NumPy 学习笔记（四）

import numpy as np

arr = np.array([[1., 2., 3.], [4., 5., 6.], [7., 8., 9.]])
print("reciprocal arr: ", np.reciprocal(arr))
print("power by 2: ", np.power(arr, 2))
print("mod by arr.T: ", np.mod(arr, arr.T))
print("remainder by arr.T: ", np.remainder(arr, arr.T))

import numpy as np

arr = np.array([[1., 2., 3.], [4., 5., 6.], [7., 8., 9.]])

# numpy.amax(a, axis=None, out=None, keepdims=<no value>, initial=<no value>)
# numpy.amin(a, axis=None, out=None, keepdims=<no value>, initial=<no value>)
# numpy.max 和 numpy.min 的格式和上面一样
print("max: ", np.amax(arr, axis=1))
print("max: ", np.max(arr))
print("min: ", np.amin(arr))
print("min: ", np.min(arr, axis=1))

# numpy.ptp(a, axis=None, out=None, keepdims=<no value>) 计算数组中元素最大值与最小值的差（最大值 - 最小值）
print("ptp(arr, axis=0): ", np.ptp(arr, axis=0))
print("ptp(arr, axis=1): ", np.ptp(arr, axis=1))
print("ptp(arr): ", np.ptp(arr))

# numpy.percentile(a, q, axis)  a 表示一个对象，q 表示要计算的百分数（0-100），axis 是轴
# 选出在数组 arr 里大于等于 50% 数的那个数
print("percentile(arr, 50): ", np.percentile(arr, 50))      # 5.0
# 按 1 轴选出大于等于 0% 的数的那个数，即最小值
# [1. 4. 7.]
print("percentile(arr, 0, axis=1): ", np.percentile(arr, 0, axis=1))
# 按 1 轴选出大于等于 100% 的数的那个数，即最大值
# [3. 6. 9.]
print("percentile(arr, 100, axis=1): ", np.percentile(arr, 100, axis=1))
# 若所求百分数不能刚好对应数组里的数，则会在前后两个数之间取平均值
print(np.percentile(np.array([1., 2., 3., 4., 5., 6.]), 50))    # 3.5
print(np.percentile(np.array([1., 2., 3., 4., 5., 6.]), 70))    # 4.5

# numpy.median(a, axis=None, out=None, overwrite_input=False, keepdims=False) 计算 a 的中位数
print("median(arr, axis=0): ", np.median(arr, axis=0))  # [4. 5. 6.]
print("median(arr, axis=1): ", np.median(arr, axis=1))  # [2. 5. 8.]

# numpy.mean(a, axis=None, dtype=None, out=None, keepdims=<no value>)
# 返回数组中元素的算术平均值。 如果提供了轴，则沿其计算
print("mean(arr): ", np.mean(arr))  # 5.0
print("mean(arr, axis=1): ", np.mean(arr, axis=1))  # [2. 5. 8.]

# numpy.average(a, axis=None, weights=None, returned=False)
# 根据 weights 中给出的各自的权重计算数组中元素的加权平均值
# returned=True 则返回权重的和
# 当 weights.shape != a.shape 时必须指定轴
print("average: ", np.average(arr, axis=1, weights=np.array([1, 2, 3]), returned=True))

# numpy.std(a, axis=None, dtype=None, out=None, ddof=0, keepdims=<no value>) 计算标准差
print("std(arr): ", np.std(arr))    # 2.581988897471611
print("std(arr, axis=1): ", np.std(arr, axis=1))    # [0.81649658 0.81649658 0.81649658]

# numpy.var(a, axis=None, dtype=None, out=None, ddof=0, keepdims=<no value>) 计算方差
print("var(arr): ", np.var(arr))    # 6.666666666666667
print("var(arr, axis=0): ", np.var(arr, axis=0))    # [6. 6. 6.]