CS/INFO 6850 The Structure of Information Networks

SpringBoot+Vue+MySQL 高校学科竞赛平台 完整可运行版(源码+文档) 本项目基于SpringBoot+Vue+MySQL构建高校学科竞赛管理平台,覆盖学生、教师、领队教师、管理员四大角色,实现竞赛发布、报名、刷题、评分、公示等全流程闭环管理。系统架构清晰,技术主流,配套完整源码、数据库脚本、万字论文、开题报告、答辩PPT及部署文档,支持一键运行,适用于毕业设计、课程设计及院系二次开发,真正实现“开箱即用”。 阅读详情

The Structure of Information Networks

Homework 1

CS/INFO 6850 Fall 2024

Due 6pm, Wednesday, Sept. 18, 2024

The goal of this problem set is to provide practice implementing some basic network analysis techniques on a moderate-sized network dataset — specifically, a coauthorship network con-structed from a bibliography file of computer science papers maintained by Joel Seiferas at the University of Rochester. To make sure that the bibliography file is easy to obtain, a copy of it is has been posted on Canvas as the file “ps1data.txt”. As a first step, we will only be using papers written between the years 1985 and 2005 inclusive, so you should only keep lines in the file corresponding to papers written in this interval.

Building the coauthorship network. Here are some instructions on how to create the coau-thorship network from the raw bibliography. The short explanation is: each line represents a paper, and we want to build the undirected graph whose nodes are the people named in the bibliography, and whose edges join those pairs of people who’ve coauthored a paper in the bib-liography. This graph should only be constructed using papers written between the years 1985 and 2005 inclusive.

The more detailed instructions now follow. Each line in the bibliography describes a distinct paper, and has the following format:

year [number] conference/journal author & author & ... & author, title

Here, conference/journal is an acronym encoding the conference or journal where the paper appeared, year is the year of the paper, and number is the volume number of the journal or conference. We write number in brackets above because it is present in some lines and (when it is not known or not applicable) absent in others. Authors are given by last name only, and separated by the & symbol. The list of authors ends with a comma, and the the remainder of the line is the title. Thus, a sample line from the file is

1994 11 ALGRTHMICA Khuller & Naor, Flow in Planar Graphs with Vertex Capacities

encoding the paper “Flow in Planar Graphs with Vertex Capacities” by Khuller and Naor in Volume 11 of the journal Algorithmica. Finally (as with any list of records of this length), it is possible that a few of the lines in the file are misformatted.

You should start by only keeping lines for which the year is in the interval [1985, 2005] (including the two endpoints 1985 and 2005). From this set of lines, you should construct a coauthorship network as follows.

• There should be one node for each person. (Note that even if a person is an author on 50 of the papers listed in the bibliography file, there should still just be one node corresponding to him or her, not 50.)

• There should be an undirected edge between nodes A and B if and only if they are coauthors on a paper in the bibliography. (If they are coauthors on multiple papers, there should still just be a single edge joining them.)

For example, if the file consisted of just the three lines

1992 27 BAMS Alon & Kleitman, Piercing Convex Sets

1994 11 ALGRTHMICA Khuller & Naor, Flow in Planar Graphs with Vertex Capacities

1996 45 IEEETC Azar & Naor & Rom, Routing Strategies for Fast Networks

then the graph should have node set

{Alon, Azar, Khuller, Kleitman, N aor, Rom}

and edge set

{(Alon, Kleitman),(Azar, N aor),(Azar, Rom),(N aor, Rom),(Khuller, N aor)}.

Caveats. Before we move on to the problems themselves, here are three points worth mention-ing about the network we’re studying here.

(i) As we’ll see at various points in the course, coauthorship networks are a popular kind of “model system” for large-scale network analysis. This is not so much because there’s universal fascination with the coauthoring habits of scientists (though it’s an interesting topic that some people study as their research area), but because coauthorship networks are a kind of social network, encoding a particular type of collaboration among people, for which extremely rich and detailed data is available. As a result, it is a chance to try out network analysis techniques at very high resolution, in a setting that possesses many of the properties exhibited by much “messier” and harder-to-measure social networks as well.

