()How to create a normal distributed random number
(*size) Gets a random number from a normal distribution with mean of 0 and variance of 1
【sample】
In [1]: import torch In [2]: (3) Out[2]: tensor([1.7896, 0.7974, 0.7416]) In [3]: (2,3) Out[3]: tensor([[ 0.4030, -0.3138, -0.7579], [-0.1486, 1.0306, 0.0734]]) In [4]: (()) Out[4]: tensor(-0.8383) # The dimension is0
Torch's random number generation method
() () () ()
1. Even distribution
(*sizes, out=None) → Tensor
Returns a tensor containing a set of random numbers extracted from the uniform distribution of intervals [0, 1). The shape of a tensor is defined by the parameter sizes.
parameter:
-
sizes (int...)- Integer sequence, defining the shape of the output tensor -
out (Tensor, optinal)- Result Tensor
example:
(2, 3) 0.0836 0.6151 0.6958 0.6998 0.2560 0.0139 [ of size 2x3]
2. Standard normal distribution
(*sizes, out=None) → Tensor
Returns a tensor containing a set of random numbers extracted from a standard normal distribution (mean value is 0, variance is 1, i.e. Gaussian white noise). The shape of a tensor is defined by the parameter sizes.
parameter:
-
sizes (int...)- Integer sequence, defining the shape of the output tensor -
out (Tensor, optinal)- Result Tensor
example:
(2, 3) 0.5419 0.1594 -0.0413 -2.7937 0.9534 0.4561 [ of size 2x3]
3. Discrete normal distribution
(means, std, out=None) → → Tensor
Returns a tensor containing a set of random numbers extracted from a discrete normal distribution of the specified mean means and standard deviation std.
The standard deviation std is a tensor containing the standard deviation of the normal distribution associated with each output element.
parameter:
-
means (float, optional)- Mean -
std (Tensor)- Standard deviation -
out (Tensor)- Output tensor
example:
(mean=0.5, std=(1, 6)) -0.1505 -1.2949 -4.4880 -0.5697 -0.8996 [ of size 5]
4. Linear spacing vector
(start, end, steps=100, out=None) → Tensor
Returns a 1-dimensional tensor containing evenly spaced step points on intervals start and end.
The length of the output tensor is determined by steps.
parameter:
-
start (float)- The starting point of the interval -
end (float)- End point of interval -
steps (int)- Number of samples generated between start and end -
out (Tensor, optional)- Result Tensor
example:
(3, 10, steps=5) 3.0000 4.7500 6.5000 8.2500 10.0000 [ of size 5]
Summarize
The above is personal experience. I hope you can give you a reference and I hope you can support me more.