try to solve some of the problems given in the documentation
- What is normal distribution
- What is univariate, bivariate, and multivariate analysis
- what is bivariate normal distribution
- Marginal PDF
- Python
- JAX library
- matplotlib
Univariate: Using only one class of feature for generating the output
y = theta1*x + theta2
Where thetas are weight and bias
Bivariate: Using two different class of features for generating the output
z = theta1*x + theta2*y + theta3
This multivariate analysis help us to find the correlation between the features
Also known as gaussian distribution Continuous probability distribution for real-valued random variable Probability destity function
- Height of people
- Prices of shares in stock market
- Income distribution in ecomnomy
- Student average marks
jax.random.multivariate_normal(key, mean, cov, shape=None, dtype=<class 'numpy.float64'>, method='cholesky') -> array
It is a matrix decomposistion method to factorize a matrix into a product of metrices
It is a decomposition of a Hermitian (positive-definite) matrix into the product of a lower triangular matrix and its conjugate transpose.
A = [L][L]T
[[4, 12, -16]
[12, 37, -43]
[-16, -43, 98]
]
-----------------------
[2 0 0] [2 6 -8]
[6 1 0] * [0 1 5]
[-8 5 3] [0 0 3]
Variance
refers to the spread of a data set around its mean value.covariance
refers to the measure of the directional relationship between two random variables.