230525

numpy/test1.py

@@ -0,0 +1,120 @@
+import numpy as np
+
+a1 = np.array([1, 2, 3, 4, 5])
+
+print(a1)
+
+a2 = np.zeros((2, 3), dtype=int)
+
+print(a2)
+
+print(a1.shape)
+print(a2.shape)
+
+a3 = np.ones((2, 4), dtype=int)
+
+print(a3)
+
+a4 = np.arange(1, 10)
+
+print(a4)
+
+a5 = np.linspace(1, 6, 4)
+
+print(a5)
+
+a6 = np.random.rand(2, 4)
+
+print("a6: " + a6.__str__())
+
+a7 = a5.astype(int)
+
+print(a7)
+
+a8 = np.linspace(1, 10, 4).astype(int)
+a9 = np.linspace(1, 6, 4).astype(int)
+
+print(a8)
+print(a9)
+
+a10 = np.dot(a8, a9)
+
+print(a10)
+
+a11 = np.zeros((2, 3), dtype=int)
+
+a11[0] = np.linspace(1, 10, 3).astype(int)
+a11[1] = np.linspace(1, 11, 3).astype(int)
+
+print(a11)
+
+# numpy数组可以直接进行线性变换 即广播
+a12 = (1 + (10 - 1) * np.random.rand(2, 2)).astype(int)
+a13 = (1 + (10 - 1) * np.random.rand(2, 2)).astype(int)
+
+print("a12: " + a12.__str__())
+print("a13: " + a13.__str__())
+
+# 矩阵乘法 写法等同
+a14 = a12 @ a13
+print(a14)
+
+a14 = np.dot(a12, a13)
+print(a14)
+
+a14 = np.matmul(a12, a13)
+print(a14)
+
+a15 = np.sqrt(a12)
+print(a15)
+
+a16 = np.sin(a12)
+a17 = np.cos(a12)
+
+print(a16)
+print(a17)
+
+a18 = np.log(a12)
+print(a18)
+
+a19 = np.power(a12, 2)
+print(a19)
+
+ax1 = (1 + (10 - 1) * np.random.rand(1, 7)).astype(int)
+print("ax1: " + ax1.__str__())
+
+a20 = np.max(np.abs(ax1))
+a21 = np.min(np.abs(ax1))
+print(a20)
+print(a21)
+
+a22 = np.argmax(np.abs(ax1))
+a23 = np.argmin(np.abs(ax1))
+print(a22)
+print(a23)
+
+a24 = np.sum(np.abs(ax1))
+print(a24)
+
+a25 = np.mean(np.abs(ax1))
+print(a25)
+
+a26 = np.median(np.abs(ax1))
+print(a26)
+
+# 方差
+a27 = ax1.var()
+print(a27)
+
+a28 = ax1.std()
+print(a28)
+
+a29 = ax1[(ax1 > 3) & (ax1 % 2 == 0)]
+print(a29)
+
+a30 = ax1[0, 0:6:2]
+print(a30)
+
+# 反转数组
+a31 = ax1[::-1]
+print(a31)

numpy/test2.py

@@ -0,0 +1,26 @@
+import numpy as np
+from PIL import Image
+
+file = Image.open('dog.png')
+file.show()
+
+image = np.array(file)
+
+print(image.shape)
+
+# 直接访问像素点颜色
+print(image[100, 100])
+
+red = image[:, :, 0]
+green = image[:, :, 1]
+blue = image[:, :, 2]
+
+Image.fromarray(red).show()
+Image.fromarray(green).show()
+Image.fromarray(blue).show()
+
+# 直接混合图片
+# conbine = a1 * 0.2 + a2 * 0.8
+# 显示图片需要换为unit8
+# combine.astype(np.unit8)
+