如果 time() 大于 5:00，如何从 ndarray 中选取时间戳

Question

我有一个时间戳数组。如果时间戳代表 5:00 之后的时间，我需要选择时间戳。

示例代码为：

import numpy as np
from datetime import datetime

arr = np.array([1672713000, 1672716600, 1672720200, 1672723800, 1672727400, 1672731000])
threshold_time = datetime.strptime('5:00', '%H:%M').time()

new_arr = np.where(datetime.utcfromtimestamp(arr).time() > threshold_time)

但是，当我尝试以下代码时，我收到此错误：

TypeError: only integer scalar arrays can be converted to a scalar index

这种情况下如何正确查询ndarray？我无法使用 pandas 来解决这个问题。

Answer 1

import numpy as np
from datetime import datetime

arr = np.array([1672713000, 1672716600, 1672720200, 1672723800, 1672727400, 1672731000])
threshold_time = datetime.strptime('5:00', '%H:%M').time()
func = np.vectorize(datetime.utcfromtimestamp)

# Create a boolean mask for accurate filtering
mask = (func(arr) > datetime.combine(datetime.today(), threshold_time)).astype(bool)

# Use the boolean mask to filter the array
new_arr = arr[mask]

print(new_arr)  # This will output the filtered timestamps

Answer 2

这样的解决方案对我有用：

import numpy as np
from datetime import datetime

arr = np.array([1672713000, 1672716600, 1672720200, 1672723800, 1672727400, 1672731000])
threshold_time = datetime.strptime('5:00', '%H:%M').time()

pick_time = np.vectorize(lambda x: datetime.utcfromtimestamp(x).time())

new_arr = np.where(pick_time(arr) > threshold_time)

print(arr[new_arr])

输出为：

[1672723800 1672727400 1672731000]