b*********h
发帖数: 103 | 1 最大 m 子段和，贴一个通用做法 时间 O(n*m) 空间 O(m) 只是 m = 2
class Solution {
public:
int maxProfit(vector &prices) {
int f[3] = {0};
int g[3] = {0};
int n = prices.size() - 1;
for (int i = 0; i < n; ++i) {
int diff = prices[i+1] - prices[i];
int m = min(i+1, 2);
for (int j = m; j >= 1; --j) {
f[j] = max(f[j], g[j-1]) + diff;
g[j] = max(g[j], f[j]);
}
}
return max(g[1], g[2]);
... 阅读全帖 |
|
b*********h
发帖数: 103 | 2 最大 m 子段和，贴一个通用做法 时间 O(n*m) 空间 O(m) 只是 m = 2
class Solution {
public:
int maxProfit(vector &prices) {
int f[3] = {0};
int g[3] = {0};
int n = prices.size() - 1;
for (int i = 0; i < n; ++i) {
int diff = prices[i+1] - prices[i];
int m = min(i+1, 2);
for (int j = m; j >= 1; --j) {
f[j] = max(f[j], g[j-1]) + diff;
g[j] = max(g[j], f[j]);
}
}
return max(g[1], g[2]);
... 阅读全帖 |
|
h******8
发帖数: 278 | 3
diff is line by line.
diff是经常用，从来不知道是line by line， compare 的时候是line不要line
compare，每行是char by char？
那天还有一个人问kill（2） signal call是怎么implement，这种问题怎么真是没法准
备。 |
|
r*******e
发帖数: 7583 | 4 sum of diff != |sum of male - sum of female|
sum of diff = |diff1| + |diff2| ... + |diffn|
height |
|
l*****s
发帖数: 279 | 5 public static int changeArray(int[] n, int a, int b) {
int len = n.length;
Arrays.sort(n);
int sumSoFar = 0;
int test = 0;
int diff = a-b;
boolean done = false;
for (int i = len-1; i>=0; i--) {
sumSoFar += n[i];
test = (sumSoFar - diff) / (len - i);
if (test > n[i-1] && test <= n[i]) {
done = true;
break;
}
}
if (done)
return test... 阅读全帖 |
|
l*****s
发帖数: 279 | 6 public static int changeArray(int[] n, int a, int b) {
int len = n.length;
Arrays.sort(n);
int sumSoFar = 0;
int test = 0;
int diff = a-b;
boolean done = false;
for (int i = len-1; i>=0; i--) {
sumSoFar += n[i];
test = (sumSoFar - diff) / (len - i);
if (test > n[i-1] && test <= n[i]) {
done = true;
break;
}
}
if (done)
return test... 阅读全帖 |
|
m********7
发帖数: 1368 | 7 大家新年好~！
leetcode上 stock III的这个 dp 解法，http://www.cnblogs.com/caijinlong/archive/2013/05/01/3053165.html
上讲的dp算法看懂了，但是代码实现和算法如何对应起来的，没看懂。。
欢迎大牛们指导~~谢！！
code:
int maxProfit(vector &prices) {
int f[3] = {0};
int g[3] = {0};
int n = prices.size() - 1;
for (int i = 0; i < n; ++i) {
int diff = prices[i+1] - prices[i];
int m = min(i+1, 2);
for (int j = m; j >= 1; --j) {
f[j] =... 阅读全帖 |
|
A*********c
发帖数: 430 | 8 不是大牛，表达能力有限，试着解释一下。
我估计你不懂的是这两行：
12 f[j] = max(f[j], g[j-1]) + diff;
13 g[j] = max(g[j], f[j]);
你看懂了算法，就按着算法说。
该算算法的核心是这个：f[i][j] = max(g[j-1] + a[i], f[i-1][j] + a[i]).
第12行就是算这个，其中diff就是a[i].
