I have recently started to learn deep learning. Generally speaking, deep learning is divided into two parts, one is the perceptron function, and the other is the neural network model. If you don't want to dig through the textbook, the following tutorial is considered a conscience, from the perceptron to the common network models such as: RNN, LSTM, etc. are introduced, and the python implementation code of its GitHub is attached. And revealed the differences, advantages and disadvantages between different models, basically there is a full range of derivation for the mathematical formula of each model. So I won't go into details here. I hope that the children's shoes who want to get started with deep learning can click the link below to open the door to a new world. (There are only 7 chapters, but most of the commonly used models have been covered, but the following requires a certain high-level foundation, and other high-level and matrix calculations need to have a certain knowledge reserve, but the difficulty is not very big, for Students who are not interested in deriving the formula can also try it out)
Zero Basic Introduction to Deep Learning
Click to open the link