[Codechef]2014 December Challenge
由于这个月有两个招先哥和一个乱搞哥就准备混个前10所以后来challenge懒得刷了
刚开始一个小哥把spj爆掉了,分数直接比乱搞哥的两倍还高。。。当时不敢相信就觉得他是把spj爆掉了。。。后来rejudge了才发现真的是把spj爆掉了。。。那个小哥以为是0-indexed,然后用0和1~n-1连边。。。spj居然没判出来。。。
SANSKAR 我直接[tex]2^{2^{21}}[/tex]爆搜的。。。听说直接搜不回溯都能A。。。无语了。。。
SEAGCD 我直接[tex]n\log n[/tex]暴力容斥+大力常数优化过的。。。
scanf("%d%d%d%d",&n,&m,&L,&R); rep(i,1,m){ if(i-1&&m/i==(m/(i-1)))f[i]=f[i-1];else f[i]=powmod(m/i,n); } dep(i,m,1){ if(m/i==m/(i+1))f[i]=f[i+1]; else{ for(int j=i+i;j<=m;j+=i)f[i]-=f[j]; } } ll ans=0;rep(i,L,R)ans=(ans+f[i])%mod; printf("%lld\n",(ans+mod)%mod);
不会更优的算法。。。
RIN 详见陈高远TC原题。网络流一遍没了。详见CF248D和SRM590Hard。
DIVIDEN 首先你得画出3°,大概是用36°和30°搞。然后所有不是3的倍数的都可以画。扭一下坐标系的话会很好写。
GOODGAL 不明觉厉的题。感觉非常傻逼但是过的人居然这么少。直接判每个点的出度和随机选两个点判和它们都相邻的点的个数是不是等于应该等于的值即可。虽然不会证明正确性但是好像不好卡。
KALKI 一开始想了很多种算法都打不过最小生成树,后来睡觉前突然想到每次选边的时候估价一下,每个当前值为x的时候就加上[tex]x^2[/tex]的代价。不过这样复杂度有点高,每次都要扫一遍所有边。所以每个点只取最近的20条边。然后卡着时限能过。这样已经够艹掉shizhouxing了。然后发现把[tex]x^2[/tex]改成[tex]x^8[/tex]直接和tankengineer一样高了。然后加了一些break优化改成计算多次好像效果也没好多少然后就弃疗了
2022年8月30日 00:54
Primary School Certificate Students of Dinajpur Board can get their PSC Result 2022 with full mark sheet available from the last week of December 2022 at DPE official website for November Primary Certificate Exams 2022, DPE Result dinajpur there are lacks of students who are participated in the terminal exams in the country along with Dinajpur Division also.Directorate of Primary Education (DPE) has successfully conducted those class 5th grade exams between 17th to 24th November 2022 across in the country for all education boards, and they have conduct answer sheet evaluation process to announce PSC Result 2022 Dinajpur Board with full mark sheet, Minister of Education has announced this is a very tuff process to complete on time for this DPE new GPA grading system of PEC Marking scheme.