第13行做得就是“memorize the optimal j-1segments”
算法仅仅用了常数空间的原因是因为f仅仅依赖于前一个f，而不是前面所有的f值。而g
仅仅依赖于f和当前的g。
完了。 |
|
m********7
发帖数: 1368 | 9 Zeal,
谢谢你帮忙讲解。。
我看懂的是算法和时间复杂度O(mn),空间复杂度O(m)，怪我，不理解的地方应该先说出
来的，那我们看着代码讨论吧。。
3 int maxProfit(vector &prices) {
4 int f[3] = {0};
5 int g[3] = {0};
6
7 int n = prices.size() - 1;
8 for (int i = 0; i < n; ++i) {
9 int diff = prices[i+1] - prices[i];
10 int m = min(i+1, 2);
11 for (int j = m; j >= 1; --j) {
12 f[j] = max(f[j], g[j-1]) + diff;
13 g[j] = ... 阅读全帖 |
|
f*******t
发帖数: 7549 | 10 主要工作是通过监测performance metrics，定位哪个diff造成了性能损失，追着
author要求修复，以此挽救公司的硬件开支。这个组visibility和impact极高，经常揪出
一个diff就能减少1% global frontend CPU usage。
我觉得进这个组别的不说，至少读代码的能力要强。另外每天看各种php code，要能忍
住不吐也不容易。。。 |
|
x****g
发帖数: 1512 | 11 这个就是O（n+m）？
比较两个值，那个小move哪个？
A[i]<=A[i+1]
B[j]<=B[j+1]
如果 A[i]
如果A[i+1]
如果A[i+1]>B[j]， 在该移动策略下，无论A[i+1]>B[j+1]与否A[i+1]和B[j],B[j+1
]都会发生比较。
反之同理。
i=0;
j=0;
int smallest = INT_MAX;
while(i
{
int diff = abs(A[i]-B[j]);
smallest = min(diff,smallest);
if(smallest == 0)
{
return 0;
}
if(A[i]>B[j])
{
j++;
}
else
{
i++;
}
}
return smallest. |
|
h******6
发帖数: 2697 | 12 请问4是下面这个代码的意思吗
public static int partition(int[] array) {
if (array == null || array.length == 0) {
return -1;
}
int sum = 0;
for (int d : array) {
sum += d;
}
int pos = -1;
int diff = Integer.MAX_VALUE;
int left = 0;
for (int i = 0; i < array.length; ++i) {
left += array[i];
int right = sum - left;
if (Math.abs(right - left) < diff) {
pos = i;
... 阅读全帖 |
|
h******6
发帖数: 2697 | 13 请问4是下面这个代码的意思吗
public static int partition(int[] array) {
if (array == null || array.length == 0) {
return -1;
}
int sum = 0;
for (int d : array) {
sum += d;
}
int pos = -1;
int diff = Integer.MAX_VALUE;
int left = 0;
for (int i = 0; i < array.length; ++i) {
left += array[i];
int right = sum - left;
if (Math.abs(right - left) < diff) {
pos = i;
... 阅读全帖 |
|
t***t
发帖数: 6066 | 14 there aren't so big diff among diff states in US before. but in recent years
, FLG pumped salary and packages to > 200k for regular software engineers.
actually most other companies still can't offer > 200k package for even
senior engineers with many years experiences. |
|
a**********0
发帖数: 422 | 15 The Unix Commands
其实就是攒了一下网上的资料
# Create a new tar archive.
# the folder dirname/ is compressed into archive_name.tar
tar cvf archive_name.tar dirname/
# Extract from an existing tar archive
tar xvf archive_name.tar
# View an existing tar archive
tar tvf archive_name.tar
# Search for a given string in a file (case in-sensitive search).
grep -i "the" demo_file
# Print the matched line, along with the 3 lines after it.
grep -A 3 -i "example" demo_text
# Search for a given string in all files recur... 阅读全帖 |
|
j*****8
发帖数: 3635 | 16 我的28行，不过看着也挺繁琐的。思路就是碰到递增的就一直加1，递减的话一直减，
碰到往上走时就回头check一下看那段递减的需不需要调整（负数，或者最小的不是1）
public int candy(int[] ratings) {
if(ratings == null || ratings.length == 0) return 0;
int len = ratings.length;
int[] num = new int[len];
for(int i = 0; i < len; i++) {
if(i == 0) num[i] = 1;
else {
if(ratings[i] > ratings[i-1]) num[i] = num[i-1] + 1;
else if(ratings[i] == ratings[i-1]) num[i] = 1;
else {
... 阅读全帖 |
|
q********c
发帖数: 1774 | 17 类似于rate limiter吗？试了下：
boolean allowRequest() {
int minInterval = 1000/N; // N is the max rate
int current = now();
int diff = current-last;
last = current; //last would be a class private member
if(diff < minInterval)
return false;
else
return true;
} |
|
x*******9
发帖数: 138 | 18 略暴力，Lintcode上的题。（没看错，是Lintcode，不是Leetcode）
class Solution {
public:
/**
* @param A: An integer array.
