YDOOKAI:Pytorch:pytorch常用的张量生成与张量定义函数tensor生成与定义函数方法

YDOOK AI : Pytorch : pytorch常用的张量生成与张量定义函数 tensor 生成与定义函数方法

1. 源代码展示&＃xff1a;

import torch
a &＃61; [[1, 2, 3], [4, 5, 6]]
print(&＃39;a &＃61; &＃39;, a)
# YDOOK JYLin : 普通的数组不能使用 tensor.dytpe 获取数据类型&＃xff0c;需要使用 type(valuable) 获取
print(&＃39;type(a) &＃61; &＃39;, type(a))
print()
t1 &＃61; torch.tensor(a)
print(&＃39;t1 &＃61; &＃39;, t1)
print(&＃39;t1.type &＃61; &＃39;, t1.dtype)
print()
t2_1 &＃61; torch.zeros(2, 3)
print(&＃39;t2_1 &＃61; &＃39;, t2_1)
print(&＃39;t2_1.type &＃61; &＃39;, t2_1.dtype)
print()
t2_2 &＃61; torch.zeros(3, 2)
print(&＃39;t2_2 &＃61; &＃39;, t2_2)
print(&＃39;t2_2.type &＃61; &＃39;, t2_2.dtype)
print()
t3 &＃61; torch.zeros_like(t1)
print(&＃39;t3 &＃61; &＃39;, t3)
print(&＃39;t3.type &＃61; &＃39;, t3.dtype)
print()
t4_1 &＃61; torch.ones(2, 3)
print(&＃39;t4_1 &＃61; &＃39;, t4_1)
print(&＃39;t4_1.type &＃61; &＃39;, t4_1.dtype)
print()
t4_2 &＃61; torch.ones(2, 3)
print(&＃39;t4_2 &＃61; &＃39;, t4_2)
print(&＃39;t4_2.type &＃61; &＃39;, t4_2.dtype)
print()
t5 &＃61; torch.ones_like(t1)
print(&＃39;t5 &＃61; &＃39;, t5)
print(&＃39;t5.type &＃61; &＃39;, t5.dtype)
print()
t6 &＃61; torch.full(t1.size(), 3.14)
print(&＃39;t6 &＃61; &＃39;, t6)
print(&＃39;t6.type &＃61; &＃39;, t6.dtype)
print()
t6_2 &＃61; torch.full(t1.size(), 3.14, dtype&＃61;torch.float64)
print(&＃39;t6_2 &＃61; &＃39;, t6_2)
print(&＃39;t6_2.type &＃61; &＃39;, t6_2.dtype)
print()
t7 &＃61; torch.full_like(t1, 6.28)
print(&＃39;t7 &＃61; &＃39;, t7)
print(&＃39;t7.type &＃61; &＃39;, t7.dtype)
print()
t7_2 &＃61; torch.full_like(t1, 6.28, dtype&＃61;torch.float64)
print(&＃39;t7_2 &＃61; &＃39;, t7_2)
print(&＃39;t7_2.type &＃61; &＃39;, t7_2.dtype)
print()
t8 &＃61; torch.empty(t1.size())
print(&＃39;t8 &＃61; &＃39;, t8)
print(&＃39;t8.type &＃61; &＃39;, t8.dtype)
print()
t8_2 &＃61; torch.empty(t1.size(), out&＃61;t1)
print(&＃39;t8_2 &＃61; &＃39;, t8_2)
print(&＃39;t8_2.type &＃61; &＃39;, t8_2.dtype)
print()
t8_3 &＃61; torch.empty_like(t1)
print(&＃39;t8_3 &＃61; &＃39;, t8_3)
print(&＃39;t8_3.type &＃61; &＃39;, t8_3.dtype)
print()
t9 &＃61; torch.eye(2, 3)
print(&＃39;t9 &＃61; &＃39;, t9)
print(&＃39;t9.type &＃61; &＃39;, t9.dtype)
print()
t9_2 &＃61; torch.eye(3, 3)
print(&＃39;t9_2 &＃61; &＃39;, t9_2)
print(&＃39;t9_2.type &＃61; &＃39;, t9_2.dtype)
print()
t10 &＃61; torch.arange(1, 10)
print(&＃39;t10 &＃61; &＃39;, t10)
print(&＃39;t10.type &＃61; &＃39;, t10.dtype)
print()
# YDOOK JYLin : 很有用的 torch.linspace(start, end, steps)
t11 &＃61; torch.linspace(1, 10, steps&＃61;10)
print(&＃39;t11 &＃61; &＃39;, t11)
print(&＃39;t11.type &＃61; &＃39;, t11.dtype)
print()
t11_2 &＃61; torch.linspace(1, 10, steps&＃61;9)
print(&＃39;t11_2 &＃61; &＃39;, t11_2)
print(&＃39;t11_2.type &＃61; &＃39;, t11_2.dtype)
print()
t11_3 &＃61; torch.linspace(1, 10, steps&＃61;8)
print(&＃39;t11_3 &＃61; &＃39;, t11_3)
print(&＃39;t11_3.type &＃61; &＃39;, t11_3.dtype)
print()


