PyTorch Lightning, Lit-Serve and Lightning Studios, with Dr. Luca Antiga

Lightning AI makes tons of tools that speed A.I. model dev and deployment, including the wildly popular open-source library PyTorch Lightning. Today, hear from hands-on CTO Dr. Luca Antiga how all the magic happens ⚡️

More on Luca:

• CTO of Lightning AI, which (as one of world’s hottest startups developing A.I. tools) have raised over $80m in venture capital.

• Is also CTO of OROBIX, an A.I. services company that Luca co-founded 15 years ago.

• Holds a PhD in biomedical engineering from Politecnico di Milano… and did his postdoc at the Robarts Research Institute in London, Ontario (coincidentally around the same time I was doing brain-imaging research there).

Today’s episode will probably appeal most to hands-on practitioners like data scientists, software developers and ML engineers, but any tech-savvy professional could find it valuable.

In today’s episode, Luca details:

• How Lightning AI's suite of tools (in addition to PyTorch Lightning, this includes Lightning Studios, LitServe and the Thunder Compiler) is making A.I. development faster and easier.

• The rise of small language models and their potential to rival LLMs.

• His journey from biomedical imaging to deep learning pioneer.

• How software developer’s work will be transformed by A.I. in the coming years.

The SuperDataScience podcast is available on all major podcasting platforms, YouTube, and at SuperDataScience.com.

PyTorch: From Zero to Hero, with Luka Anicin

Today's episode is on Python's most popular auto-differentiation library, PyTorch, and how you can use it to design, train and deploy deep neural nets, including LLMs. Acclaimed PyTorch instructor Luka Anicin is our guide.

Luka:

• Is one of Udemy’s all-time bestselling instructors on A.I.; over 500,000 students have taken his courses.

• His latest course, available exclusively at SuperDataScience.com, is called “PyTorch: From Zero to Hero”.

• CEO of full-lifecycle A.I. consultancy Datablooz.

• Holds a Bachelor’s in Computer Science, a Master’s in Data Science and is nearing completion of his PhD in Applied A.I.

Today’s episode will probably appeal most to hands-on practitioners like data scientists, software developers and ML engineers.

In it, Luka details:

• What the popular Python library PyTorch is for.

• Why you would select PyTorch over TensorFlow or Scikit-learn.

• The tensor building blocks PyTorch provides for designing, training and deploying state-of-the-art deep neural networks, including Large Language Models (LLMs).

• His top tips for accurate and efficient deep learning.

• Guidance on PyTorch portfolio projects.

• Real-world PyTorch case-studies from his experience leading an A.I. consultancy.

The SuperDataScience podcast is available on all major podcasting platforms, YouTube, and at SuperDataScience.com.

Deep Reinforcement Learning — with Wah Loon Keng

For an intro to Deep Reinforcement Learning or to hear about the latest research and applications in the field (which is responsible for the most cutting-edge "A.I."), today's episode with Wah Loon Keng is for you.

Keng:
• Co-authored the exceptional book "Foundations of Deep Reinforcement Learning" alongside Laura Graesser.
• Co-created SLM-Lab, an open-source deep reinforcement learning framework written with the Python PyTorch library.
• Is a Senior A.I. Engineer at AppLovin, a marketing solutions provider.

In this episode, Keng details:
• What reinforcement learning is.
• A timeline of major breakthroughs in the history of Reinforcement Learning, including when and how Deep RL evolved.
• Modern industrial applications of Deep RL across robotics, logistics, and climate change.
• Limitations of Deep RL and how future research may overcome these limitations.
• The industrial robotics and A.I. applications Deep RL could play a leading role in in the coming decades.
• What it means to be an A.I. engineer and the software tools he uses daily in that role.

The SuperDataScience show's available on all major podcasting platforms, YouTube, and at SuperDataScience.com.

The Gradient of Mean Squared Error

For this week's YouTube video, we manually derive the gradient of Mean Squared Error (popular in ML to quantify accuracy). We then use the Python library PyTorch to obtain the same result and to visualize ML in action.

