randint(), randperm() and normal() in PyTorch

*Memos:

[Warning]-normal() is really tricky.

My post explains rand() and rand_like().

My post explains randn() and randn_like().

randint() create the 1D or more D tensor of the zero or more random integers(Default) or floating-point numbers betwe…


This content originally appeared on DEV Community and was authored by Super Kai (Kazuya Ito)

*Memos:

randint() create the 1D or more D tensor of the zero or more random integers(Default) or floating-point numbers between low and high-1(low<=x<=high-1) as shown below:

*Memos:

  • randint() can be used with torch but not with a tensor.
  • The 1st argument with torch is low(Optional-Default:0-Type:int): *Memos:
    • It must be lower than high.
    • The 0D or more D tensor of one integer works.
  • The 2nd argument with torch is high(Required-Type:int): *Memos:
    • It must be greater than low.
    • The 0D or more D tensor of one integer works.
  • The 3rd argument with torch is size(Required-Type:tuple of int or list of int).
  • There is dtype argument with torch(Optional-Type:dtype): *Memos:
    • If dtype is not given, dtype is torch.int64.
    • int or float can be used.
    • dtype= must be used.
    • My post explains dtype argument.
  • There is device argument with torch(Optional-Type:str, int or device()): *Memos:
    • device= must be used.
    • My post explains device argument.
  • There is requires_grad argument with torch(Optional-Type:bool): *Memos:
    • requires_grad= must be used.
    • My post explains requires_grad argument.
  • There is out argument with torch(Optional-Type:tensor): *Memos:
    • out= must be used.
    • My post explains out argument. 
import torch

torch.randint(high=10, size=(3,))
# tensor([7, 4, 8])

torch.randint(high=10, size=(3, 2))
# tensor([[8, 9], [6, 5], [5, 2]])

torch.randint(high=10, size=(3, 2, 4))
# tensor([[[1, 5, 9, 0], [4, 6, 7, 2]],
#         [[5, 2, 1, 5], [9, 3, 2, 6]],
#         [[9, 3, 6, 4], [0, 4, 7, 5]]])

torch.randint(low=10, high=20, size=(3,))
# tensor([17, 12, 10])

torch.randint(low=10, high=20, size=(3, 2))
# tensor([[14, 18], [10, 19], [15, 16]])

torch.randint(low=10, high=20, size=(3, 2, 4))
# tensor([[[16, 14, 11, 19], [19, 15, 18, 13]],
#         [[14, 10, 11, 13], [16, 11, 10, 16]], 
#         [[17, 12, 17, 10], [13, 16, 11, 10]]])

torch.randint(low=-5, high=5, size=(3,))
# tensor([-1,  2, -3])

torch.randint(low=-5, high=5, size=(3, 2))
# tensor([[-5,  4], [ 1, -1], [-4, -3]])

torch.randint(low=-5, high=5, size=(3, 2, 4))
# tensor([[[-2, 0, 1, -5], [4, -5, -3, 1]],
#         [[-4, -1, -1, -1], [-3, 2, -4, -1]],
#         [[4, -1, -5, -3], [2, -3, -2, 2]]])

torch.randint(low=-5, high=5, size=(3, 2, 4), dtype=torch.float32)
torch.randint(low=torch.tensor(-5),
              high=torch.tensor([5]),
              size=(3, 2, 4),
              dtype=torch.float32)
# tensor([[[-4., 1., -1., -3.], [-3., -5., -4., 1.]],
#         [[-5., 3., 3., 1.], [-1., 4., -5., 2.]],
#         [[-2., -4., -5., 3.], [4., 1., -3., 3.]]])

torch.randint(high=1, size=(0,))
torch.randint(low=0, high=1, size=(0,))
torch.randint(low=10, high=20, size=(0,))
# tensor([], dtype=torch.int64)

randperm() can create the 1D tensor of zero or more random integers(Default) or floating-point numbers between 0 and n-1(0<=x<=n-1) as shown below:

*Memos:

  • randperm() can be used with torch but not with a tensor.
  • The 1st argument with torch is n(Required-Type:int): *Memos:
    • It must be greater than or equal to 1.
    • The 0D or more D tensor of one integer works.
  • There is dtype argument with torch(Optional-Type:dtype): *Memos:
    • If dtype is not given, dtype is torch.int64.
    • int or float can be used.
    • dtype= must be used.
    • My post explains dtype argument.
  • There is device argument with torch(Optional-Type:str, int or device()): *Memos:
    • device= must be used.
    • My post explains device argument.
  • There is requires_grad argument with torch(Optional-Type:bool): *Memos:
    • requires_grad= must be used.
    • My post explains requires_grad argument.
  • There is out argument with torch(Optional-Type:tensor): *Memos:
    • out= must be used.
    • My post explains out argument. 
import torch

torch.randperm(n=0)
# tensor([], dtype=torch.int64)

torch.randperm(n=5)
# tensor([3, 0, 4, 2, 1])

torch.randperm(n=10)
# tensor([8, 6, 9, 2, 1, 3, 5, 0, 7, 4])

torch.randperm(n=10, dtype=torch.float32)
torch.randperm(n=torch.tensor([[10]]), dtype=torch.float32)
# tensor([7., 4., 2., 1., 8., 3., 0., 6., 9., 5.])

normal() can create the 1D or more D tensor of zero or more random floating-point numbers or complex numbers from normal distribution as shown below:

*Memos:

