stander random forest:  random K features, enum all values as split, find best split.

LINKS:https://en.wikipedia.org/wiki/Random_forest



LINKS:http://docs.opencv.org/2.4/modules/ml/doc/ertrees.html

1. Extremely randomized trees don’t apply the bagging procedure to construct a set of the training samples for each tree. The same input training set is used to train all trees.
2. Extremely randomized trees pick a node split very extremely (both a variable index and variable splitting value are chosen randomly), whereas Random Forest finds the best split (optimal one by variable index and variable splitting value) among random subset of variables.
   
   Extremely randomized trees用了所有的樣本作為訓(xùn)練集；Extremely randomized trees隨機(jī)選一個特征和一個值作為分割標(biāo)準(zhǔn)；
   
   LINKS:http://scikit-learn.org/stable/modules/generated/sklearn.tree.ExtraTreeRegressor.html#sklearn.tree.ExtraTreeRegressor
   
   This class implements a meta estimator that fits a number of randomized decision trees (a.k.a. extra-trees) on various sub-samples of the dataset and use averaging to improve the predictive accuracy and control over-fitting.
   
   Extra-trees differ from classic decision trees in the way they are built. When looking for the best split to separate the samples of a node into two groups, random splits are drawn for each of the max_features randomly selected features and the best split among those is chosen. When max_features is set 1, this amounts to building a totally random decision tree.
   
   extra-trees 的ensemble用了bagging，然后選取多個特征，每個特征隨機(jī)選一個值作為分割標(biāo)準(zhǔn)建樹。
   
   一種實現(xiàn)方法：
          樣本bagging, random n features & random k values ，求最優(yōu)，建樹。 
   
   

主要類：
CCNode

   CCDirector
   CCScene
   CCLayer


定時更新：

   [[[CCDirector sharedDirector] scheduler] scheduleUpdateForTarget:self priority:0 paused:NO];

   //[[[CCDirector sharedDirector] scheduler] unscheduleUpdateForTarget:self];

接收輸入：
   v0.99

      CCStandardTouchDelegate

      CCTargetedTouchDelegate

   v2.10

      CCTouchOneByOneDelegate

      CCTouchAllAtOnceDelegate

   [[[CCDirector sharedDirector] touchDispatcher] addTargetedDelegate:self priority:0 swallowsTouches:YES];

   //[[[CCDirector sharedDirector] touchDispatcher] removeDelegate:self];

坐標(biāo)系統(tǒng)：
   position是設(shè)置相對于父親節(jié)點的坐標(biāo)
   self.anchorPoint和self.position重合

多層：
   [cclayer.addchild cclayer];
   一層疊一層

     摘要:   閱讀全文

     摘要:   閱讀全文

     摘要:   閱讀全文

     摘要:   閱讀全文

     摘要: UVA 10801 Lift Hopping  閱讀全文

     摘要:   閱讀全文

     摘要:   閱讀全文

今天做了次CF，兩個小時比賽時間。。用一小時水了兩題之后，又用一個小時的龜速想了一個不知道什么玩意的玩意，比賽沒A掉，比賽結(jié)束Ａ掉了。為什么要想那么久呢。。。。。。水題也要想那么久。。。。。小細(xì)節(jié)處理不好。。。。。
不過話說CF的題目不錯，不像廣大中文OJ的無腦題
http://codeforces.com/problemset/problem/180/E 
不貼代碼了。