hunkim/deeplearningzerotoall

DeepLearningZeroToAll is a collection of lab code files that accompany a deep learning tutorial series originally delivered in Korean and published on YouTube. The code is intended to help people learn the basics of building and training machine learning models using TensorFlow, with some labs also covering Keras and MXNet (two other tools in the same space). The files are numbered and named by topic, with a prefix indicating which framework they use: files starting with lab- are TensorFlow labs, klab- files are Keras labs, and mxlab- files are MXNet labs. The style throughout prioritizes being easy to read and understand over being computationally efficient, because the primary purpose is teaching, not production use. The README describes the project as a work in progress and is sparse on details about which specific topics each lab covers. Slide decks for the lectures are linked externally. The project accepts contributions and comments, and was intended at the time of writing to eventually produce an English version of the video series as well. This repository is most useful if you are following along with the original YouTube lecture series, since the lab files are keyed to those videos. As a standalone code collection, the README does not describe the full scope of what is covered, so someone arriving without that context would need to browse the individual files to understand what each one does. The license and specific TensorFlow version requirements are not stated in the README.