Get started with your machine learning journey for free
There are many ways to get started with studying machine learning. I have previously written a lot about how to design your own curriculum and roadmap as an alternative to taking courses. This approach allows you to pick and choose free, or low-cost, resources from across the internet that suit both your learning style and budget.
However, when you are just starting out on the beginning of your journey into machine learning it can often be useful to follow at least a short course that will guide you through the basic concepts first. This will give you a good foundational overview of the field and it will make it easier to design your own learning path and then continue on with deeper self-directed learning.
There are many machine learning courses available online. They range from short courses to much longer MOOCs (Massive Open Online Courses), and they can vary considerably in price.
I am an advocate for sharing free resources for learning and fortunately, there are several free machine learning courses available. In this article, I am going to share my three favourite free courses. They each have their own angle on machine and deep learning and contain a variety of content types, from videos to practical coding exercises, and it is, therefore, worth a beginner following all three.
If you are just getting started with studying machine learning I would recommend getting started with the courses covered in this post. They are presented here in the order I would recommend learning them in and I have also provided links to some additional resources that will give you the prerequisite knowledge needed for the last two courses.
From Cassie Kozyrkov
Length: 6.5 hours
Best suited for: Everyone
Core subject: Applied machine learning
Originally put together as an internal course at Google, this was released to the general public via YouTube in 2021. This is both a highly entertaining big picture introduction to applied machine learning and an exceptionally practical and beginner-friendly guide — understandable by anyone.
Rather than focussing on the theory or implementation details for machine learning it aims to give a high-level overview of the core concepts. This makes it digestible for anyone looking to get an overview of machine learning, not just technical people.
By taking this course you will gain an understanding of the end to end machine learning process, achieve an intuitive knowledge of the available algorithms including how they should be used and take a tour of some real-world machine learning use cases.
This is hands-down one of the best machine learning courses I have ever seen. It covers an introduction to all concepts and also covers common pitfalls and gotchas when applying machine learning in the real world. In my opinion, this course should be mandatory viewing for anyone working in or thinking of working in the field of applied machine learning!
If you are completely new to machine learning I would suggest starting with this course for your very first introduction. There are no prerequisites such as knowing how to code so it is an ideal place to get started. Once you have taken this course I recommend developing at least basic Python programming skills before moving on to the courses recommended in subsequent parts of this article. Codecademy is a great place to get started with learning how to program.
From Google
Length: 15 hours
Best suited for: Learners who can code with Python and already have a good understanding of linear algebra and statistics.
Core subject: Practical machine learning
This relatively short course covers an extensive breadth of machine learning topics. It is very hands-on, with much of the machine learning code focussed on TensorFlow APIs.
The course consists of 25 lessons each covering a specific area of machine learning. The content consists of a mixture of video lectures, written guides and practical exercises. The exercises are a combination of short tests to check your understanding of the concepts and hands-on programming on the Collaboratory platform.
The course covers much of the theory behind machine learning as well as its practical application. Topics covered include an introduction to TensorFlow, model training and evaluation, algorithms, model optimisation, generalization and a deep dive into neural networks. It also includes a series of case studies for machine learning and some useful content around problem framing that is rarely seen in these types of courses.
This is the perfect course to take after you have grasped the high-level concepts behind machine learning and have gained some Python programming experience. It is also helpful to have some understanding of linear algebra and statistics so if you need to brush up on these subjects check out some of the courses on khanacademy.org first.
From FastAI
Length: 7 weeks
Best suited for: Learners with at least one year’s coding experience
Core subject: Deep learning
Deep learning is a subset of machine learning and taking a course that will dive deeper into this field is a perfect follow-up to Google’s course described above. FastAI’s “practical deep learning for coders” is a great introduction to deep learning and neural networks in particular.
The creators of this course state that their aim is to “make deep learning accessible to as many people as possible”. It is primarily designed for learners who already know how to write Python code and focuses heavily on giving a practical introduction with guided Jupyter Notebooks available, alongside a video for each section. You should therefore take this course later in your learning journey once you have at least one year of programming experience.
This course has a great structure that starts with a high-level introduction to deep learning including the history, and then gently transitions into more complex topics. You will learn about how to train deep learning models, how to optimise them and how to put them into production.
More specialised areas of deep learning are also covered here with a lesson on collaborative filtering and one on natural language processing (NLP).
In this post, I have shared three free online courses for studying machine learning. Each of the three courses covers varying aspects of the field at a relatively high level acting as a perfect introduction.
If you are just getting started, following the material presented in this article will give you solid foundational knowledge which you can then build on with a customised learning pathway. To recap these courses should be taken as follows:
- Making friends with machine learning — a gentle, entertaining introduction to machine learning (no programming required)
- Machine learning crash course — a detailed end to end machine learning course which will teach you how to code machine learning models with Python
- Practical deep learning for coders — a fantastic, practical introduction to deep learning
Once you have taken these courses I would recommend taking a look at my posts, linked earlier in this article that will guide you in designing a complete learning pathway. There are many free, or very low-cost resources available online that will build on your knowledge of data science and machine learning.
If you are interested in finding out more about resources that are available I have also previously published a complete list of free materials that are widely available online.
Thanks for reading!
