All in Deep Learning

This project is to help people get started with deep learning from scratch and further be able to apply deep learning techniques to their own research fields.

1. Set Your Deep Learning Environments

Start deep learning on local computer

The installation guide helps you set your deep learning environments on your local computer, in which the installation steps of Nvidia GPU driver, CUDA, TensorFlow, Keras, PyTorch, etc. are provided.

Start deep learning on GPU Cluster

The cluster tutorial helps you submit your deep learning tasks to the ustc GPU cluster.

This tutorial was originally provided by Zhanpeng Zhang @(https://github.com/Phosphenesvision) and Li Li. Some minor modifications were made by me to enhance the readability.

2. Deep Learning Tutorial with HYLee

If you want to know deep learning in a few hours, Hungyi Lee's slides One day to understand deep learning is a good choice!

3. Deep Learning Tutorial with CS231n

This is an awesome online free course which teaches deep learning in computer vision. If you are going to do research in computer vision area, this course suits you best.

• The course of Chinese version: http://www.mooc.ai/course/268

I think the best part of this course is its assignments. If you can finish it by yourself, you will have a deep understanding of deep learning.

4. Recent Advancements in Deep Learning

This part contains some of the recent advancements in Deep Learning along with codes for implementation in keras library. The links of the original papers are also provided in case you are interested in reading them.

The main content of this part is XXNet.md. XXNet is a review of classical deep learning network architectures from 2012 to the present, including

• A brief introduction of the frequently used deep learning models.
• An keras implementation for each model.
• The detailed analysis of each model based on its corresponding paper.
• Some applications based on these models.

5. Deep Learning with TensorFlow

TensorFlow is a framework that provides both high and low level APIs.

The O'Reilly's "Learning TensorFlow" is Great for beginners who are looking forward to implement your ideas with tensorflow as fast as possible.

The notes and code examples of this book have been implemented in the jupyter notebook.

6. Deep Learning with Keras

Keras is a high-level API capable of running on top of TensorFlow, CNTK and Theano. It has gained favor for its ease of use and syntactic simplicity, facilitating fast development.

This part aims to help you learn deep learning by practice. The tutorial was written based on the Chinese version of Keras official documents. Some theory background is added to understand and use the demo code more easily.

7. Deep Learning with PyTorch

Pytorch is a lower-level API focused on direct work with array expressions. It has gained immense interest in the last year, becoming a preferred solution for academic research, and applications of deep learning requiring optimizing custom expressions.

8. Deep Learning Research area

Some of the current popular deep learning research directions are introduced in this section, including

• Interpretability of neural networks
• Point cloud processing and Learning
• Model visualization
• Time series forecasting
• ...

9. Machine Learning

10. Reinforcement Learning

11. Interview Questions

Reference book