(ii) Any time one tries to build a network from a file containing a list of names, there’s the concern that different people can have the same name, and hence these different people are being “merged” into a single node. This is definitely something to worry about when one tries to draw inferences about social structure from the resulting network. However, in our case, we are using this dataset simply to build an interesting graph on which to practice various analysis techniques, so for our limited purposes there’s no problem: if two authors have the same last name, then for us they are the same person.

(ii') In fact, because of the issue in (2), there are papers where someone appears to coauthor with themselves. We will omit from the network those edges that link some node to itself.

(iii) Point (ii) is a particular instance of a broader principle, that in building a network from everyday data such as this list of papers, some of the entries will be misformatted or contain idiosyncracies that conflict with the general assumptions we’re making about how the data is structured. For these unusual cases, it’s fine to make up a consistent set of rules for how you’re handling situations that don’t follow the expected structure, and to document the choices you made in doing this. The goal is to make choices that don’t have a significant effect on the result, where possible.

What to hand in

You should upload the following files to Canvas; please read this section carefully, since the format is important. In particular, for the first file, there is a specific line format we need, since we will be using scripts as part of the grading.

The files to hand in:

(1) An ascii .txt file named “hw1solution.txt”. This should have results for the questions below, with each line on which you are reporting part of the answer beginning with a “@”. The form. for these lines will be described in the questions below. The first line of this file hw1solution.txt should be just your NetID, as a single string.

(2) Four files named “plot1”, “plot2”, “plot3”, and “plot4”, containing the plots associated with Questions 1, 2, 3, and 4 respectively. These can be in any standard image format (e.g. plot1.png, or plot1.jpg, and so forth.).

(3) The source code you used to compute the answers. By default, we won’t be grading the quality of the code itself, but it will be useful to have it in case we run into any confusion. It is fine to use packages or software specifically designed for handling graphs, in which case you should just include what you wrote. If you answer the questions by some means where the notion of “source code” doesn’t exactly apply, such as an interactive session with a software package, then submit whatever analogue of source code we’d need to see how you answered the questions — for example, a script. you wrote as part of some larger existing package, or a transcript. of an interactive session in which you did it. (This file should be named “code” and can be in any format; if you need to bundle together multiple files, for example, it can be a zipped or gzipped folder or tar file.)

(4) A brief description of how to apply your code (or code analogue, or transcript) to the data, together with any decisions you made about how to handle the data that would be useful for us to know about. (This file should be named “explanation” and can be in any format. It’s this file that can address any decisions on data-handling that you had to make in accordance with general caveat (iii) above.)

Again, for most of the solutions, we’ll simply be evaluating (1) and (2), and only consulting (3) or (4) as background if necessary.

The Problems

(1) Recall that the degree of a node is the number of edges it’s incident to. We start by considering how the degrees of the nodes are distributed.

Thus, for a number j, let nj denote the number of nodes with degree exactly j. Let d ∗ be the maximum degree of any node in the network. (This is the maximum total number of co-authors that any one author has — the maximum j for which nj > 0.)

(a) For each j from 0 to d ∗ , output the number nj . Each of these should correspond to a line in the file hw1solution.txt with the following four fields

@ 1 j nj

(The second field here simply denotes that you’re answering the first question.) So for example, in the file above consisting only of the three lines

1992 27 BAMS Alon & Kleitman, Piercing Convex Sets

1994 11 ALGRTHMICA Khuller & Naor, Flow in Planar Graphs with Vertex Capacities

1996 45 IEEETC Azar & Naor & Rom, Routing Strategies for Fast Networks

the correct output would be

@ 1 0 0

@ 1 1 3

@ 1 2 2

@ 1 3 1

(Recall that before any of these lines, the first line of your file hw1solution.txt should be your NetID.)

(b) Produce a scatterplot in the plane of the ordered pairs (log j, log nj ) for those j such that both j > 0 and nj > 0. Hand this in as the file plot1. Later in the course, we’ll see some proposed explanations for why such scatterplots can often be approximated fairly well by a straight line.

(2) Now we consider the sizes of the connected components in the network.

(a) Let n* be the number of nodes in the largest connected component, and let n be the number of nodes in graph overall. Report these two quantities and their ratio, as a line in the file hw1solution.txt of the form.

@ 2 n* n n*/n

For example, on our sample graph above, you would report

@ 2 4 6 .667

Looking at the ratio of these two quantities is a good way to assess whether we should think of the network as having a “giant” component, or whether it consists entirely of small components.

