We have used these resources, and do not receive any compensation for recommending them

It is better to pick one or two resources and learn them well, rather than jumping between the latest and greatest resources

Table of Contents

Websites
Books
Video Courses
Deep Learning
Miscellaneous

---

Websites

• Air Force Digital U
  • The Air Force has partnered with Udemy to create Digital U, which provides curated training plans for different careers and skills.
  • Career options include: data scientist, machine learning engineer, project manager, and more.
  • Skill options include: AI, CSS, computer science, Agile methodology, and many more.
  • Each career or skill training plan includes numerous Udemy course offerings organized to build the foundations needed to get going and then buid expertise over time.
• Towards Data Science, a Medium blog with endless tutorials and code examples
  • Medium has a $5/month charge for unlimited articles. Some of us find found this cost incredibly worth the value for how much we use TDS.
• Machine Learning Mastery, a blog by Jason Brownlee, has great tutorials
• Kaggle, a data science competition platform, has lots of great data sets and notebooks sharing users’ problem-solving process and code
• Codeacademy, free coding classes for a ton of different languages. Great for brushing up on syntax if you need to switch languages.
• Twitter has a large data science community
  • The DL/ML Twitter Loop has some good ML people to follow
  • Santiago has informative threads on ML and Python

---

Books

Free

• Approaching (Almost) any Machine Learning Problem by Abhishek Thakur (4x Kaggle Grandmaster)
• Machine Learning Systems Design by Chip Huyen
• Data Engineering Cookbook by Andreas Kretz
• PyData Book by Wes McKinney (free companion to his paid book)
• Python Data Science Handbook by Jake VanderPlas

Paid

(O’Reilly books are free for Service Members)

• Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython by Wes McKinney
• Hands-On Machine Learning With Scikit-Learn and Tensorflow by Aurélien Géron
• The Analytics Edge by Allison K. O’Hair, Dimitris Bertsimas, William R. Pulleyblank

---

Video Courses

Free

• Practical Deep Learning For Coders by FastAI (Jeremy Howard + Rachel Thomas)
  • Uses fastai library, which is a mid/high level PyTorch library
• Free courses for AWS EdX
• Stanford ML Courses class notes available in PDFs, videos sometimes on youtube

Paid

Udemy (courses often list for $100, but often drop to < $20)

• Python for Data Science and Machine Learning Bootcamp by Jose Portilla (Jose offers many great courses on time series, NLP, etc)
• AWS Solutions Architect and Developer A Cloud Guru has good Udemy courses

Coursera (pay only if you want the certificate at the end)

• Andrew Ng’s ML course is famous
• Andrew Ng’s Deep Learning Specialization Courses

---

Deep Learning

• Dive Into Deep Learning (free + code snippets in Pytorch/TF + nbs + github)
• Deep Learning Book - the original DL book. $49 on amazon, free on web

Keras

• Keras Code Examples for almost any use case
• Keras Recipes
• Intro to Keras for Engineers
• Intro to Keras for Researchers

Pytorch

• Deep Learning with Pytorch (pdf book)
• Awesome Pytorch Examples (all problem types)
• PyTorch Tutorial Github

---

Newsletters

• Import AI by Jack Clark, Strategy and Communications Director at OpenAI
• The Batch by deeplearning.ai and Andrew Ng
• ChinAI Newsletter by Jeff Ding
  • “Weekly translations of writings from Chinese thinkers on China’s AI landscape”

---

Miscellaneous

• ML technical debt paper (Google)
• Causality and ML. For the curious: causality is a new field of study that deals with the lack of causal representations in the field of ML. Yes, it’s the “causation” we’ve been avoiding from the “correlation does not imply causation”. Check out this talk and paper from Yoshua Bengio (AI pioneer) and this article explaining.