Get started with your machine learning journey for free
There are many ways to get started with studying machine learning. I have previously written a lot about how to design your own curriculum and roadmap as an alternative to taking courses. This approach allows you to pick and choose free, or low-cost, resources from across the internet that suit both your learning style and budget.
However, when you are just starting out on the beginning of your journey into machine learning it can often be useful to follow at least a short course that will guide you through the basic concepts first. This will give you a good foundational overview of the field and it will make it easier to design your own learning path and then continue on with deeper self-directed learning.
There are many machine learning courses available online. They range from short courses to much longer MOOCs (Massive Open Online Courses), and they can vary considerably in price.
I am an advocate for sharing free resources for learning and fortunately, there are several free machine learning courses available. In this article, I am going to share my three favourite free courses. They each have their own angle on machine and deep learning and contain a variety of content types, from videos to practical coding exercises, and it is, therefore, worth a beginner following all three.
If you are just getting started with studying machine learning I would recommend getting started with the courses covered in this post. They are presented here in the order I would recommend learning them in and I have also provided links to some additional resources that will give you the prerequisite knowledge needed for the last two courses.
From Cassie Kozyrkov
Length: 6.5 hours
Best suited for: Everyone
Core subject: Applied machine learning
Originally put together as an internal course at Google, this was released to the general public via YouTube in 2021. This is both a highly entertaining big picture introduction to applied machine learning and an exceptionally practical and beginner-friendly guide — understandable by anyone.
Rather than focussing on the theory or implementation details for machine learning it aims to give a high-level overview of the core concepts. This makes it digestible for anyone looking to get an overview of machine learning, not just technical people.
By taking this course you will gain an understanding of the end to end machine learning process, achieve an intuitive knowledge of the available algorithms including how they should be used and take a tour of some real-world machine learning use cases.
This is hands-down one of the best machine learning courses I have ever seen. It covers an introduction to all concepts and also covers common pitfalls and gotchas when applying machine learning in the real world. In my opinion, this course should be mandatory viewing for anyone working in or thinking of working in the field of applied machine learning!
If you are completely new to machine learning I would suggest starting with this course for your very first introduction. There are no prerequisites such as knowing how to code so it is an ideal place to get started. Once you have taken this course I recommend developing at least basic Python programming skills before moving on to the courses recommended in subsequent parts of this article. Codecademy is a great place to get started with learning how to program.
From Google
Length: 15 hours
Best suited for: Learners who can code with Python and already have a good understanding of linear algebra and statistics.
Core subject: Practical machine learning
This relatively short course covers an extensive breadth of machine learning topics. It is very hands-on, with much of the machine learning code focussed on TensorFlow APIs.
The course consists of 25 lessons each covering a specific area of machine learning. The content consists of a mixture of video lectures, written guides and practical exercises. The exercises are a combination of short tests to check your understanding of the concepts and hands-on programming on the Collaboratory platform.
The course covers much of the theory behind machine learning as well as its practical application. Topics covered include an introduction to TensorFlow, model training and evaluation, algorithms, model optimisation, generalization and a deep dive into neural networks. It also includes a series of case studies for machine learning and some useful content around problem framing that is rarely seen in these types of courses.
This is the perfect course to take after you have grasped the high-level concepts behind machine learning and have gained some Python programming experience. It is also helpful to have some understanding of linear algebra and statistics so if you need to brush up on these subjects check out some of the courses on khanacademy.org first.
From FastAI
Length: 7 weeks
Best suited for: Learners with at least one year’s coding experience
Core subject: Deep learning
Deep learning is a subset of machine learning and taking a course that will dive deeper into this field is a perfect follow-up to Google’s course described above. FastAI’s “practical deep learning for coders” is a great introduction to deep learning and neural networks in particular.
The creators of this course state that their aim is to “make deep learning accessible to as many people as possible”. It is primarily designed for learners who already know how to write Python code and focuses heavily on giving a practical introduction with guided Jupyter Notebooks available, alongside a video for each section. You should therefore take this course later in your learning journey once you have at least one year of programming experience.
This course has a great structure that starts with a high-level introduction to deep learning including the history, and then gently transitions into more complex topics. You will learn about how to train deep learning models, how to optimise them and how to put them into production.
More specialised areas of deep learning are also covered here with a lesson on collaborative filtering and one on natural language processing (NLP).
In this post, I have shared three free online courses for studying machine learning. Each of the three courses covers varying aspects of the field at a relatively high level acting as a perfect introduction.
If you are just getting started, following the material presented in this article will give you solid foundational knowledge which you can then build on with a customised learning pathway. To recap these courses should be taken as follows:
- Making friends with machine learning — a gentle, entertaining introduction to machine learning (no programming required)
- Machine learning crash course — a detailed end to end machine learning course which will teach you how to code machine learning models with Python
- Practical deep learning for coders — a fantastic, practical introduction to deep learning
Once you have taken these courses I would recommend taking a look at my posts, linked earlier in this article that will guide you in designing a complete learning pathway. There are many free, or very low-cost resources available online that will build on your knowledge of data science and machine learning.
If you are interested in finding out more about resources that are available I have also previously published a complete list of free materials that are widely available online.
Thanks for reading!