(b) Let kj denote the number of connected components of size j, and let c* denote the size of the second-largest component. For each j from 1 to c*, output the number kj . Each of these should correspond to a line in the file hw1solution.txt with the following four fields

@ 2 j kj

For example, on our sample graph above, you would report

@ 2 1 0

@ 2 2 1

(c) Produce a scatterplot in the plane of the ordered pairs (log j, log kj ) for those j such that both 1 ≤ j ≤ c* and also kj > 0. Hand this in as the file plot2. The extent to which logarithmic plots of component sizes should look like straight lines is less heavily studied, but there is evidence for this as well.

(3) We next consider node-to-node distances in the largest component.

(a) We start by fixing the author name Gries (i.e. David Gries, an eminent faculty member of Cornell CS, who joined in its early years) as our “root node.” For each j, let rj denote the number of nodes at distance exactly j from Gries. (So r0 = 1, and r1 is equal to the degree of Gries.) Let s* denote the largest j for which rj > 0 — this is the farthest anyone in the bibliography is from Gries, yet still connected to him by a path.

For each j from 1 to s*, output the number rj . Each of these should correspond to a line in the file hw1solution.txt with the following four fields

@ 3 j rj

For example, on our sample graph above, if the starting node were Khuller (rather than Gries), you would report

@ 3 1 1

@ 3 2 2

(b) Produce a histogram that plots rj as a function of j, for j from 1 to s*. Hand this in as the file plot3.

(4) Finally, we continue the analysis from the previous question by considering the structure of breadth-first search trees. In a graph G, a breadth-first search tree rooted at a node r is simply a tree T that we build as follows. We first make node r the root of T. Then, for j = 1, 2, 3, ... in increasing order, we include in T all the nodes whose distance from r in G is equal to j. When we get to a value of j (meaning that we’ve already included all nodes at distance up to j − 1 in T), we consider each node v whose distance from r is equal to j. For each such v, we include it in T by making it the child of some node u at distance j −1 from r, where u and v are neighbors in G. Note that since there might be multiple choices of u for a given v, this means that there may be multiple breadth-first search trees rooted at r.

A breadth-first search tree thus satisfies the following properties.

(i) The root of T is r, and T contains precisely the nodes in r’s connected component.

(ii) Each edge of T is also in G.

(iii) For each node v in T, if the shortest-path distance from r to v in G is equal to j, then v is at depth j in T.

(a) Construct a breadth-first search tree T rooted at the node corresponding to Gries. Above, we noted why there can be many different breadth-first search trees with a given root: each time a node v at distance j from the root has multiple neighbors at distance j − 1, each of these neighbors is eligible to be the parent of v.

One way to quantify this variability is to count the number of options that nodes have for their parents. For a node v at distance j from the root, let’s say that u is a potential parent associated with v if u is at distance j − 1 from the root, and u is a neighbor of v in the graph G. Note that the parent of v in the breadth-first search is one of the potential parents associated with v.

For each j, let pj denote the average number of potential parents associated with nodes at distance j from the root node (in this case, Gries). That is, each node v at distance j from the root has a number of potential parents associated with it; we take the average of these numbers over all nodes v at distance j from the root.

Recall from the previous question that s* denotes the largest j for which there is some node at distance j from Gries.

For each j from 1 to s*, output the number pj . Each of these should correspond to a line in the file hw1solution.txt with the following four fields

@ 4 j pj

To construct an example with respect to potential parents, suppose we add two more papers to our sample input from earlier (to go along with the three we already have), so that the full input is now:

1992 27 BAMS Alon & Kleitman, Piercing Convex Sets

1994 11 ALGRTHMICA Khuller & Naor, Flow in Planar Graphs with Vertex Capacities

1996 45 IEEETC Azar & Naor & Rom, Routing Strategies for Fast Networks

1987 16 SICOMP Azar & Vishkin, Tight Comparison Bounds on the Complexity of

Parallel Sorting

1992 24 STOC Khuller & Vishkin, Biconnectivity Approximations and Graph Carvings

Now, if we construct a breadth-first search tree with Khuller as the root node, using the co-authorship graph built from these five papers (the two above and the three original ones), then the resulting tree will have Naor and Vishkin at depth 1, and Azar and Rom at depth 2. Naor, Vishkin, and Rom each have one potential parent in any breadth-first search tree rooted at Khuller, but Azar has two potential parents: Naor and Vishkin.