2. 输出&＃xff1a;

a &＃61; [[1, 2, 3], [4, 5, 6]]
type(a) &＃61;
t1 &＃61; tensor([[1, 2, 3],
[4, 5, 6]])
t1.type &＃61; torch.int64
t2_1 &＃61; tensor([[0., 0., 0.],
[0., 0., 0.]])
t2_1.type &＃61; torch.float32
t2_2 &＃61; tensor([[0., 0.],
[0., 0.],
[0., 0.]])
t2_2.type &＃61; torch.float32
t3 &＃61; tensor([[0, 0, 0],
[0, 0, 0]])
t3.type &＃61; torch.int64
t4_1 &＃61; tensor([[1., 1., 1.],
[1., 1., 1.]])
t4_1.type &＃61; torch.float32
t4_2 &＃61; tensor([[1., 1., 1.],
[1., 1., 1.]])
t4_2.type &＃61; torch.float32
t5 &＃61; tensor([[1, 1, 1],
[1, 1, 1]])
t5.type &＃61; torch.int64
t6 &＃61; tensor([[3.1400, 3.1400, 3.1400],
[3.1400, 3.1400, 3.1400]])
t6.type &＃61; torch.float32
t6_2 &＃61; tensor([[3.1400, 3.1400, 3.1400],
[3.1400, 3.1400, 3.1400]], dtype&＃61;torch.float64)
t6_2.type &＃61; torch.float64
t7 &＃61; tensor([[6, 6, 6],
[6, 6, 6]])
t7.type &＃61; torch.int64
t7_2 &＃61; tensor([[6.2800, 6.2800, 6.2800],
[6.2800, 6.2800, 6.2800]], dtype&＃61;torch.float64)
t7_2.type &＃61; torch.float64
t8 &＃61; tensor([[3.1400, 3.1400, 3.1400],
[3.1400, 3.1400, 3.1400]])
t8.type &＃61; torch.float32
t8_2 &＃61; tensor([[1, 2, 3],
[4, 5, 6]])
t8_2.type &＃61; torch.int64
t8_3 &＃61; tensor([[4614253070214989087, 4614253070214989087, 4614253070214989087],
[4614253070214989087, 4614253070214989087, 4614253070214989087]])
t8_3.type &＃61; torch.int64
t9 &＃61; tensor([[1., 0., 0.],
[0., 1., 0.]])
t9.type &＃61; torch.float32
t9_2 &＃61; tensor([[1., 0., 0.],
[0., 1., 0.],
[0., 0., 1.]])
t9_2.type &＃61; torch.float32
t10 &＃61; tensor([1, 2, 3, 4, 5, 6, 7, 8, 9])
t10.type &＃61; torch.int64
t11 &＃61; tensor([ 1., 2., 3., 4., 5., 6., 7., 8., 9., 10.])
t11.type &＃61; torch.float32
t11_2 &＃61; tensor([ 1.0000, 2.1250, 3.2500, 4.3750, 5.5000, 6.6250, 7.7500, 8.8750,
10.0000])
t11_2.type &＃61; torch.float32
t11_3 &＃61; tensor([ 1.0000, 2.2857, 3.5714, 4.8571, 6.1429, 7.4286, 8.7143, 10.0000])
t11_3.type &＃61; torch.float32