æºå¨å¦ä¹ ä¸çæ°æ®éå
æ°æ®éåç±»
å¨æçç£(supervise)çæºå¨å¦ä¹ ä¸ï¼æ°æ®é常被åæ2~3个ï¼å³ï¼è®ç»é(train set) éªè¯é(validation set) æµè¯é(test set)ã
è®ç»éç¨æ¥ä¼°è®¡æ¨¡åï¼éªè¯éç¨æ¥ç¡®å®ç½ç»ç»ææè
æ§å¶æ¨¡åå¤æç¨åº¦çåæ°ï¼èæµè¯éåæ£éªæç»éæ©æä¼ç模åçæ§è½å¦ä½ã
Ripley, B.Dï¼1996ï¼å¨ä»çç»å
¸ä¸èPattern Recognition and Neural Networksä¸ç»åºäºè¿ä¸ä¸ªè¯çå®ä¹ã
Training set:
A set of examples used for learning, which is to fit the parameters [i.e., weights] of the classifier.
Validation set:
A set of examples used to tune the parameters [i.e., architecture, not weights] of a classifier, for example to choose the number of hidden units in a neural network.
Test set:
A set of examples used only to assess the performance [generalization] of a fully specified classifier.
æ¾ç¶ï¼training setæ¯ç¨æ¥è®ç»æ¨¡åæç¡®å®æ¨¡ååæ°çï¼å¦ANNä¸æå¼çï¼ validation setæ¯ç¨æ¥å模åéæ©ï¼model selectionï¼ï¼å³å模åçæç»ä¼ååç¡®å®çï¼å¦ANNçç»æï¼è test setå纯粹æ¯ä¸ºäºæµè¯å·²ç»è®ç»å¥½ç模åçæ¨å¹¿è½åãå½ç¶ï¼test setè¿å¹¶ä¸è½ä¿è¯æ¨¡åçæ£ç¡®æ§ï¼ä»åªæ¯è¯´ç¸ä¼¼çæ°æ®ç¨æ¤æ¨¡åä¼å¾åºç¸ä¼¼çç»æãä½å®é
åºç¨ä¸ï¼ä¸è¬åªå°æ°æ®éåæ两类ï¼å³training set åtest setï¼å¤§å¤æ°æç« å¹¶ä¸æ¶åvalidation setã
éæ©è®ç»éåæµè¯é
å
¶ä¸ä¸ä¸ªå
¸åçååæ¯è®ç»éå æ»æ ·æ¬ç50ï¼
ï¼èå
¶å®åå 25ï¼
ï¼ä¸é¨åé½æ¯ä»æ ·æ¬ä¸éæºæ½åãæ ·æ¬å°çæ¶åï¼ä¸é¢çååå°±ä¸åéäºã常ç¨çæ¯çå°é¨ååæµè¯éãç¶å对å
¶ä½Nä¸ªæ ·æ¬éç¨Kæ交åéªè¯æ³ãå°±æ¯å°æ ·æ¬æä¹±ï¼ç¶ååååæK份ï¼è½®æµéæ©å
¶ä¸Kï¼1份è®ç»ï¼å©ä½çä¸ä»½åéªè¯ï¼è®¡ç®é¢æµè¯¯å·®å¹³æ¹åï¼æåæK次çé¢æµè¯¯å·®å¹³æ¹åååå¹³åä½ä¸ºéæ©æä¼æ¨¡åç»æçä¾æ®ãç¹å«çKåNï¼å°±æ¯çä¸æ³ï¼leave one outï¼ã
æ°æ®å½ä¸åé®é¢
æ°æ®å½ä¸åå±äºæ°æ®çé¢å¤çãå 为sigmoidå½æ°æ ¹æ®ä¸åçåæ¢ï¼è¾åºå¨0å°1æè
ï¼1å°1ä¹é´ï¼å æ¤å¦æä¸åå½ä¸ï¼å°±ä¼åºç°æ ·æ¬è¾åºè¶
åºç¥ç»ç½ç»è¾åºçèå´ãéæ©æ大çå¼maxåæå°å¼minï¼åå¦ä¸åæ¢
xï¼(xï¼min)/(max-min)
å°±æ¯å½ä¸åã
éè¦æ³¨æçæ¯maxåminä¸åºè¯¥ç´æ¥éæ©ä¸ºxä¸çæ大å¼åæå°å¼ãåå æ¯æ ·æ¬åªæ¯æéçè§æµï¼æå¯è½è¿ææ´å¤§æè
æ´å°çè§æµï¼å æ¤åéçéæ©åºè¯¥maxéåxmax大ä¸äºåminæ¯xminå°ä¸äºçãå½ä¸å并ä¸æ»æ¯åéçé¢å¤çï¼å 为å®ä¸è½ä½¿åå¸ä¸å¯¹ç§°çæ ·æ¬åå¾æ´å¯¹ç§°ä¸äºï¼æ ååè¦å¥½ä¸äºãå¦å¤ï¼ææ¶å主æååæä¹è½èµ·å°éç»´çä½ç¨ã
温馨提示:答案为网友推荐,仅供参考