Thus, on this example input with five papers, the output would be

@ 4 1 1

@ 4 2 1.5

(b) Produce a histogram that plots pj as a function of j, for j from 1 to s*. Hand this in as the file plot4.

FM收音机调试-RDA5807M 收音机调试 IC型号:RDA5807M 阅读详情

相关推荐

【低空经济】无人机充电站设计方案

随着无人机技术的快速发展和广泛应用,其在物流、农业、巡检、救援等领域的应用日益普及。然而,无人机的续航能力一直是限制其大规模应用的关键因素之一。为了解决这一问题,无人机充电站的设计与建设显得尤为重要。本项目旨在设计一套高效、智能、可扩展的无人机充电站系统,以满足不同场景下无人机的充电需求,提升其作业效率和应用范围。无人机充电站的设计将围绕以下几个核心目标展开:首先,充电站应具备快速充电能力,以减少无人机的停机时间,提高作业效率;

2960

CS/INFO 6850 Fall 2024 The Structure of Information Networks Homework 1R

Homework 1@ 1 j nj@ 1 0 0@ 1 1 3@ 1 2 2@ 1 3 1@ 2 j kj@ 2 1 0@ 2 2 1@ 3 j rj@ 3 1 1@ 3 2 2@ 4 j pj@ 4 1 1@ 4 2 1.5。

qq__99515681的博客 944

手把手拆解DGX GB200 NVL72:1个机柜如何塞进72块GPU?详解NVLink Switch铜缆背板设计

本文深度解析英伟达DGX GB200 NVL72系统如何通过NVLink Switch铜缆背板设计,在标准42U机柜中集成72块B200 GPU,实现1800GB/s的超高带宽。文章详细介绍了模块化架构、铜缆互连技术、液冷散热方案和高密度供电系统等核心创新,为AI算力需求提供突破性解决方案。

red88的博客 1595

177.Agent-LangGraph-MCP的构成说明

MCP架构通过MCP主机、客户端与服务器的协同,实现工具、资源与提示词的统一管理。MCP服务器提供Tools、Resources、Prompts能力,MCP客户端则负责发现、调用与读取,使Agent可按需获取上下文。其核心在于将工具从Agent中抽离,封装为可复用的服务,提升灵活性与可维护性,推动AI应用高效协作。

计算机王的博客 315

爬取信用中国里面的行政处罚

result字段名为行政处罚文书号、处罚决定日期、处罚内容、处罚金额、违法事实、处罚机关、公司名称,行政处罚次数。file_path=r'E:\Data analyze\python\行政处罚爬虫/'#路径需要按实际修改。writer = pd.ExcelWriter('统计.xlsx')# 写入Excel文件。src_workbook=file_path+r'行政处罚(名单).xls' #表格名称也要更改。# keyword = input("输入爬取的公司全称必须是全称")

qq_42863025的博客 1677

system_stm32f10x.c文件分析

配置系统时钟(包括系统时钟源、锁相环(PLL)倍频系数、高级高性能总线(AHB)/ 高级外设总线(APBx)预分频器以及闪存相关设置)。该函数在系统复位后的启动阶段被调用,且调用时机早于跳转到主程序之前。

Jia shuheng的博客 604

Docker Compose:搭建本地多服务开发环境

从一个 Java API、PostgreSQL 和 Redis 开始,配置服务名、端口、环境变量、健康检查、数据卷和依赖关系。理解 Compose 文件如何被解析、项目如何隔离,以及哪些数据不能依赖“重建容器”保留,能够把个人开发环境整理成团队可复用的启动入口。

m0_62221377的博客 478

VSCode PlantUML 插件超详细教程:从安装配置到实战绘图全流程

VSCode PlantUML 插件超详细教程:从安装配置到实战绘图全流程

weixin_69237381的博客 911

AI全栈实战 | 3.9-01 爬虫入门:requests 加 BeautifulSoup 三分钟跑通,但 90% 的人栽在这 4 个“礼貌“问题上

⚠️:本文内容仅用于。请遵守目标站点的服务条款与相关法律法规,不得对目标服务造成负担,不得采集、传播个人隐私数据。技术是中性的,边界在你手里。爬虫的入门技术(requests 发请求 + BeautifulSoup 解析)三分钟就能学会。真正区分"会用爬虫"和"会做爬虫的人"的,是这篇我们跑通"请求→解析→存储"的最小闭环,然后重点把"合规与礼貌"这根弦给你绷上。