* @param target: An integer.
*/
int MinAdjustmentCost(vector A, int target) {
if (A.empty()) {
return 0;
}
int n = A.size();
int ptr = 0;
memset(dp, 0, sizeof(dp));
for (int i = 0; i < n; i++) {
int cur = A[i];
int next = ptr ^ 1;
memset(dp[next], INF, sizeof(dp[nex... 阅读全帖 |
|
t********5
发帖数: 522 | 19 git diff
git add .
git commit .
git push origin
git tag v0.0.1
while true:
git diff
git add .
git commit --amend
git push origin -f
git tag v0.0.2 |
|
h**********c
发帖数: 4120 | 20 这个 题
如果有m对情侣，假设坐满，假设全是情侣n
那么正确做法坐法应该有 p m m * ２, hashmap or list
当前坐法 string diff 正确做法，实际上string diff就是两个矢量差normal form。
比distance 最小的一个出来，然后折腾 |
|
s*******h
发帖数: 3219 | 21 【 以下文字转载自 Working 讨论区 】
发信人: itcecsa (itcecsa), 信区: Working
标 题: 恶心的烙印同事
发信站: BBS 未名空间站 (Wed Apr 8 23:08:46 2015, 美东)
烙印比我早几个月加入这家公司，我们都是码工，快在这家公司工作3年了。我当初来
的时候他对我一点也不客气，一副倚老卖老的样子。
从去年开始，我听说烙印自己在外面搞项目。渐渐得公司这边在一直偷懒。由于老板是
我们组promote上去的，所以老板对他没有管得太紧。我可是清楚得开在眼里。烙印在
老板离开table后讲笑话，不好好干活。老板在的时候，他装得很老实的样子。
去年他作了几个项目，结果是烂得一塌糊涂。而且专门挑简单，不重要的活干。能偷懒
尽量偷懒，daily scrum（11：45AM）30%左右时间缺席。我对他的偷懒是看在眼里，但
我从来没有在背后说过他的坏话。
最近，他的一个local environment不能build了，都几个礼拜了也不去修。他需要改几
个javascript文件，要我帮忙在我的machine上改。我当初就跟他说了，试一下可... 阅读全帖 |
|
n*****5
发帖数: 984 | 22 Amplify : 做K12教育的公司
HackerRank online coding。需要自己写读入函数。有点像nearest common ancestor
with parents。 最后快一半的test case 没过，我也不知道问啥，也没搞明白怎么调
试... 就没有然后了。
MLB ： 专做baseball相关的公司，好像很大，还在扩招。
core java question, hashcode equals.
然后问了很多spring的问题。 why use spring mvc, 怎么deploy jar， container
是啥。。。
没算法，电面就挂了。
Portware: 好像做交易平台的。 电面全是core java 相关的小问题。
Diff : String , StringBuffer, StringBuilder
Serialize, deserialize, transient
final var, method, class
try catch finally. can go with only try and finally . When ... 阅读全帖 |
|
l******s
发帖数: 3045 | 23 private static bool catchable(int[] strategy, int numCaves){
bool bOdd = false, bEven = false;
for(int i = 0; i < strategy.Length && strategy[i] < numCaves; i++){
int left = i;
if (strategy[i] != 1 && strategy[i] != numCaves - 2) continue;
if (numCaves > 3 && Math.Abs(strategy[i] - strategy[i + 1]) > 1)
continue;
if (numCaves > 3 && i < strategy.Length - 1){
int diff = strategy[i + 1] - strategy[i];
while (i - left + 1 < numCaves - ... 阅读全帖 |
|
d******b
发帖数: 73 | 24 一行搞定。。。。。
int main() {
std::vector v1 {"Zebra"};
std::vector v2 {"Antelope", "Bison", "Cow"};
std::vector diff;
std::set_difference(v1.begin(), v1.end(), v2.begin(), v2.end(), back_
inserter(diff));
return 0;
} |
|
|
l******n
发帖数: 641 | 26 like a car, lx or dx, no diff when sale.
