摘要： 一文详解卷积与逆卷积 从cnn到mask-rcnn到yolo, part1 图卷积神经网络GCN详解part3 VAE详解(参杂AE和PCA) AAE(Adversarial auto-encoder Contrastive learning 阅读全文

摘要： Convolutional layer with kernel size of 1 * 1 can be regarded as a fully connected layer on channels. 阅读全文

摘要： Notes about Jinja2 Inheritence in jinja2: If a child html file extends a parent html file, and {% extends 'parant.html' %} is written, when the Jinja 阅读全文

摘要： Pytorch Notes Be aware of the tensor shape. The shape of torch.tensor([1]) and torch.tensor(1) is different. One is torch.Size([1]), the other is torc 阅读全文

摘要： Debug Xv6 Kernel with GDB, QEMU and VScode The launch.json is as follows: { "version": "0.2.0", "configurations": [ { "name": "C/C++: gcc 生成和调试活动文件", 阅读全文

摘要： Open a file: :e /path/file Tag multilines: In visual mode, press shift + .. 阅读全文

摘要： Risc-V Xv6 Syscall Procedure User syscall function loads corresponding syscall number into register a7, then execute ecall. What ecall does in risc-v 阅读全文

摘要： gcc has some useful tools to help us find bugs. Example: gcc -lpthread -fsanitize=thread #check data race gcc -fsanitize=address #check memory error A 阅读全文

摘要： Makefile Notes 1. Core Idea of Makefile The structure of a typical makefile is as below: target: dependency1 denpendency2... command ....... When we t 阅读全文