源码获取 | 运行部署 | 定制二开修改,如果回复不及时请耐心等待,请见谅。 442

CMOS 图像传感器——Skipping 和 Binning 模式

Skipping 和 Binning 模式

沧海一升的博客 6005

【四足机器人项目实战】四足机器人弹簧倒立摆(SLIP)动力学模型

系列文章目录 提示:这里可以添加系列文章的所有文章的目录,目录需要自己手动添加 TODO:写完再整理 文章目录系列文章目录前言1.动力学建模构型方法2.四足机器人动力学模型(1)多体动力学模型【针对躯干+脚建模】方法一:VMC(VirtualModelController)模型方法二:SLIP模型(2)浮基单体动力学模型【针对躯干建模】【用于MPC】 前言 认知有限,望大家多多包涵,有什么问题也希望能够与大家多交流,共同成长! 本文先对四足机器人动力学模型-VMC、SLIP和浮动机体模型做个简

qq_35635374的博客 7316

spring boot 4集成SQLLite,及其SQLLite的建库

spring boot 4集成SQLLite,及其SQLLite的建库

weixin_41463944的博客 231

【Flutter 问题系列第 16 篇】如何防止软键盘弹出后顶掉内容

Flutter 如何防止软键盘弹出后顶掉内容

“Allen Su”的博客 3317

【基于 WebSocket 的多用户网页五子棋】—— 数据管理模块

但行好事,莫问前程

qq_42996461的博客 260

WINCC7.3破解文件

WINCC7.3硬件够破解文件,有需要的下载!WINCC7.3硬件够破解文件,有需要的下载!

android 蓝牙通讯demo

android 下蓝牙通讯demo,实现了发现蓝牙,使能蓝牙,通过蓝牙发送消息。

OpenClaw 深度调查:把 AI 助手做成你自己的控制平面

对 openclaw/openclaw 的完整技术尽调:Gateway 架构、Agent 运行时、工具与沙箱、30 个消息通道、181 个插件、安全信任边界、工程代价与采用建议。17 章 / 10 张图,全部数字标注口径,数据截至 2026-10-31。

厚积薄发 349

大中型超市收银系统(可直接运行)

光步前台收银系统系统功能:商品销售、商品退货、挂单、收银员权限、会员销售、单品取消、整单取消、商品查询、脱机销售、脱机数据上传,基本资料下载,开钱箱、小票打印、收款结算,支持多币种付款方式(现金、储值卡、礼券、微信等),支持多种促销。运行环境:.net 4.0登陆信息用户名:1001 密码:123456内置商品信息,商品编码可输入00001000 - 00010000,测试效果相关功能键F5 结算 F9挂单 F11商品查询 F8退货 Insert单品取消 Del整单取消 F12开钱箱F2 会员 Ctrl+Z退出系统 Home 锁屏 相关源码待公开

IDEA 简介:从零认识 IntelliJ IDEA

IDEA 全称,是由 JetBrains 公司开发的一款主流 Java 集成开发环境(IDE)。它集代码编辑、编译、调试、测试、版本控制、构建管理等功能于一体,被大量 Java 开发者视为日常开发的首选工具。所谓 IDE(Integrated Development Environment,集成开发环境),就是把写代码所需的各类工具整合到同一个界面中。相比只用文本编辑器加命令行编译,IDE 能显著减少重复操作,让开发者把精力集中在业务逻辑上。

ztI10的博客 185

数学建模模型总结

这是本人在两年内准备参加数学建模竞赛搜集到的模型资料,已经经过了本人的筛选和整理,希望能够对准备参加数学建模竞赛的学弟学妹们提供一些帮助!

