如何在Python中按指定区间概率生成出勤模拟随机数？

Great question! You're right that numpy.random.choice is typically used for discrete values, but we can adapt it (along with other tools) to generate continuous data that fits your interval-based probability requirements. Let's walk through two approaches: one using NumPy (the most efficient for large datasets) and a pure Python standard library method.

方法1：使用NumPy实现（推荐） #

NumPy doesn't have a single built-in function to directly generate continuous data with interval probabilities, but we can combine numpy.random.choice (to pick intervals by probability) and numpy.random.uniform (to generate values within each selected interval) to get exactly what you need.

Here's a complete example:

import numpy as np

# 定义区间和对应的概率
attendance_intervals = [(70, 100), (40, 60), (0, 40)]
interval_probabilities = [0.6, 0.25, 0.15]

# 要生成的学生出勤记录数量
num_students = 1000

# 步骤1：按概率随机选择每个学生所属的区间
selected_indices = np.random.choice(len(attendance_intervals), size=num_students, p=interval_probabilities)

# 步骤2：为每个学生在选中的区间内生成随机出勤值
attendance_data = np.array([
    np.random.uniform(*attendance_intervals[idx]) 
    for idx in selected_indices
])

# 可选：如果需要整数出勤值，转换为int类型
attendance_data = attendance_data.astype(int)

验证结果是否符合概率要求 #

为了确认输出符合预期的分布，你可以统计每个区间内的样本占比：

# 计算各区间的样本数
count_70_100 = np.sum((attendance_data >= 70) & (attendance_data <= 100))
count_40_60 = np.sum((attendance_data >= 40) & (attendance_data < 70))
count_0_40 = np.sum((attendance_data >= 0) & (attendance_data < 40))

# 打印占比
print(f"70-100区间占比: {count_70_100 / num_students:.2%}")
print(f"40-60区间占比: {count_40_60 / num_students:.2%}")
print(f"0-40区间占比: {count_0_40 / num_students:.2%}")

方法2：纯Python标准库实现（无需NumPy） #

如果你不想使用NumPy，用random模块也能实现同样的效果，逻辑和上面一致：先按权重选区间，再在区间内生成随机值。

import random

attendance_intervals = [(70, 100), (40, 60), (0, 40)]
interval_probabilities = [0.6, 0.25, 0.15]

num_students = 1000
attendance_data = []

for _ in range(num_students):
    # 按指定概率选择区间
    low, high = random.choices(attendance_intervals, weights=interval_probabilities)[0]
    # 在区间内生成随机值
    attendance = random.uniform(low, high)
    # 转换为整数后加入列表（可选）
    attendance_data.append(int(attendance))

关键说明 #

• 不管是Python标准库还是NumPy，都没有直接生成带区间概率的连续随机数的单一内置函数，但上面的两步法简单高效，完全能满足需求。
• 用numpy.random.choice选择区间的方式，在处理大样本（比如1万+条记录）时效率更高；纯Python循环在样本量极大时会稍慢一些。
• 如果你需要区间内的非均匀分布（比如70-100区间内更多值集中在90左右），可以把uniform换成其他分布函数，比如numpy.random.normal（调整参数使其落在目标区间内）。

内容的提问来源于stack exchange，提问作者Sagarika Nangia