The Top Data Scientists to Follow and Best Machine Learning Tutorials on GitHub

GitHub has become the heart of the data science community. It‘s where the top experts and practitioners share their work, collaborate on cutting-edge open source projects, and compile incredible resources for learning and applying data science, machine learning, AI, and analytics.

As of 2023, over 100 million repositories and more than 30 million users call GitHub home, many of them in data-related fields. Nearly all of the most popular and innovative data science tools are developed openly on GitHub – libraries like NumPy, pandas, SciPy, scikit-learn, TensorFlow, PyTorch, Keras, and beyond. Following the individuals and teams behind these transformative projects gives you insights you can‘t get anywhere else.

At the same time, many of these experts use GitHub to collect, curate and share educational resources. From hands-on tutorials to in-depth topic guides to annotated paper collections, GitHub has become a go-to platform for learning data science. High-quality content is posted daily, saving you the time and effort of sorting through the endless sea of articles, videos, and courses available.

In this post, we‘ll highlight some of the most influential data scientists to follow on GitHub right now, as well as the best tutorials, courses, and educational repositories they‘ve developed. You‘ll also learn tips for getting the most out of GitHub yourself to accelerate your data science learning and career.

Why GitHub is an Essential Platform for Data Scientists

What makes GitHub uniquely valuable for data science compared to other platforms? Let‘s look at some key reasons:

1. Discover and contribute to open source projects. The open source community on GitHub is deeply intertwined with data science. Following the top repositories keeps you informed on the latest developments in foundational tools. Many welcome contributions, giving you the opportunity to hone your skills, network, and build an impressive public portfolio.

2. Learn by example from experts. Data scientists of all levels, from beginners to world-class practitioners, share notebooks, scripts, tutorials, and projects showing their process and techniques. Seeing how experts approach and solve problems with real code is one of the best ways to grow your own capabilities.

3. Keep up with the latest trends and techniques. Data science is an incredibly fast-moving field. New approaches, algorithms, and tools are constantly emerging. Data scientists frequently announce and discuss the latest advancements on GitHub. Following the top voices helps you stay informed and identify which skills to learn next.

4. Extend your knowledge into new domains. Increasingly, data science is not a singular field, but rather the application of tools and techniques to specific domains. Whatever your area of interest, from astronomy to zoology, there is a community of data scientists applying the latest approaches and sharing their work on GitHub.

5. Access high-quality educational resources. A growing number of data scientists are creating and compiling exceptional learning resources on GitHub – curated lists of tutorials, papers, courses, books, and more. These are often more up-to-date and hands-on than textbooks or traditional curriculums.

As data science and AI continue their rapid advancement and integration across industries and domains, staying connected to the heart of the community via GitHub will only become more valuable and essential.

How to Use GitHub Effectively as a Data Scientist

If you‘re new to GitHub or looking to get more out of the platform, here are some tips:

Create a strong profile.

Your GitHub profile is like a public resume showcasing your skills, projects, and contributions. Take the time to write a clear bio, pin your best repos, and keep your activity up-to-date. Highlight your areas of expertise and the types of projects you‘re interested in.

Actively engage with others.

Don‘t just follow and read, but participate! Ask questions, report bugs, suggest new features, and submit pull requests to interesting projects. This is how you build relationships and learn collaboratively. Make your expertise available to others as well by responding to issues and mentoring new contributors.

Keep your code and projects organized.

Use informative READMEs, descriptive commit messages, and clear documentation. Not only will this help other people understand and potentially contribute to your work, but it demonstrates positive habits to potential collaborators and employers.

Build a portfolio of your work.

In addition to contributing to existing projects, create and share your own! Open source your personal projects, notebooks, and experiments. A demonstrated history of technical work and growth is a huge asset, especially if you‘re earlier in your career or transitioning into data science.

Manage your learning with collections.

Star interesting repositories and organize them into collections by topic for quick reference. This could include both educational resources as well as tools and projects you want to explore more deeply. Curating your own go-to learning library will streamline your growth.

Now let‘s look at some of the best data scientists to follow and educational repositories to focus on.

Top Data Scientists to Follow on GitHub

These experts are not only advancing the field, but actively sharing valuable knowledge and perspectives:

1. Hadley Wickham (@hadley) – Chief Scientist at RStudio and creator of many of the most widely-used R packages like ggplot2, dplyr, and tidyr. Hadley‘s repositories offer master classes in data wrangling, visualization, and tool-building. He‘s also a speaker and author focusing on data science education and best practices.

2. Sebastian Raschka (@rasbt) – Experienced machine learning researcher and educator. Author of the bestselling book "Python Machine Learning" and creator of popular tutorials and resources. Sebastian frequently shares cutting-edge techniques with detailed walkthroughs and code. A perfect follow for staying current with the latest in applied ML.

3. Cassie Kozyrkov (@kozyrkov) – Chief Decision Scientist at Google, Cassie is one of the best communicators in data science. Her repositories contain highly accessible tutorials and guides on applied statistics, machine learning, and data-informed decision-making. She excels at explaining complex topics for both technical and non-technical audiences.

4. Rachel Thomas (@racheltho) and Jeremy Howard (@jph00) – Co-founders of fast.ai, which offers free, cutting-edge courses on deep learning. Both Rachel and Jeremy are highly respected practitioners and educators. Their repositories feature not just the fast.ai course materials, but also key papers, insightful analysis, and helpful resources on effectively learning and applying ML and data science.

5. Chris Albon (@chrisalbon) – Director of Machine Learning at Wikimedia, Chris is well-known for creating highly practical tutorials and educational resources. His repositories, like his Machine Learning Flashcards, focus on developing actionable skills and understanding. He‘s an excellent follow for leveling up your day-to-day capabilities.

6. Wes McKinney (@wesm) – Creator of the pandas library and author of "Python for Data Analysis." Wes is a leader in the PyData ecosystem and frequently shares thoughts on data engineering and analysis best practices. His work is especially helpful for understanding how to build scalable data pipelines and work efficiently with large datasets.

7. Kimberly Fessel (@kimfetti) – Data scientist and developer advocate at Weights & Biases, Kimberly creates exceptional content on machine learning best practices, experiment tracking, and model evaluation. Her repositories include hands-on tutorials and insightful overviews of key ML concepts. A great follow for learning how to build reliable, production-ready models.

A few other notable data scientists to follow include François Chollet (@fchollet), creator of Keras, Thomas Wolf (@thomwolf), co-founder of Hugging Face and NLP expert, and Andreas Mueller (@amueller), core contributor to scikit-learn.

Best Machine Learning & Data Science Tutorials on GitHub

Beyond following individual experts, some of the most valuable educational content on GitHub are curated collections and in-depth tutorials created and maintained by the community. Here are a few of the best:

1. Awesome Data Science (@academic/awesome-datascience) – Perhaps the most comprehensive collection of data science resources on GitHub. Includes sections on bloggers, newsletters, podcasts, cheat sheets, online courses, degree programs, books, tools, and much more. An incredibly thorough overview of the field and how to learn it.

2. Machine Learning for Software Engineers (@ZuzooVn/machine-learning-for-software-engineers) – Specifically oriented towards developers looking to transition into machine learning and data science. Provides a suggested curriculum and links to key resources for each topic. The focus on succinct, actionable steps makes it a great jumpstart guide.

3. Machine Learning Course (@machinelearningmindset/machine-learning-course) – A well-structured, slides-based introduction to machine learning covering essential theory and key algorithms. The associated Jupyter notebooks allow you to implement and experiment with the concepts yourself. Valuable for gaining a solid foundational understanding.

4. Dive into Machine Learning (@hangtwenty/dive-into-machine-learning) – A carefully crafted path through the most essential machine learning skills and techniques, including recommended resources and hands-on Jupyter notebooks. Especially good for Python developers looking for a thorough but efficient introduction.

5. Machine Learning from Scratch (@eriklindernoren/ML-From-Scratch) – Implementations of essential machine learning algorithms in bare Python code, without using libraries like scikit-learn. An excellent way to truly understand what‘s happening "under the hood" and build your technical foundations.

6. Homemade Machine Learning (@trekhleb/homemade-machine-learning) – Similar to ML From Scratch, this repository features Python implementations of dozens of ML algorithms with detailed explanations and links for further learning. Helpful for grasping the core concepts behind the most important techniques.

7. 100 Days of ML Coding (@Avik-Jain/100-Days-Of-ML-Code) – A guided journey through essential machine learning concepts, algorithms, and projects, with clear code examples and explanations. Great for committing to consistent practice and tracking your progress over time.

Remember, the real power of these resources comes not just from reading them, but from engaging with the material deeply. Run the code, experiment with modifications, and extend the techniques to your own projects and datasets. Hands-on application is key to transforming knowledge into practical skills.

The Future of Open Source and GitHub in Data Science

The open source ethos and resources like GitHub have already had a profound impact on the development of data science and machine learning. The most powerful tools and frameworks are created collaboratively and released freely for the community to use, inspect, and extend. Research and novel techniques are frequently published first as open source code and interactive notebooks rather than in traditional journals.

This trend shows no signs of slowing down. If anything, data science and open source will continue to intertwine more deeply:

• Shifting incentives. As data literacy becomes essential across fields, the incentives for keeping work proprietary diminish. Sharing knowledge and tools openly leads to more rapid innovation and "hive mind" problem-solving. Especially in research and education, success is increasingly defined by open contributions rather than gated access.

• Access and equity. Open source code and educational resources lower the barriers to entry for data science. Learners and practitioners from all backgrounds can access top-tier tools and training without steep financial burdens. This allows a larger and more diverse pool of talent to participate in the field‘s advancement.

• Reproducibility and trust. Especially in high-stakes domains like healthcare, open source tools and methodologies are essential for verifying results and building trust with stakeholders. When code is available for inspection, issues can be identified and improvements made transparently and collaboratively.

• Rapid experimentation. The ability to quickly test and iterate on ideas is crucial in a fast-moving field like data science. Open source tools and the ability to fork and modify code on GitHub accelerates research and product development cycles. Approaches can be evaluated and refined rapidly by the community.

As these trends progress, staying plugged into the open source data science community via GitHub will be increasingly vital for practitioners at all levels. The platform is where the latest advancements surface first and where the most impactful tools are forged through collaborative effort.

Conclusion

We‘ve explored why GitHub is an essential hub for modern data science and how you can get the most out of the platform by:

1. Following influential experts advancing the field
2. Watching and contributing to powerful open source projects
3. Studying and engaging with the highest-quality learning resources

By plugging into the GitHub data science community and participating actively, you can stay on the cutting edge of new techniques and tools while also honing your skills and building a body of work. You‘ll gain not just technical know-how, but also the collaborative, open-oriented mindset driving the future of data science forward.

To help you get started, we highlighted some of the top data science experts to follow on GitHub today, as well as several of the best educational resources and tutorial collections. Remember to focus on hands-on application and experimentation, not just passive consumption.

Don‘t be afraid to dive in, make mistakes, and learn by doing. The supportive open source community is there to help you grow. Follow the experts shaping the future of the field, master the best tools and techniques available today, and most importantly, share your own learnings and projects back to the community. That‘s how you can make the leap from data science student to influential practitioner.

As Vicki Boykis (@vboykis), machine learning engineer and open source advocate, puts it: "The beauty of open source is that it‘s accessible by anyone, from anywhere. You can be a student in a developing country, or a busy parent squeezing in learning time after day job and family duties. If you have an internet connection and a willingness to dig in, you can gain data science skills and contribute meaningfully, often more efficiently than through traditional paths. Open source contributions are a ticket to building in-demand expertise while doing fulfilling work that makes an impact on the world."