From the course: Machine Learning Foundations: Statistics

Introduction to probability distribution - Python Tutorial

From the course: Machine Learning Foundations: Statistics

Introduction to probability distribution

“

- [Instructor] If you have previously taken my ML probability course, you have explored different types of probability distributions. If you haven't, I highly recommend taking it as probability and statistics go hand in hand, like coffee and milk. But don't worry, we'll cover the basics related to statistics. In the previous chapters, we have covered the basics of descriptive statistics. And now we'll cover inferential statistics that cover methods built upon probability theory and distributions. Distribution is defined as a function that shows the possible values for a variable and how often they occur. The distribution type depends upon the types of variables. Random variables can be divided into discrete and continuous. A discrete variable takes only a limited set of values from a given range of values. For example, the number of seats in a cinema, the number of questions in a quiz, the number of pets in a family, et cetera, are all discrete variables. A continuous variable takes any value from a given range of values. For example, temperature, a person's height, a customer's waiting time at the drive-in restaurant are all continuous variables. So statisticians divide probability distributions into the following two types: discrete probability distribution and continuous probability distribution. It is important to understand the difference between a distribution and a graph. As many people think they are the same. The graph is just a representation of a function. For example, here we have a graph of the most popular distribution called normal or Gaussian distribution. We define the distribution based on the underlying probabilities, and we use the graph to show the probability of every event visually. Probability distributions play a crucial role in ML as they help you make predictions in case our models involve randomness, or in the case you're uncertain about outcomes.

Contents