We publish a new video from my "Calculus for Machine Learning" course to YouTube every Wednesday. Playlist is here.

More detail about my broader "ML Foundations" curriculum and all of the associated open-source code is available in GitHub here.

Deep Learning Battle: Pytorch vs. Tensorflow

When I teach Deep Learning, the question I get most often is: "Should I be using TensorFlow or PyTorch?" In this recent talk at the DataScienceGO conference, I provide my most thorough and polished response yet.

Thanks to Harpreet Sahota for hosting the session masterfully and leading the audience Q&A at the end.

"Math for Machine Learning" course on Udemy complete!

After a year of filming and editing, my "Math for Machine Learning" course on Udemy is complete! To celebrate, we put together this epic video that overviews the 15-hour curriculum in under two minutes.

Over 83,000 students have registered for the course, which provides an introduction to all of the essential Linear Algebra and Calculus that one needs to be an expert Machine Learning practitioner. It's full of hundreds of hands-on code demos in the key Python tensor libraries — NumPy, TensorFlow, and PyTorch — to make learning fun and intuitive.

You can check out the course here.

O'Reilly + JK October & November ML Foundations LIVE Classes open for registration

We're halfway through my live 14-class "ML Foundations" curriculum, which I'm offering via O'Reilly Media. The first Probability class is on Wednesday with five of the seven remaining classes open for registration now:

• Oct 6 — Intro to Probability
• Oct 13 — Probability II and Information Theory
• Oct 27 — Intro to Statistics
• Nov 3 — Statistics II: Regression and Bayesian
• Nov 17 — Intro to Data Structures and Algorithms

The final two classes will be in December and are on Computer Science topics: Hashing, Trees, Graphs, and Optimization. Registration for them should open soon.

All of the training dates and registration links are provided at jonkrohn.com/talks.

A detailed curriculum and all of the code for my ML Foundations series is available open-source in GitHub here.

Calculating Partial Derivatives with PyTorch AutoDiff

My recent videos have detailed how to calculate partial derivatives by hand. In today's, I demo how we can compute them automatically using PyTorch, enabling us to easily differentiate complex equations like ML models.

We publish a new video from my "Calculus for Machine Learning" course to YouTube every Wednesday. Playlist is here.

More detail about my broader "ML Foundations" curriculum and all of the associated open-source code is available in GitHub here.

Machine Learning from First Principles, with AutoDiff

Today's brand-new, epic 40-minute YouTube tutorial ties together the preceding 27 Calculus videos to enable us to perform Machine Learning from first principles and fit a line to data points.

To make learning interactive and intuitive, this video focuses on hands-on code demos featuring PyTorch, the popular Python library for Automatic Differentiation.

If you're familiar with differential calculus but not machine learning, this video will make clear for you how ML works. If you're not familiar with differential calculus, the preceding videos in my "Calculus for Machine Learning" course will provide you with all of the foundational theory you need for ML.

We publish a new video from my "Calculus for Machine Learning" course to YouTube every Wednesday. Playlist is here.

More detail about my broader "ML Foundations" curriculum and all of the associated open-source code is available in GitHub here.

O'Reilly + JK October ML Foundations LIVE Classes open for registration

The Linear Algebra classes of my "ML Foundations" curriculum, offered via the O'Reilly Media platform, are in the rear-view mirror. Two Calculus classes are coming up soon and the Probability classes just opened for registration:

• Sep 15 — Calculus III: Partial Derivatives
• Sep 22 — Calculus IV: Gradients and Integrals
• Oct 6 — Intro to Probability
• Oct 13 — Probability II and Information Theory
• Oct 27 — Intro to Statistics

Overall, four subject areas are covered:
• Linear Algebra (3/3 classes DONE)
• Calculus (2/4 classes DONE)
• Probability and Statistics (4 classes)
• Computer Science (3 classes)

Hope to see you in class! Sign up opens about two months prior to each class. All of the training dates and registration links are provided at jonkrohn.com/talks

A detailed curriculum and all of the code for my ML Foundations series is available open-source in GitHub here.

