Learn by Source
IBM · on edX

PyTorch Basics for Machine Learning

Beginner 15 hours English Course CertificateFREE
SkillsPyTorchTensorsAutomatic differentiationMachine learningPythonModel building

Is this course right for you?

Our take
The first half of IBM's two-course PyTorch sequence on edX, covering the fundamentals: tensors, automatic differentiation, and building your first simple models.

Good for: A hands-on start with PyTorch fundamentals.

Skip if: You cannot code Python or want the deep-learning material directly.

It's the on-ramp before the deep-learning course that follows, so it's technical and practical but kept deliberately foundational — right for people who can code Python and want to start with PyTorch properly, and not where to go if you want the deep-learning material directly. You can audit it free on edX, with a verified certificate as a paid option (financial assistance available). PyTorch is a stable, mainstream framework, so these basics stay current (as of 2026).

Comparison · LBS

Compare alternatives for PyTorch Basics for Machine Learning

Same topic, different options. We surface the trade-offs others hide so you can pick the course that actually fits your time, budget, and goals.
edX
PyTorch Basics for Machine Learning
Price
Free
Audit free · paid verified certificate
Duration
15 hrs
Level
Beginner
Certificate
Course Certificate
freeCodeCamp4.7(5,800)
Machine Learning with Python Certification
Price
Free
Completely free, forever
Duration
300 hrs
Level
Intermediate
Certificate
Course Certificate
View course →
Google4.7(8,500)
Machine Learning Crash Course
Price
Free
Completely free, self-paced
Duration
15 hrs
Level
Beginner
Certificate
View course →
edX
Deep Learning with Python and PyTorch
Price
Free
Audit free · paid verified certificate
Duration
20 hrs
Level
Intermediate
Certificate
Course Certificate
View course →
Prices & availability can change — confirm on the provider's site. We're not affiliated with any single provider.

About this course

PyTorch Basics for Machine Learning is the first half of IBM's PyTorch sequence, covering the fundamentals: tensors, automatic differentiation, and building your first simple models. It's the on-ramp before the deep learning course that follows it, and it keeps the scope small on purpose so tensors and gradients click before you add network architectures on top.

Browse all AI & ML courses Browse all edX courses

Instructor

I
IBM
edX instructor

Taught by IBM's data science and AI training team.

Frequently asked questions

It is a foundational, first-level PyTorch course from IBM — introducing the framework's basics: tensors, building simple models, and the core mechanics of training in PyTorch. It typically sits early in IBM's deep-learning path, before more advanced PyTorch and neural-network courses. So treat it as the entry point into PyTorch, with deeper networks and applications coming in the courses that follow it.

Yes. It teaches machine learning with PyTorch in Python, so you need comfortable Python fundamentals before starting. It is not a Python primer. Some prior grasp of basic machine-learning ideas helps too, since it moves into building models, but the essential prerequisite is being able to write and read Python confidently enough to focus on PyTorch rather than syntax.

Within IBM's deep-learning path, it leads into more advanced PyTorch courses — deeper neural networks, and applications like computer vision or sequence models — building toward a fuller deep-learning skill set. You can take this basics course alone for a PyTorch introduction, but it is designed as the first step, so the natural progression is the more advanced courses that follow it.

You can audit it on Coursera (or access via edX depending on the listing) to watch the material at no cost, with graded labs and the certificate behind a subscription. Since the hands-on coding is where PyTorch skills develop, the paid track is worth considering if you want to actually build the models rather than only follow the lectures.

It is a focused foundational course — typically a few weeks at a modest weekly pace, faster with solid Python. It is hands-on, so the time goes into working through the PyTorch coding exercises, with the wider IBM deep-learning path a considerably longer commitment if you continue through the more advanced courses.
Free
to audit
Enroll now