This is where most of the people should begin as Python is the base of modern data work, automation and machine learning. Short, exercise-driven lessons include variables, functions, conditionals, lists, loops, strings, dictionaries and external libraries. It’s designed to be used to prepare you for the upcoming Pandas and ML courses from Kaggle, which is why the course is best suited for complete beginners. If you have played around with another language in the past, the pace of the exercises won’t seem so slow that you feel you’re wasting time on concepts you know well.

After learning the fundamentals of Python, the next step is Pandas. You will do some work with Series and DataFrames, starting to read, select, group, combine and clean real tabular data. The skills that you develop in these exercises will be useful in nearly all data science tasks you undertake in the future, and the exercises are short enough to make you confident faster. You will be returning to this course later as well because Pandas syntax can get lost in the mind when you don’t use it on a daily basis.

It is difficult to get people to buy in when you use only numbers. This course will familiarize you with Seaborn and the typical types of charts (line charts, bar charts, heatmaps, scatter plots) that you can use to communicate trends, comparisons and relationships instead of calculating them. Everything is applied in a short final project. Ideal for a person that is already familiar with basic Python and Pandas.

Real data is messy and this course is a result of this. You will learn to manage missing data, inconsistent formats, date parsing, character encoding problems and scalable workflows. It’s for analysts/aspiring data scientists and it’s focused on one of the parts of the job that is most often forgotten: reading and parsing. This course is for anyone who ever opened a CSV file and then said, “Oh! No!This is the course for those who have ever opened a CSV file and then said, “Oh! No!”.

In this course, you will learn how to develop a baseline model in the most efficient way.In this course, you will learn how to build a basic model in the most efficient manner.

For this reason, SQL continues to be important in the context of most business data residing within relational databases. This course provides an introduction to SELECT statements, filtering, grouping and joins with BigQuery examples. When you add Python to the mix, you’ve got a lot more of an analytical toolkit than most beginners care to develop. While many jobs might have a strong preference for candidates with experience in Python, it is not a requirement.Many jobs may have a strong preference for candidates who have experience in Python, but it is not a requirement.

After mastering basic queries, this course takes you deeper into data analysis with more advanced join and analytic functions and techniques that you will often encounter in an analytics career. It’s the transition between being able to write a SELECT statement, and being able to solve problems that involve SQL in the workplace, so don’t take it until basic SQL is comfortable.

This is a lot for many people’s first real taste of machine learning. It includes decision trees, model validation, underfitting and overfitting, random forests and is short enough that you can get a working ML model done before you’ve used up your entire weekend. A good book for someone who wants to move from data analysis to data science and the experience of submitting their first prediction to Kaggle is a worthwhile few hours in and of itself.

This one is about the messier, more realistic aspect of modelling: missing values, categorical variables, pipelines, cross-validation, XGBoost and data leakage. It teaches not just the names of the algorithms, but real modelling practices, where many ML courses fall short.

The often key to model performance is how you present your data, not which algorithm you choose. This course will cover the basics of creating features, transformations and categorical representations, and it’s a skills-heavy course that is directly tied to how well your models will perform when it comes to data prep.

Useful if your data is sorted in a time order. You’ll study time-series features, forecasting, trends, seasonality and validation techniques, and be able to start your forecasting career in sales, demand, finance and operations forecasting—an extremely valuable specialization to add to your resume.

The ability to communicate why a model predicted something is an important component of modeling as well as constructing the model. This course will teach you feature importance, permutation importance and partial dependence, as well as adding in interpretability skills that become important when you have to inform real decisions with your models.

Combines data science with location data. You’ll create interactive maps and master the skills to identify spatial patterns, which are actually very useful in logistics, real estate, environmental work and location-based business problems. A good way of separating your portfolio from the rest.

Your bridge between classical data science and neural networks, with a focus on TensorFlow and Keras. Discusses neurons, neural networks, stochastic gradient descent, dropout and batch normalization. Small enough to not require an entire specialization to learn the fundamentals of deep learning.

There’s more to accuracy than meets the eye. This one is about fairness, ethical decision making and responsible AI practices and it’s definitely worth doing no matter how good your technical program is, as it includes a professional dimension that isn’t taught by just any technical program.

Google’s crash course combines small bites of knowledge with visualizations and activities. It covers regression, classification, numerical and categorical data, neural networks, embeddings, production ML and fairness. It works best if you already have some experience of Python and manipulation of data.

Earning 25 points for completing this course will award you a course certificate.You will receive a course certificate when you complete this course, with 25 points.

This is not an ML-specific course but rather a comprehensive course on Analytics and the use of R from spreadsheets to SQL, data cleaning, analytics, visualisation, and case studies. This is a good tool for absolute beginners who prefer a step by step approach to a library tutorial instead of a career-oriented one.

IBM’s certificate strings together Python, SQL, data analysis, visualization, machine learning and projects into one sequence. Great for people looking to change their career but are not interested in random tutorials and sessions.

Very hands-on and heavily practice oriented, NumPy, Pandas, Matplotlib and projects. When you are a hands-on learner, this is great, as the projects you complete are also great portfolio items.

The course is designed with a programming emphasis: Python basics, data structures, algorithms and projects. If your understanding of Python is a bit shaky, this course is a good starting point before you enter Pandas or statistics courses, as it is the basis for the coding fluency those courses expect.

Prerequisites: This course covers the basics of web development and assumes no prior experience in coding.Audience: The course is designed for beginners to intermediate levels in web development with no prior coding experience.

Although R is still very important in the fields of statistics, research and analysis, if you prefer to do this course on the R-based approach, rather than the Python-first approach, this is a viable and alternative choice. Focuses on R programming, concepts of data analysis and statistics with real-life projects.