df[\'b\'] = df[\'b\'].astype(float)
print(type(df[\'b\'].values[1]))
# <class \'str\'>
# <class \'str\'>
# <class \'numpy.float64\'>
# <class \'numpy.float64\'>

View Code

２建表时，如果赋的是空值，默认是float类型，所以要指定数据类型。

特别是建了空表后，与其它表融合时，要注意数据类型的改变。

df = pd.DataFrame({ \'a\': [],
                    \'b\': [],
                  })
print(df.info())
df = pd.DataFrame({ \'a\': [],
                    \'b\': [],
                 },         dtype = np.int64)
print(df.info())
# <class \'pandas.core.frame.DataFrame\'>
# RangeIndex: 0 entries
# Data columns (total 2 columns):
# a    0 non-null float64
# b    0 non-null float64
# dtypes: float64(2)
# memory usage: 76.0 bytes
# None
# <class \'pandas.core.frame.DataFrame\'>
# RangeIndex: 0 entries
# Data columns (total 2 columns):
# a    0 non-null int64
# b    0 non-null int64
# dtypes: int64(2)
# memory usage: 76.0 bytes
# None

参考：https://blog.csdn.net/jinruoyanxu/article/details/79150896