NOIP2017 GG 记

NOIP2017 GG 记

Day0 <script type="math/tex" id="MathJax-Element-1">Day\;0</script>

<script type="math/tex" id="MathJax-Element-2">\quad</script>吃完饭后买了一杯 Star's <script type="math/tex" id="MathJax-Element-3">\text{Star's}</script>圣诞拼配,浓郁的雪松香气。
<script type="math/tex" id="MathJax-Element-4">\quad</script>……结果晚上睡不着,只能瞎想,大概想了几点注意事项吧。

Day1 <script type="math/tex" id="MathJax-Element-5">Day\;1</script>

<script type="math/tex" id="MathJax-Element-6">\quad</script>分到一个很神的位置,左边是数论神犇 ZJY <script type="math/tex" id="MathJax-Element-7">\text{ZJY}</script>,前后各一个二中神犇,都比我强。
<script type="math/tex" id="MathJax-Element-8">\quad</script>比赛开始,先把题过一遍: T1 <script type="math/tex" id="MathJax-Element-9">\text{T1}</script>应该是结论题; T2 <script type="math/tex" id="MathJax-Element-10">\text{T2}</script>怎么看都是模拟; T3 <script type="math/tex" id="MathJax-Element-11">\text{T3}</script>乍一看好像是 Astar <script type="math/tex" id="MathJax-Element-12">\text{Astar}</script>。
<script type="math/tex" id="MathJax-Element-13">\quad</script>大概 8:40 <script type="math/tex" id="MathJax-Element-14">8:40</script>,开始做 T1 <script type="math/tex" id="MathJax-Element-15">\text{T1}</script>。结果直接把结论猜出来,写完对拍貌似才 8:50 <script type="math/tex" id="MathJax-Element-16">8:50</script>左右。跑了一下极限数据,觉得稳了。感觉今年签到题真水。
<script type="math/tex" id="MathJax-Element-17">\quad</script>开始做 T2 <script type="math/tex" id="MathJax-Element-18">\text{T2}</script>,列举了一些情况,决定用 DFS <script type="math/tex" id="MathJax-Element-19">\text{DFS}</script>写出来。写完调了几个错,过了大样例就不管了。大概 9:45 <script type="math/tex" id="MathJax-Element-20">9:45</script>左右吧。
<script type="math/tex" id="MathJax-Element-21">\quad</script>开始做 T3 <script type="math/tex" id="MathJax-Element-22">\text{T3}</script>,发现有模数,那应该和 Astar <script type="math/tex" id="MathJax-Element-23">\text{Astar}</script>无关了。想了一个用 BFSDP <script type="math/tex" id="MathJax-Element-24">\text{BFS}\;\text{DP}</script>的方法,每个点存 k+1 <script type="math/tex" id="MathJax-Element-25">k+1</script>个状态。虽然觉得复杂度很玄,但似乎挺正确,就开始码了。码到一半,发现貌似 DP <script type="math/tex" id="MathJax-Element-26">\text{DP}</script>值会循环转移,赶紧弃了。我想了很久,感觉没什么办法避免循环转移,就多开了一维状态存路径经过的点数,新开了一个程序。又打到一半,不知是脑抽了还是什么,突然就觉得之前的想法挺对,然后就回去码之前的程序了。码玩跑大样例,跑了很久才  WA <script type="math/tex" id="MathJax-Element-27">\text{ WA}</script>,冷静想一想,发现还是错的,就又回去写部分分了。最后过了小样例就不管了。
<script type="math/tex" id="MathJax-Element-28">\quad</script>出来知道了 T3 <script type="math/tex" id="MathJax-Element-29">\text{T3}</script>正解,怀疑自己没长脑子……
<script type="math/tex" id="MathJax-Element-30">\quad</script>自我感觉极差,感觉要翻。

Day2 <script type="math/tex" id="MathJax-Element-31">Day\;2</script>

<script type="math/tex" id="MathJax-Element-32">\quad</script>今天一定不能翻。
<script type="math/tex" id="MathJax-Element-33">\quad</script>一开始先把 T1 <script type="math/tex" id="MathJax-Element-34">\text{T1}</script>切了,跑去想 T2 <script type="math/tex" id="MathJax-Element-35">\text{T2}</script>。想了半天没想出好做法,只是感觉是状压 DP <script type="math/tex" id="MathJax-Element-36">\text{DP}</script>。打了个 BFS <script type="math/tex" id="MathJax-Element-37">\text{BFS}</script>暴力,感觉有一半分,但觉得还可以用 hash <script type="math/tex" id="MathJax-Element-38">\text{hash}</script>代替 map <script type="math/tex" id="MathJax-Element-39">\text{map}</script>优化。
T3 <script type="math/tex" id="MathJax-Element-40">\quad\text{T3}</script>写了个 mn <script type="math/tex" id="MathJax-Element-41">mn</script>的暴力,看数据范围以为能有 50 <script type="math/tex" id="MathJax-Element-42">50</script>。
<script type="math/tex" id="MathJax-Element-43">\quad</script>终于打出 hash <script type="math/tex" id="MathJax-Element-44">\text{hash}</script>,却发现要么爆空间,要么不对,就懒得挂链了。到 11:20+ <script type="math/tex" id="MathJax-Element-45">11:20+</script>,终于想出如何状压 DP <script type="math/tex" id="MathJax-Element-46">DP</script>,破釜沉舟 10min <script type="math/tex" id="MathJax-Element-47">10\text{min}</script>打完,发现 n=12 <script type="math/tex" id="MathJax-Element-48">n=12</script>时貌似超时了,赶紧优化掉一个 Θ(n) <script type="math/tex" id="MathJax-Element-49">\Theta(n)</script>。
<script type="math/tex" id="MathJax-Element-50">\quad</script>最后 15min <script type="math/tex" id="MathJax-Element-51">15\text{min}</script>开始检查,发现 T1 <script type="math/tex" id="MathJax-Element-52">\text{T1}</script>可能会爆 long long <script type="math/tex" id="MathJax-Element-53">\text{long long}</script>,但觉得 CCF <script type="math/tex" id="MathJax-Element-54">\text{CCF}</script>不会这么恶心就不管了。
<script type="math/tex" id="MathJax-Element-55">\quad</script>赛后同学说一定会卡 long long <script type="math/tex" id="MathJax-Element-56">\text{long long}</script>, T2 <script type="math/tex" id="MathJax-Element-57">\text{T2}</script>的做法可以再改少 Θ(n2) <script type="math/tex" id="MathJax-Element-58">\Theta(n^2)</script>,而且 T3nm <script type="math/tex" id="MathJax-Element-59">\text{T3}\;nm</script>预处理只有 30 <script type="math/tex" id="MathJax-Element-60">30</script>分。
<script type="math/tex" id="MathJax-Element-61">\quad</script>感觉自己没救了……

赛后

<script type="math/tex" id="MathJax-Element-62">\quad</script>好彩 Day 1 <script type="math/tex" id="MathJax-Element-63">\text{Day 1}</script>两题都没炸, CCF <script type="math/tex" id="MathJax-Element-64">\text{CCF}</script>也没丧心病狂卡 long long <script type="math/tex" id="MathJax-Element-65">\text{long long}</script>,算是苟进 400 <script type="math/tex" id="MathJax-Element-66">400</script>了。
<script type="math/tex" id="MathJax-Element-67">\quad</script>但这次 NOIP <script type="math/tex" id="MathJax-Element-68">\text{NOIP}</script>总体表现还是太差了。要多多吸取教训。

©️2020 CSDN 皮肤主题: 编程工作室 设计师:CSDN官方博客 返回首页