There is an increasing demand for people who can analyse and make sense out of the data. Practical math is crucial because it allows you to draw insights from data. It is difficult to know which topics are relevant and applicable in mathematics for data science careers. These essential math topics are key to integrating data science, statistics, or machine learning knowledge.
Thomas Nield, data science expert, will guide you through a carefully selected list of mathematical topics that will help you to become proficient in the areas you can apply immediately. Probability, statistics, hypothesis testing and linear regression are all covered. You will also learn practical calculus. You'll learn how to apply what you have learned to real-world problems.
Week 1 - Probability
Probability is an integral part of data science, machine-learning, statistics, and analytics. Although measuring random events may seem complicated and abstract, there is a practical and intuitive way to determine probability. This tour will cover foundational concepts of probability such as conditional probability and Bayes' Theorem. These ideas will be illustrated with Python code.
Week 2 - Statistics and Hypothesis Testing
This session will cover descriptive metrics such as variance and their nuances. It also covers inferential statistics topics like hypothesis testing, normal distributions and t–distributions.
Week 3: Linear Algebra
Linear algebra is an enormous and sometimes an intimidating topic, but in this session you'll explore a more visual and intuitive approach to vectors, matrices, and their transformations--which will help you write readable Python code. This intuitive visual understanding will change the way you look at NumPy.
Week 4 - Calculus and Functions
A little bit of calculus is helpful in machine learning. However, this method will not be used in academia. Instead, you will use plain English and avoid using expressions that are stuffed with Greek symbols. You'll begin with an intuitive understanding and use of functions and numbers. Then you will move on to calculus concepts such as integrals and derivatives. You'll also learn how to create Python code.
Week 5: Linear Regression
Linear regression can be used to predict the relationship between two variables. Linear regression is a popular tool for many types of regression tasks due to its simplicity and resistance to overfitting. This session will cover multiple methods of a linear regression. These include library and "from scratch" implementations, using least squares, hill climbing and gradient descent techniques. The day will also cover variance and overfitting, performance metrics such as R2, and cross validation techniques to avoid overfitting.
Week 6 - Logistic Regression & Classification
Logistic regression can be a powerful and simple form of regression. It is useful for classification tasks. This is why it is often the first algorithm that practitioners use. This session will cover logistic regression and its many applications. You'll also learn techniques for evaluating classification performance such as confusion matrices and ROC. You'll be able to see Python examples such as predicting employee retention or the Space Shuttle Challenger catastrophe.
NOTE: By registering today, you will be signed up to all six sessions. You can attend any session individually but we recommend that you participate in all six weeks.
Challenges
Thomas Nield will present you with a challenge at the end of each week. This interactive scenario-based assessment will help you assess whether or not you have mastered the skills learned in live training.
We recommend that you complete each challenge before moving on to the next week. This will reinforce your learning. For tips and assistance if you are unable to complete the challenge successfully, you can review the video recording of each live training session (emailed 24 hours after each session).