but 2 door and 4 door have diff. |
|
t**s
发帖数: 1393 | 27 If you are comfortable to pay off loan in 7 years, then 7ARM, otherwise 15-
year loan.
0.125% rate is essentially nothing. For a $400k loan, 0.125% rate diff means
$30 diff per month |
|
t**s
发帖数: 1393 | 28 I will choose B，unless the pond view is a minus, not a plus。
(一般除了很小且脏而且对 kids 安全有隐患的小 pond 才是不好的)。
300 sqft diff 至少可以 cover 19k price diff，对不对？
大理石 counter tops 也不完全是必要的，比较好的 corian 材料也不错。
High ceiling 并非缺点，对我而言还是优点，空旷能给人以享受。冬天确实冷一点，
但是夏天也凉爽点，两两相抵。 |
|
t**s
发帖数: 1393 | 29 diff states may have diff "rules". |
|
k***a
发帖数: 2400 | 30 如果是no closing cost的refi，包括所有的费用。没有impound. closing时你只付
interests， 新贷款比原来的少借了，你还要付laon amount的diff， 多借了，loan
amount的diff是你的credit， 你付的appraisal report fee 和credit report fee也
要credit给你。 |
|
k***a
发帖数: 2400 | 31 我们refi cover所有费用。如果是no closing cost的refi，没有impound. closing时
你只付interests， 新贷款比原来的少借了，你还要付laon amount的diff， 多借了，
loan amount的diff是你的credit， 你付的appraisal report fee 和credit report
fee也要credit给你。 |
|
|
c*********r
发帖数: 1312 | 33 来自主题: NextGeneration版 - 震惊加后悔 注意干净卫生是一定必须的，我无可否认。我吃完饭还是要洗碗的，呵呵。但是无须搞
得正常饮食都要想着把细菌病毒都除去，你在医院工作应该知道，正常生活中是无法做
到无菌无病毒的。最怕走两个极端：极端的干净和极端的不干净。所以我的回复里用到
的词语是“适当的”接触这些细菌病毒是建立和调节人体自身免疫力所必须的。。。
你提到的HIV ,乙肝，美国流行的C-diff, MRSA, VRE 等，适用于楼主这里的例子并不
多。
HIV的传播途径肯定不是水杯；乙肝也许有可能，但是更重要的是切断与携带者的接触
；C-diff我查了一下，主要是通过感染者的粪便传染；MRSA, VRE是耐受菌，主要的感
染源是医院，主要发生在医院里。不知道我说的对不对，如果不对的还请医院工作的专
业人士指正。所以要避免这些传染，我想更重要的是避免传染源吧。
现代的很多自身免疫性疾病、过敏等等，都和过于干净有关，而不是过于不干净。人体
内的细菌数目都是人自身细胞数目的十倍了，而且这些细菌都是人体健康必不可少的。
绝大多数情况下，如果一个人的正常菌群建立好了，吃进去几个细菌病毒啥的，应该是
没啥危害的。
我是做生物的，以前做微生... 阅读全帖 |
|
s****n
发帖数: 248 | 34 来自主题: NextGeneration版 - 震惊加后悔 同意！
小孩没拉肚子，没有肚子疼这些受病菌毒素攻击的症状，就没问题。
那些病菌的确主要是在医院里才有，尤其C.diff.