  • normal() can be used with torch but not with a tensor.
  • The 1st argument with torch is mean(Required-Type:float or complex or tensor of float or complex): *Memos:
    • Setting mean without std and size is tensor of float or complex.
    • Setting mean and std without size is float or tensor of float or complex.
    • Setting mean, std and size is float or tensor of float. *The 0D tensor of float also works.
  • The 2nd argument with torch is std(Optional-Type:float or tensor of float): *Memos:
    • It is standard deviation.
    • It must be greater than or equal to 0.
    • Setting std without size is float or tensor of float.
    • Setting std with size is float or tensor of float. *The 0D tensor of float also works.
  • The 3rd argument with torch is size(Optional-Type:tuple of int or list of int): *Memos:
    • It must be used with std.
    • It must not be negative.
  • There is dtype argument with torch(Optional-Type:dtype): *Memos:
    • If dtype is not given, dtype is inferred from mean or std or dtype of set_default_dtype() is used for floating-point numbers.
    • Only float and complex can be used.
    • dtype= must be used.
    • My post explains dtype argument.
  • There is device argument with torch(Optional-Type:str, int or device()): *Memos:
    • device= must be used.
    • My post explains device argument.
  • There is requires_grad argument with torch(Optional-Type:bool): *Memos:
    • requires_grad= must be used.
    • My post explains requires_grad argument.
  • There is out argument with torch(Optional-Type:tensor): *Memos:
    • out= must be used.
    • My post explains out argument. 
import torch

torch.normal(mean=torch.tensor([1., 2., 3.]))
# tensor([1.2713, 0.7271, 3.5027])

torch.normal(mean=torch.tensor([1.+0.j, 2.+0.j, 3.+0.j]))
# tensor([1.1918-0.9001j, 2.3555+0.2956j, 2.5479-0.4672j])

torch.normal(mean=torch.tensor([1., 2., 3.]),
             std=torch.tensor([4., 5., 6.]))
# tensor([2.0851, -4.3646, 6.0162])

torch.normal(mean=torch.tensor([1.+0.j, 2.+0.j, 3.+0.j]),
             std=torch.tensor([4., 5., 6.]))
# tensor([1.7673-3.6004j, 3.7773+1.4781j, 0.2872-2.8034j])

torch.normal(mean=torch.tensor([1., 2., 3.]), std=4.)
# tensor([2.0851, -3.0917, 5.0108])

torch.normal(mean=torch.tensor([1.+0.j, 2.+0.j, 3.+0.j]), std=4.)
# tensor([1.7673-3.6004j, 3.4218+1.1825j, 1.1914-1.8689j])

torch.normal(mean=1., std=torch.tensor([4., 5., 6.]))
# tensor([2.0851, -5.3646, 4.0162])

torch.normal(mean=1., std=4., size=(3,))
torch.normal(mean=torch.tensor(1.), std=torch.tensor(4.), size=(3,))
# tensor([2.0851, -4.0917, 3.0108])

torch.normal(mean=1., std=4., size=(3, 2))
torch.normal(mean=torch.tensor(1.), std=torch.tensor(4.), size=(3, 2))
# tensor([[2.0851, -4.0917],
#         [3.0108, 2.6723],
#         [-1.5577, -1.6431]])

torch.normal(mean=1., std=4., size=(3, 2, 4))
torch.normal(mean=torch.tensor(1.), std=torch.tensor(4.), size=(3, 2, 4))
# tensor([[[-3.7568, 6.5729, 9.4236, -0.4183],
#          [2.4840, 5.3827, 9.5657, 1.5267]],
#         [[8.0575, -0.5000, -0.3416, 5.3502],
#          [-4.3835, 1.6974, 2.6226, -1.9671]],
#         [[1.1422, 1.7790, 4.5886, -0.3273],
#          [2.8941, -3.3046, 1.1336, 2.8792]]])


This content originally appeared on DEV Community and was authored by Super Kai (Kazuya Ito)


Print Share Comment Cite Upload Translate Updates
APA

Super Kai (Kazuya Ito) | Sciencx (2024-07-10T13:49:24+00:00) randint(), randperm() and normal() in PyTorch. Retrieved from https://www.scien.cx/2024/07/10/randint-randperm-and-normal-in-pytorch/

MLA
" » randint(), randperm() and normal() in PyTorch." Super Kai (Kazuya Ito) | Sciencx - Wednesday July 10, 2024, https://www.scien.cx/2024/07/10/randint-randperm-and-normal-in-pytorch/
HARVARD
Super Kai (Kazuya Ito) | Sciencx Wednesday July 10, 2024 » randint(), randperm() and normal() in PyTorch., viewed ,<https://www.scien.cx/2024/07/10/randint-randperm-and-normal-in-pytorch/>
VANCOUVER
Super Kai (Kazuya Ito) | Sciencx - » randint(), randperm() and normal() in PyTorch. [Internet]. [Accessed ]. Available from: https://www.scien.cx/2024/07/10/randint-randperm-and-normal-in-pytorch/
CHICAGO
" » randint(), randperm() and normal() in PyTorch." Super Kai (Kazuya Ito) | Sciencx - Accessed . https://www.scien.cx/2024/07/10/randint-randperm-and-normal-in-pytorch/
IEEE
" » randint(), randperm() and normal() in PyTorch." Super Kai (Kazuya Ito) | Sciencx [Online]. Available: https://www.scien.cx/2024/07/10/randint-randperm-and-normal-in-pytorch/. [Accessed: ]
rf:citation
» randint(), randperm() and normal() in PyTorch | Super Kai (Kazuya Ito) | Sciencx | https://www.scien.cx/2024/07/10/randint-randperm-and-normal-in-pytorch/ |

Please log in to upload a file.



There are no updates yet.
Click the Upload button above to add an update.

You must be logged in to translate posts. Please log in or register.

Related Posts