Python列表和数组的深入解析与最佳选择指南

my_list = [42, "Python", 3.14, True, [1, 2, 3]]  # 整数、字符串、浮点数、布尔值、甚至另一个列表

# 初始化学生列表
students = [
    {"name": "Alice", "scores": [85, 90, 88]},
    {"name": "Bob", "scores": [78, 82, 80]},
    {"name": "Charlie", "scores": [92, 95, 89]}
]

# 添加新学生
students.append({"name": "David", "scores": [80, 85, 90]})

# 计算平均分
for student in students:
    avg_score = sum(student["scores"]) / len(student["scores"])
    print(f"{student['name']}的平均分: {avg_score:.2f}")

import array
# 创建一个整数数组
int_array = array.array('i', [1, 2, 3, 4, 5])  # 'i'表示整数类型

import numpy as np

# 创建NumPy数组
np_array = np.array([1, 2, 3, 4, 5], dtype=np.float32)

# 向量化运算
squared = np_array ** 2  # 对每个元素平方

import numpy as np
import matplotlib.pyplot as plt

# 生成正弦波信号
t = np.linspace(0, 1, 1000)  # 1秒内1000个点
frequency = 5  # 5Hz
signal = np.sin(2 * np.pi * frequency * t)

# 添加噪声
noise = np.random.normal(0, 0.1, 1000)
noisy_signal = signal + noise

# 绘制信号
plt.plot(t, noisy_signal)
plt.title("带噪声的正弦波信号")
plt.xlabel("时间(s)")
plt.ylabel("幅度")
plt.show()

import timeit
import array
import numpy as np

# 准备数据
list_data = [float(i) for i in range(1000000)]
arr_data = array.array('d', list_data)
np_data = np.array(list_data)

# 测试平方运算
def test_list():
    return [x**2 for x in list_data]

def test_array():
    return array.array('d', [x**2 for x in arr_data])

def test_numpy():
    return np_data ** 2

# 计时
print("列表耗时:", timeit.timeit(test_list, number=10))
print("数组耗时:", timeit.timeit(test_array, number=10))
print("NumPy耗时:", timeit.timeit(test_numpy, number=10))

import numpy as np

# 使用列表收集异构数据
raw_data = [
    {"sensor": "A", "readings": [23.4, 24.1, 22.9]},
    {"sensor": "B", "readings": [19.8, 20.2, 21.0]}
]

# 转换为NumPy数组进行批量计算
all_readings = np.array([item["readings"] for item in raw_data])
mean_readings = np.mean(all_readings, axis=1)

# 将结果存回字典
for i, item in enumerate(raw_data):
    item["mean"] = mean_readings[i]

# 使用最小够用的数据类型
small_ints = np.array([1, 2, 3], dtype=np.int8)  # 1字节/元素

big_array = np.random.rand(10000, 10000)
subset = big_array[100:200, 100:200]  # 创建视图，不复制数据

# 处理超大文件
mmap_arr = np.memmap('large_array.npy', dtype='float32', mode='r', shape=(1000000,))

import numpy as np
from multiprocessing import Pool

def process_chunk(data_chunk):
    return np.sum(data_chunk ** 2)

# 分块处理大数据
big_data = np.random.rand(10000000)
chunks = np.array_split(big_data, 8)  # 分为8块

with Pool(8) as p:
    results = p.map(process_chunk, chunks)

total = sum(results)