TransactionInactiveError:await 之后事务已经不活跃了

TransactionInactiveError 出现于 IndexedDB 事务已结束(非 active 状态)时调用 put、get 等操作。常见于 await 后或 setTimeout 中执行请求,因任务已完成,事务已提交或进入 finished 状态。请求必须在创建事务的同一任务中挂起,否则无效。解决方法:在 onsuccess 回调内同步发起后续操作,或提前将所有请求挂入事务。避免在异步上下文中使用事务操作,确保请求在事务活跃期内完成。

qq_40822000的博客 159

Docker部署linkding:HTTPS与数据备份

部署私人书签库,除了让页面能够打开,还需要考虑几个实际问题:重建容器后数据是否保留、登录和保存是否正常、升级前怎样备份,以及出现 502 或 CSRF 错误时从哪里检查。本文使用 Docker Compose 部署 linkding,采用默认 SQLite 数据库,通过 Caddy 提供 HTTPS,并完成账户创建、数据持久化验证和备份。如果重建后账户或书签消失,应先检查是否换了项目目录、改了挂载路径,或者启动了另一套实例,不要急着重新初始化。遇到权限问题,应根据日志确认具体目录和运行用户,再调整权限。

tedcloud123的博客 194

云计算课程小组PPT汇报任务书

36人分为6组,每组6人;形式:生活案例调研 + 理论解读;汇报时长7‑9min;PPT建议10‑14页。

Michelle8023的博客 226

企业官网源码与数据资产交接:四项权利、五组自动检查

企业官网源码交接不仅要看压缩包,还要核域名、云资源、构建脚本、数据库恢复、权限与密钥。

sweetofmine1的博客 187

软件项目交付时,源代码之外还应该交付哪些技术资料?

软件项目交付时,除可运行程序外,还应核对源代码、数据库结构、接口文档、部署运维、测试验收和账号资产。本文从实际交付环节拆解完整资料清单,并说明合同如何约定交付范围、知识产权与维护边界。

chengduyoushu的博客 308

大表 ALTER TABLE 怎么做?NineData OnlineDDL 配置与执行实践

介绍NineData OnlineDDL大表变更实践,涵盖执行策略、切表控制、拷贝批次与可视化进度追踪。

dbasenewbie的博客 295

Java 项目实战:交友项目后台系统——网关统一鉴权与 B 端管理后台

本次改造通过引入SpringCloud Gateway实现网关统一鉴权与路由,新建mf-admin管理端复用C端域服务(Dubbo调用),实现管理员登录与用户管理四件查询。核心套路:网关白名单+双体系JWT(同工具不同payload)、管理端零直连业务库、分页形态区分(MySQL自动count/Mongo手动count)、浏览器端独立聚合。踩坑12条,附自检清单与后续改造方向,可复用于多角色后台系统。

Wang的专栏 64

MySQL报Too many open files?句柄耗尽的排查与参数配置

文章讲清 max_connections 与 open_files_limit 两个额度、句柄被谁占走、分区表为什么最容易踩,给出从 error log 到 /proc/pid/limits 的定位顺序与句柄预算算法。

「一个在数据库赛道上边学边分享的学妹」——聚焦国产数据库/信创方向、分享学习笔记、踩坑经验和行业观察。 338

SQL Server 数据库双备份:简化配置,兼顾本地留存与异机归档

本文针对中小企业SQLServer数据库备份难题,提出简化配置与本地+异机双副本方案。传统原生备份依赖手动脚本,存在配置复杂、传输易中断、日志分散、静默失败等问题。推荐使用图形化工具实现备份、本地留存与异机同步一体化管理,支持断点续传、加密传输、自动清理与版本控制,降低运维负担。双副本策略提升数据安全性,本地快速恢复,异地防硬件故障。建议定期验证恢复流程,确保备份可用性。

2501_93776716的博客 313

Agent 安全实战:间接提示注入的四个攻击面与防御体系

当 Agent 从 "生成内容" 走向 "执行动作",安全边界就从内容合规变成了权限越界。本文拆解间接提示注入在网页抓取、工具返回值、MCP 连接器、长期记忆四个位置的真实攻击路径,对照 Claude、GPT、Gemini 在工具调用权限模型上的设计差异,给出可直接落地的五层防御体系。

阡陌数智,专注大模型网关与 LLM 工程落地 289
上一篇: FIT5137 M-Stay Residential service
下一篇: ITD102: Introduction to Computer Systems
wx__codinghelp
博客等级 码龄2年 514粉丝 61原创
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付:10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值