日常生活一般干净就好了。
还要注意的是抗生素不得已的情况下才吃，否则破坏了肠道菌群的平衡，就给病菌滋长
机会了，C.diff就是一例。 |
|
c*********r
发帖数: 1312 | 35 来自主题: NextGeneration版 - 震惊加后悔 注意干净卫生是一定必须的，我无可否认。我吃完饭还是要洗碗的，呵呵。但是无须搞
得正常饮食都要想着把细菌病毒都除去，你在医院工作应该知道，正常生活中是无法做
到无菌无病毒的。最怕走两个极端：极端的干净和极端的不干净。所以我的回复里用到
的词语是“适当的”接触这些细菌病毒是建立和调节人体自身免疫力所必须的。。。
你提到的HIV ,乙肝，美国流行的C-diff, MRSA, VRE 等，适用于楼主这里的例子并不
多。
HIV的传播途径肯定不是水杯；乙肝也许有可能，但是更重要的是切断与携带者的接触
；C-diff我查了一下，主要是通过感染者的粪便传染；MRSA, VRE是耐受菌，主要的感
染源是医院，主要发生在医院里。不知道我说的对不对，如果不对的还请医院工作的专
业人士指正。所以要避免这些传染，我想更重要的是避免传染源吧。
现代的很多自身免疫性疾病、过敏等等，都和过于干净有关，而不是过于不干净。人体
内的细菌数目都是人自身细胞数目的十倍了，而且这些细菌都是人体健康必不可少的。
绝大多数情况下，如果一个人的正常菌群建立好了，吃进去几个细菌病毒啥的，应该是
没啥危害的。
我是做生物的，以前做微生... 阅读全帖 |
|
s****n
发帖数: 248 | 36 来自主题: NextGeneration版 - 震惊加后悔 同意！
小孩没拉肚子，没有肚子疼这些受病菌毒素攻击的症状，就没问题。
那些病菌的确主要是在医院里才有，尤其C.diff.
日常生活一般干净就好了。
还要注意的是抗生素不得已的情况下才吃，否则破坏了肠道菌群的平衡，就给病菌滋长
机会了，C.diff就是一例。 |
|
|
c**t
发帖数: 9197 | 38 【 以下文字转载自 Chinook 俱乐部 】
发信人: cbot (cbot), 信区: Chinook
标 题: balanced trading approach
发信站: BBS 未名空间站 (Wed Aug 3 13:08:59 2011, 美东)
make a list every time before enter a position.
bull list: everything you think for market going up.
bear list: the reverse.
Then compare the length of the two list.
The bigger the diff, the better.
if no much diff, then wait. |
|
p******e
发帖数: 17163 | 39 diff exchange has diff trading hrs. nyse and nasdaq go from 9:30 est to 4:00
pm est. cboe i believe got 15 min xtra. and cme electronic product
generally round the clock except 15 min halt after 4;15 and 45 min halt
between 5:15 and 6, tho u gotta go to their site to ck it out.
as with weekend trading, there is just not much liquidity and so no price
discovery so no exchange offering to trade cuz price will gyrate too much
and we gambler need a break too.tho i believe some Mideast country is
tra... 阅读全帖 |
|
f*****g
发帖数: 15860 | 40 if you're 80% sure about 再等6个月才能批（如果顺利），then stay. GC makes
HUGE diff for your future.
EB2 now is almost a disaster. 工资股票待遇会好很多 is nothing compared to GC.
if you're married or getting married, GC makes a huge diff. for your LD for
her job opportunities too.
拿到绿卡后还要等6个月才能跳（180天政策）, as far as i know, there's no such
official rule, it's only some laywers' own interpretation of "the intention
to work for X employer on a permanent basis". in another word, 180天 may
make you feel a little "safer" but t... 阅读全帖 |
|
p*****y
发帖数: 529 | 41 most likely, it won't make any difference if you complain or not. and as
other people said, HR may not even ask for your comments.
So for your benefits, if it makes you feel better, let it out.
If you work in a small industry that you may cross your manager again in the
future, don't say anything. I don't think he'll know your comments from HR,
just in case.
In any case, it's such a small case that you don't need to review so many
diff opinions from diff people to make a decision. Just do it! |
|
f*****g
发帖数: 15860 | 42 that's why i didn't type a thousand words to explain this to you.
"contract" or "contractor" is a confusing term, diff. ppl mean diff. things.
