Python 随机数据生成器预期的 str 实例，numpy.datetime64 找到

Question

>您好，一直在尝试将具有随机日期的随机数据创建到csv文件中，但收到以下错误expected str instance, numpy.datetime64 found

数据生成器的代码

import pandas as pd
import numpy as np
import string
import random
def gen_random_email():
domains = [ "hotmail.com", "gmail.com", "aol.com", "mail.com" , "mail.kz", "yahoo.com"]
letters = string.ascii_letters +'.'*5

email = ''.join(np.random.choice(list(letters),10))+'@'+ np.random.choice(domains)
email = email.replace('.@', '@')
return email, "Email"
def gen_random_float():
num = np.random.random()*np.random.randint(2000)

decimal_points = np.random.randint(8)
num = int(num*10**(decimal_points))/10**decimal_points

return str(num), 'Float'
def gen_random_sentence():
nouns = ["puppy", "car", "rabbit", "girl", "monkey"]
verbs = ["runs", "hits", "jumps", "drives", "barfs"]
adv = ["crazily", "dutifully", "foolishly", "merrily", "occasionally"]
adj = ["adorable.", "clueless.", "dirty.", "odd.", "stupid."]
random_entry = lambda x: x[random.randrange(len(x))]

random_entry = " ".join([random_entry(nouns), random_entry(verbs), 
random_entry(adv), random_entry(adj)])
return random_entry, 'String'
def gen_random_int():
num = np.random.randint(1000000)

return str(num), 'Int'
def gen_random_date():
monthly_days = np.arange(0, 30)
base_date = np.datetime64('2020-01-01')
random_date = base_date + np.random.choice(monthly_days)
return random_date, 'Date'
def gen_dataset(filename, size=5000):
randomizers = [gen_random_email, gen_random_float, gen_random_int, gen_random_sentence,gen_random_date]
with open(filename, 'w') as file:
file.write("Text, Typen")
for _ in range(size):
file.write(",".join(random.choice(randomizers)())+"n") 
gen_dataset('dataaaa.csv')   

TypeError: sequence item 0: expected str instance, numpy.datetime64 found 

Answer 1

首先，捕获错误并查看导致错误的原因。

def gen_dataset(filename, size=5000):
randomizers = [gen_random_email, gen_random_float, gen_random_int, gen_random_sentence,gen_random_date]
with open(filename, 'w') as file:
file.write("Text, Typen")
for _ in range(size):
f = random.choice(randomizers)
result = f()
try:
file.write(",".join(result)+"n")
except TypeError:
print(result)
raise

<小时 />

>>>
(numpy.datetime64('2020-01-09'), 'Date')
Traceback (most recent call last):
File "C:pyProjectstmp.py", line 80, in <module>
gen_dataset('dataaaa.csv')
File "C:pyProjectstmp.py", line 75, in gen_dataset
file.write(",".join(result)+"n")
TypeError: sequence item 0: expected str instance, numpy.datetime64 found

嗯，我想知道join是否只除了字符串作为参数？

是的，来自文档：

如果可迭代对象中有任何非字符串值(包括字节对象(，则会引发 TypeError。

我想知道如何将 numpy datetime64 变成字符串。使用numpy datetime64 to string搜索是富有成效的：在python中将numpy.datetime64转换为字符串对象

这些工作

>>> q = gen_random_date()[0]
>>> q
numpy.datetime64('2020-01-27')
>>> np.datetime_as_string(q)
'2020-01-27'
>>> q.astype(str)
'2020-01-27'
>>>

然后只需修改try/except即可。

def gen_dataset(filename, size=5000):
randomizers = [gen_random_email, gen_random_float, gen_random_int, gen_random_sentence,gen_random_date]
with open(filename, 'w') as file:
file.write("Text, Typen")
for _ in range(size):
f = random.choice(randomizers)
a,b = f()
try:
q = ",".join([a,b,"n"])
except TypeError:
a = np.datetime_as_string(a)
q = ",".join([a,b,"n"])
file.write(q)

或者干脆抢先将第一项设为字符串。

def gen_dataset(filename, size=5000):
randomizers = [gen_random_email, gen_random_float, gen_random_int, gen_random_sentence,gen_random_date]
with open(filename, 'w') as file:
file.write("Text, Typen")
for _ in range(size):
f = random.choice(randomizers)
a,b = f()
q = ",".join([str(a),b,"n"])
file.write(q)