[pytorch 07] hands on deep learning chapter_ Preliminaries/ndarray exercises hands-on version

  Catalog

1. Import torch

2. Use arange Create a row vector x, This row vector contains the following elements 0 Before we start 12 It's an integer .

3. Through tensor shape Property to access the tensor x（ Length along each axis ） The shape of the

4. Just want to know the total number of elements in the tensor , You can check its size （size）

5. Put tensor x From shape to （12,） The row vector of is transformed into a shape of （3,4） Matrix .

6. Create a shape as （2,3,4） Tensor , All elements are set to 0.

7. Create a shape as (2,3,4) Tensor , All elements are set to 1.

8. Create a shape as （3,4） Tensor . Each of these elements has a mean value of 0、 The standard deviation is 1 The standard Gaussian distribution of （ Normal distribution ） Medium random sampling .

9. torch.tensor Create a （3,4） Two dimensional array of .

10. Realize the addition, subtraction, multiplication and division of the following two arrays

11. Exponentiate the following array

12. The following two arrays , Press the line / Splicing by columns

13. Judge whether each element of the following two homogeneous arrays is equal

 14. Calculation X The sum of all elements in the array

15. Use arange Create a 3*1 Array of  a and 1*2 Array of b 

16. Try adding the above two arrays directly , View the situation .

17. Take out the two-dimensional array X Choose from The last line of elements 、 The second to third lines of elements

18、 Modify two-dimensional array X The first 1 Xing di 2 The value of the column is 9 And write the matrix

19、 Modify two-dimensional array X The first 0 Xing He 1 All elements of the row are 12.

20、 take X Turn into numpy Assign to A, then A Turn into tensor In the form of B

 21、 Output a Original number of 、 Character 、 floating-point

---

1. Import torch

2. Use arange Create a row vector x, This row vector contains the following elements 0 Before we start 12 It's an integer .

3. Through tensor shape Property to access the tensor x（ Length along each axis ） Of shape

4. Just want to know the total number of elements in the tensor , You can check its size （size）

5. Put tensor x From shape to （12,） The row vector of is transformed into a shape of （3,4） Matrix .

6. Create a shape as （2,3,4） Tensor , All elements are set to 0.

7. Create a shape as (2,3,4) Tensor , All elements are set to 1.

8. Create a shape as （3,4） Tensor . Each of these elements has a mean value of 0、 The standard deviation is 1 The standard Gaussian distribution of （ Normal distribution ） Medium random sampling .

9. torch.tensor Create a （3,4） Two dimensional array of .

【 Operator 】

10. Realize the addition, subtraction, multiplication and division of the following two arrays

x = torch.tensor([1.0, 2, 4, 8])
y = torch.tensor([2, 2, 2, 2])

11. Exponentiate the following array

x = torch.tensor([1.0, 2, 4, 8])

12. The following two arrays , Press the line / Splicing by columns

X = torch.arange(12, dtype=torch.float32).reshape((3,4))
Y = torch.tensor([[2.0, 1, 4, 3], [1, 2, 3, 4], [4, 3, 2, 1]])

13. Judge whether each element of the following two homogeneous arrays is equal

X = torch.arange(12, dtype=torch.float32).reshape((3,4))
Y = torch.tensor([[2.0, 1, 4, 3], [1, 2, 3, 4], [4, 3, 2, 1]])

 14. Calculation X The sum of all elements in the array

X = torch.arange(12, dtype=torch.float32).reshape((3,4))

【 Broadcast mechanism 】

15.  Use arange Create a 3*1 Array of  a and 1*2 Array of b 

16. Try adding the above two arrays directly , View the situation .

【 Index and slice 】

17. Take out the two-dimensional array X in choice The last line of elements 、 The second to third lines of elements

X = torch.arange(12, dtype=torch.float32).reshape((3,4))

18、 Modify two-dimensional array X The first 1 Xing di 2 The value of the column is 9 and Write matrix

X = torch.arange(12, dtype=torch.float32).reshape((3,4))

19、 Modify two-dimensional array X The first 0 Xing He 1 All elements of the row are 12.

X = torch.arange(12, dtype=torch.float32).reshape((3,4))

【 Convert objects 】

20、 take X Turn into numpy Assign to A, then A Turn into tensor In the form of B

X = torch.arange(12, dtype=torch.float32).reshape((3,4))

 21、 Output a Original number of 、 Character 、 floating-point

a = torch.tensor([3.5])