据说1099比W2好, for your own good, you better google for a better
understanding, there's nothing called "据说". |
|
i*****a
发帖数: 572 | 43 烙印比我早几个月加入这家公司，我们都是码工，快在这家公司工作3年了。我当初来
的时候他对我一点也不客气，一副倚老卖老的样子。
从去年开始，我听说烙印自己在外面搞项目。渐渐得公司这边在一直偷懒。由于老板是
我们组promote上去的，所以老板对他没有管得太紧。我可是清楚得开在眼里。烙印在
老板离开table后讲笑话，不好好干活。老板在的时候，他装得很老实的样子。
去年他作了几个项目，结果是烂得一塌糊涂。而且专门挑简单，不重要的活干。能偷懒
尽量偷懒，daily scrum（11：45AM）30%左右时间缺席。我对他的偷懒是看在眼里，但
我从来没有在背后说过他的坏话。
最近，他的一个local environment不能build了，都几个礼拜了也不去修。他需要改几
个javascript文件，要我帮忙在我的machine上改。我当初就跟他说了，试一下可以，
但我不管后面的测试和submit。可是我怎么就上了他的当呢，他后来还要我submit
pull request。由于我不想要不必要的麻烦，就send给了他一个diff image，希望他能
在自己的local上改代码，然后他自己sub... 阅读全帖 |
|
y**c
发帖数: 447 | 44 I called the tier-2 officer three times today, and have diff answer/
attitudes from diff ppl. some said they cannot see/provide whether the name/background check is done or not, but another said the check result has not been sent back from FBI, but they can see my FP is done which I already know from FBI.
Is there anyone who was told a positive answer regarding the name/
background check from the tier-2 officer?
I was also told my case was assigned to the "decision division" already. How
long w... 阅读全帖 |
|
b*********g
发帖数: 951 | 45 抽个空，跟大家反馈一下我在干啥：
1） 我自己在web of science 总结了一遍独立引用。律师说不行，用google scholar
的，因为和我交的140一致。
有个网站可以部分统计google scholar结果（http://cids.fc.ul.pt/cids_3_0/index.php），但还是担心不够精确。保险起见，还是自己一点点做。
所以我现在再一个个的核对独引， 一边核对，心里一遍遍问候着uscis。
google scholar难得是不能像wos一样自动输出格式化列表。就算程序抓取，很容易被
google block。
最难的一篇文章，你们知道左边屏幕上是百十个作者，另一个屏幕是100篇文章，挨个对
比有没有共同作者的滋味吗？
前面有版友，回复可以用google scholar -auth:"LASTNAME" -author:"First LAST"
... 的方法，对这种多作者的失效，因为google 对字符串有限制，填不了那么多作者。
后来我写了个小程序 （那些Diff 软件也试了，不行。python 的 diff libraray也不
行)，把... 阅读全帖 |
|
g***m
发帖数: 465 | 46 no.
What ever you actually earned (as indicated on your W-2 or other forms) is the
only taxable resources you need to file.
BTW, diff. states have diff. policies regarding state tax. You'd better
consult with tax secretary or other Chinese scholars in your school. |
|
r**********e
发帖数: 28 | 47 Hi, I run into those German and can't understand, would anyone please
translate them into English? Your help is appreciated greatly.
Thanks alot
Grenzen der Spektralintervalle im KURZWELLIGEN
Intervalle, direkter Strahlungsfluss (ALL)
Intervalle, direkter Strahlungsfluss (ERDE)
Intervalle, diff. abwaerts. Strahl.
Intervalle, diff. aufwaerts. Strahl. (ALL)
Grenzen der Spektralintervalle im LANGWELLIGEN
Intervalle, diffuser abwaerts. Strahl. (ERDE)
Intervalle, diffuser aufwaerts. Strahl. (ALL) |
|
P*****a
发帖数: 2464 | 48 diff people are allergic to diff pollens ba,
so it depends...
mine ended last year in mid-May... |
|
P*****a
发帖数: 2464 | 49 hmm...diff people are allergic to diff types of tree pollens...
I have suffered from mid/late April till 1-2 weeks ago,
while my hubby is still going through the process - his started in early/mid
May |
|
b********r
发帖数: 7725 | 50 budget? diff building level is very diff |
|
