5 Statistical Functions in PyTorch
PyTorch functions useful for machine learning
PyTorch is a Python package that provides two high-level features:
- Tensor computation (like NumPy) with strong GPU acceleration
- Deep neural networks built on a tape-based autograd system
PyTorch is designed to be intuitive, linear in thought and easy to use. When you execute a line of code, it gets executed. It has minimal framework overhead. It is fast — whether you run small or large neural networks.
In this blog we will discussed 5 Statistical Functions in PyTorch that I find interesting:
- torch.bernoulli()
- torch.poisson()
- torch.normal()
- torch.normal_()
- torch.norm()
We create a custom tensor and passed to bernoulli function it return a binary number (0 or 1).
If we create a tensor of All one’s .Probability of drawing “1” is 1.
All zero’s in the matrix.probability of drawing "1" is 0
These are 3 types of example of torch.bernoulli() function.
Bernoulli Distribution is a random experiment that has only two outcomes (usually called a “Success” or a “Failure”). It is best used when we have two outcomes of a given event.
A input tensor contain rate parameter between 0 and 6 with 3 rows and 3 columns, returns a output from a poission distribution.
A input tensor contain rate parameter between 0 and 7 with 5 rows and 4 columns, returns a output from a poission distribution
The Poisson distribution is popular for modeling the number of times an event occurs in an interval of time or space.
It returns a tensor of random numbers drawn from separate normal distributions.
It returns a tensor of random numbers drawn from separate normal distributions whose standard deviation is 1.
It does not work when mean is 0 and std is 1 so you will learn about tensor.normal_() in the next function.
The normal distribution is a probability function that describes how the values of a variable are distributed.
It creates a standard normal distribution with mean =0 and std=1.
It create a matrix Z (a 1d tensor) of dimension 1 × 5, filled with random elements samples from the normal distribution parameterized by mean = 4 and std = 0.5.
It helps to create standard normal distribution
It returns vector norm of a given tensor where dim =0.
It returns vector norm of a given tensor where dim =1.
It returns vector norm of a given tensor where dim =1 and p=1.
Matrix norms are indirectly used in any applications that require matrix functions and/or matrix series.
Conclusion
These are the 5 statistical PyTorch function that I find interesting as discussed above and you can find more in the PyTorch documentation.
Reference Links
Provide links to your references about tensors
- Official documentation for
torch.Tensor: https://pytorch.org/docs/stable/tensors.html
