python numpy读取csv文件_numpy模块处理csv文件

数据：

X,Y,NAME,LABEL

120.7512427,30.75084798,嘉兴市,2

120.0830671,30.89524644,湖州市,80

120.574379,30.00700998,绍兴市,140

122.202972,29.98754304,舟山市,165

121.546246,29.87620299,宁波市,86

119.642848,29.08127199,金华市,184

118.869413,28.93892703,衢州市,72

121.416636,28.65889302,台州市,38

119.912514,28.455276,丽水市,49

120.695457,27.99819198,温州市,127

103.71468,27.341637,昭通市,23

简单读取

import numpy as np

path = 'city_aic.csv'

with open(path, encoding='utf-8') as f:

data = np.loadtxt(path, delimiter=',')

print(data[:5])

这样直接读取的结果是，报错： ValueError: could not convert string to float: ‘X’

字符串处理-str参数

原因是，默认情况下，数据被认为是float类型，因此，在上面读取csv文件第1行时，遇到’X’，尝试进行数据类型转换，转换失败报错。

经查可以使用str参数，让方法读取数据时，支持str类型。

import numpy as np

path = 'city_aic.csv'

with open(path, encoding='utf-8') as f:

data = np.loadtxt(path, dtype=str, delimiter=',')

print(data[:5])

#结果

[['X' 'Y' 'NAME' 'LABEL']

['120.7512427' '30.75084798' '嘉兴市' '2']

['120.0830671' '30.89524644' '湖州市' '80']

['120.574379' '30.00700998' '绍兴市' '140']

['122.202972' '29.98754304' '舟山市' '165']]

跳过首行 - skiprow = 1

import numpy as np

path = 'city_aic.csv'

with open(path, encoding='utf-8') as f: