Free course:

The good news is: Lecture notes for a $ 2,000 online AI course are now open source!

Former Kaggle chief scientist and founder of fast.AI, Jeremy Howard, who has been working on AI education, released free benefits and announced his AI course lectures this spring at the University of San Francisco.

The notebook code for the course is now available and for free. Jeremy Howard posted the draft of the lecture notes on Github over the weekend. He gained 2k likes in two days and quickly reached the top of the daily trend list.

In addition, this project is also a draft of Jeremy Howard’s new book, which has not yet been officially released, which is equivalent to saving you another $ 60.

This book is his new book, Deep Learning for Coders with fast.AI and PyTorch, co-authored with Sylvain Gugger.

Although the book is currently in pre-order status, it is highly anticipated by readers and has long been ranked first on Amazon’s new computer graphics list.

Content

The draft of the book has been published in 22 chapters (including introduction and conclusion). The content naturally starts with the “Hello Word problem” in the AI world—MNIST image classification, and then NLP, recurrent neural network, convolutional neural network, and interpretability.

 

 

This course is not for zero-founders, the necessary Python and PyTorch knowledge is still needed.

To run the code in Notebook, the software you need to install is:

fastai v2、Graphviz、ipywidgets、matplotlib、nbdev、pandas、scikit-learn、Microsoft Azure Cognitive Services Image Search

They can all be installed directly via PyPI.

This fastbook is not only a textbook but also an AI community resource. In the final message, the author hopes that everyone who has completed this book will exchange successful experiences with everyone.

Finally, the book emphasizes the copyright issue of the project, because the project contains paid content for online courses and books, which cannot be copied and pasted at will.

The GPLv3 open source license only covers the code part of the project. As for the Markdown section in Notebook, it is not listed here. It can not be distributed or changed without permission. The project is also banned for commercial use.

If a copy of this code is hosted elsewhere, the author may be sued. Disregarding copyright provisions, the authors stated that other materials may not be considered for publication in this way in the future.

Therefore, we will not show the text and pictures in the project here. Interested friends can download it by themselves and use it as private learning materials.

Links:

Project: https://github.com/fastai/fastbook

Course: https://www.usfca.edu/data-institute/certificates/deep-learning-part-one

Original source from QbitAI

Follow FineReport Software on Facebook to get more data science news!