English | 简体中文
This repository is a companion code to the book "Fundamentals and Applications of Artificial Intelligence Techniques" written by VSISLAB
Chapters 1 do not involve practice, no supporting code
This chapter mainly introduces the basic design of neural networks, using numpy to implement commonly used modules in neural networks from scratch, and finally building a network that can recognize handwritten digits
This chapter mainly introduces the PyTorch deep learning computing framework and uses PyTorch to implement the network in Chapter 2 in a more concise way.
This chapter introduces the classical convolutional neural network structure, the dataset used for training and testing is FashionMNIST, which will be automatically downloaded to the directory chapter_4/data/FashionMNIST when building Dataloader
Sample in FashionMNIST
This chapter introduces the classical sequence-to-sequence network, the dataset used for training and testing is aclImdb, which needs to be downloaded to the directory chapter_5/data/aclImdb by the reader.
positive and negative review samples
This chapter implements a practical case of human detection, the dataset used for training and testing is Penn-Fudan, there is download scripts in jupyter notebook
human detection
This chapter implements a practical case of urban street segmentation, the dataset used for training and testing is CamVid, which needs to be downloaded to the directory chapter_7/CamVid.
original image
output of semantic segmentation