把tensor转为numpy_Torch和Numpy

古城微笑少年丶 2023-01-02 05:21 317阅读 0赞

![7ffc7aa19a7325252af4804845f3fa80.png][]

\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*本文参考\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*

Torch 或 Numpy - PyTorch | 莫烦Pythonmofanpy.com

![e2893b7b59ff375dfef9e4d567b87abc.png][]

\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*

## 1 Numpy array 和 Torch tensor 的转换 ##

【1】

import torch
    import numpy as np
    
    np_data = np.arange(6).reshape((2, 3))
    torch_data = torch.from_numpy(np_data)
    tensor2array = torch_data.numpy()
    print(
        'nnumpy array:', np_data,
        'ntorch tensor:', torch_data,
        'ntensor to array:', tensor2array,
    )

得到

numpy array: [[0 1 2]
     [3 4 5]] 
    torch tensor: tensor([[0, 1, 2],
            [3, 4, 5]], dtype=torch.int32) 
    tensor to array: [[0 1 2]
     [3 4 5]]

【2】

import torch
    import numpy as np
    
    data = [[1, 2], [3, 4]]
    tensor = torch.FloatTensor(data)
    data_array = np.array(data)
    print(data)
    print(data_array)
    print(tensor)

得到

[[1, 2], [3, 4]]
    [[1 2]
     [3 4]]
    tensor([[1., 2.],
            [3., 4.]])

## 2 Numpy 和 Torch 中的数学运算 ##

其实torch中tensor的运算和numpy array的如出一辙，下面以对比的形式来看看。

Ps: 如果想了解torch中其它更多有用的运算符，参考

torch - PyTorch 1.7.0 documentationpytorch.org

**2.1 abs 绝对值**

import torch
    import numpy as np
    
    # abs 绝对值计算
    data = [-1, -2, 1, 2]
    tensor = torch.FloatTensor(data)  # 转换成32位浮点 tensor
    print(
        'nabs',
        'nnumpy: ', np.abs(data),       
        'ntorch: ', torch.abs(tensor)
    )

得到

abs 
    numpy:  [1 2 1 2] 
    torch:  tensor([1., 2., 1., 2.])

**2.2 三角函数 sin**

import torch
    import numpy as np
    
    # sin   三角函数 sin
    data = [-1, -2, 1, 2]
    tensor = torch.FloatTensor(data)  # 转换成32位浮点 tensor
    print(
        'nsin',
        'nnumpy: ', np.sin(data),
        'ntorch: ', torch.sin(tensor)  
    )

得到

sin 
    numpy:  [-0.84147098 -0.90929743  0.84147098  0.90929743] 
    torch:  tensor([-0.8415, -0.9093,  0.8415,  0.9093])

**2.3 mean 均值**

import torch
    import numpy as np
    
    # mean  均值
    data = [-1, -2, 1, 2]
    tensor = torch.FloatTensor(data)  # 转换成32位浮点 tensor
    print(
        'nmean',
        'nnumpy: ', np.mean(data),      
        'ntorch: ', torch.mean(tensor)
    )

得到

mean 
    numpy:  0.0 
    torch:  tensor(0.)

**2.4 矩阵乘法**

【形式一】：

import torch
    import numpy as np
    
    data = [[1, 2], [3, 4]]
    tensor = torch.FloatTensor(data)  # 转换成32位浮点 tensor
    print(
        'nmatrix multiplication (matmul)',
        'nnumpy: ', np.matmul(data, data),
        'ntorch: ', torch.mm(tensor, tensor)
    )

得到

matrix multiplication (matmul) 
    numpy:  [[ 7 10]
     [15 22]] 
    torch:  tensor([[ 7., 10.],
            [15., 22.]])

【形式二】(torch变得不一样)：

import torch
    import numpy as np
    
    data = [[1, 2], [3, 4]]
    data = np.array(data)             # 将data转为array形式
    data1 = [5, 7]
    tensor = torch.FloatTensor(data1)
    print(
        'nmatrix multiplication (dot)',
        'nnumpy: ', data.dot(data),                  # [[7, 10], [15, 22]] 在numpy 中可行
        # tensor.dot()只能针对于一维的数组
        'ntorch: ', tensor.dot(tensor)
    )

得到

matrix multiplication (dot) 
    numpy:  [[ 7 10]
     [15 22]] 
    torch:  tensor(74.)

[7ffc7aa19a7325252af4804845f3fa80.png]: /images/20221120/06181291b9d74895b2702d2d162bc843.png
[e2893b7b59ff375dfef9e4d567b87abc.png]: /images/20221120/631b1cb40cb34fa2bb7d5b9490e3e04a.png