The Line Equation as a Tensor Graph

New YouTube video today — it's meaty! In it, we get ourselves set up for applying Machine Learning from scratch by using the popular Python library PyTorch to create a Tensor Graph representation of a simple line equation.

Next week, we'll publish a massive 40-minute video that builds on the Tensor Graph representation introduced this week in order to use Automatic Differential Calculus within a Machine Learning loop and fit a Regression line to data points.

If you're familiar with differential calculus but not machine learning, this pair of videos will fill in all the gaps for you on how ML works. If you're not familiar with differential calculus, the preceding videos in my "Calculus for Machine Learning" course will provide you with all of the foundational theory you need for ML.

We publish a new video from my "Calculus for Machine Learning" course to YouTube every Wednesday. Playlist is here.

More detail about my broader "ML Foundations" curriculum and all of the associated open-source code is available in Github here.

AutoDiff with PyTorch

Over the past month, we've covered all the key rules for differentiating equations by hand. In today's YouTube video, we use PyTorch to differentiate equations automatically and instantaneously.

A new video for my "Calculus for ML" course published on YouTube every Wednesday. Playlist is here.

More detail about my broader "ML Foundations" curriculum and all of the associated open-source code is available in GitHub here.

Automatic Differentiation – Segment 3 of Subject 3, "Limits & Derivatives" – Machine Learning Foundations

Automatic Differentiation is a computational technique that allows us to move beyond calculating derivatives by hand and scale up the calculation of derivatives to the massive scales that are common in machine learning.

The YouTube videos in this segment, which we'll release every Wednesday, introduce AutoDiff in the two most important Python AutoDiff libraries: PyTorch and TensorFlow.

My growing "Calculus for ML" course is available on YouTube here.

More detail about my broader "ML Foundations" curriculum and all of the associated open-source code is available in GitHub here.

DataScienceGo This Weekend

The DataScienceGO conference is this weekend — registration for Friday and Saturday is 100% free! I'm speaking Saturday on the pros and cons of TensorFlow vs PyTorch for training and deploying deep-learning models.

Awesome speakers — whom you may already be familiar with from recent SuperDataScience episodes — include:
• Erica Greene (episode # 435)
• Harpreet Sahota (# 457)
• Andrew Jones (# 483)

I don't (yet!) personally know the other speakers pictured here but their weighty reputations precede them and I'm looking forward to getting to know them better over the course of the weekend: Gabriela de Queiroz, Karen JEAN-FRANCOIS, Yudan Lin, Ken Jee, and Danny Ma.

Free registration here!

TensorFlow vs PyTorch @ DataScienceGo Virtual

The DataScienceGO Virtual conference is coming up next Saturday and it is FREE! I'm giving a talk on TensorFlow vs PyTorch with lots of time for audience questions.

Upcoming O'Reilly Calculus Classes

Starting a week today, I'm offering my entire "ML Foundations" curriculum as a series of 14 live, interactive workshops via O'Reilly Media. The first five classes are open for registration; two are already waitlist-only, so grab a spot now:

• Jul 14 — Intro to Linear Algebra (waitlisted)
• Jul 21 — LinAlg II: Matrix Tensors (5 spots remaining)
• Jul 28 — LinAlg III: Eigenvectors (waitlisted)
• Aug 12 — Intro to Calculus (143 spots remaining)
• Aug 18 — Calc II: AutoDiff (148 spots remaining)

REGARDING THE WAITLIST: I have a made a request with O'Reilly to increase the maximum class size from 600 students to 1000, so if you sign up for a waitlisted class now, you should still be able to get in.

Overall, there will be four subject areas covered:
• Linear Algebra (3 classes)
• Calculus (4 classes)
• Probability and Statistics (4 classes)
• Computer Science (3 classes)

Sign up opens about two months prior to each class. All 14 training dates, running from next week through December, are provided at jonkrohn.com/talks

A detailed curriculum and all of the code for my ML Foundations series is available open-source in GitHub here.