2019å¹´9æ27æ¥ã®PyData.Tokyo Meetup #21ã§ã®çºè¡¨è³æã§ãã Optuna (https://github.com/pfnet/optuna) ã®ä½¿ãæ¹ãã½ããã¦ã§ã¢ãã¶ã¤ã³ãLightGBMåãã®æ°æ©è½ã«ã¤ãã¦ç´¹ä»ãã¦ãã¾ããRead less
ãã¤ãã¼ãã©ã¡ã¼ã¿ãæ¢ç´¢ãããããã°ãªãããµã¼ããOptunaãªã©ãå©ç¨ãããã¨ãããã¨æãã¾ãã ããããããã¤ãã¼ãã©ã¡ã¼ã¿æ¢ç´¢ãã¦ã¿ããã®ãããªQiitaè¨äºãªã©ã§ã¯ééã£ã¦æ¸ããã¦ãããã¨ãå¤ãã®ã§ãããXGBoostãLightGBMã® n_estimators ( num_boosting_rounds )ãKerasã® epochs ããã©ã¡ã¼ã¿ãµã¼ãã®å¯¾è±¡ã«ãã¦ã¯ããã¾ããã ããã¾ãããã¨ãããããããç¡é§ãªãã¨ããã¦ãã¾ãã ãªããn_estimatorsãepochsãæ¢ç´¢ããã¨ç¡é§ãªã®ã çç±ã¯ã·ã³ãã«ã§ããããã®ãã©ã¡ã¼ã¿ã¯ã大ããå¤ã§ç²¾åº¦ç¢ºèªããéç¨ã§å°ããå¤ã®çµæãåãããããã§ãã LightGBMã®n_estimatorsã¯æ§ç¯ãã決å®æ¨ã®æ°ã表ãã¦ãã¾ãã ä¾ã¨ãã¦ãn_estimators=5 (ãããªå°ããå¤ã§å¦ç¿ãããã¨ã¯ãªãã§ãããç°¡åã®ãã)ã§
(Image by Pixabay) "Top 10 Statistics Mistakes Made by Data Scientists"ã¨ããåºæ¿çãªã¿ã¤ãã«ã®è¨äºãåºã¦ããã®ãKDnuggetsçµç±ã§ç¥ãã¾ãããããã¼ã¿ãµã¤ã¨ã³ãã£ã¹ãããããããã¡ãªçµ±è¨å¦çãªèª¤ãããã10ãã¨ãããã¨ã§ãããã«ããªããããäºä¾ãè²ã è¼ã£ã¦ãã¦é¢ç½ãã§ãã ã¨ãããã¨ã§ãä»åã¯ãã®è¨äºãå ¨è¨³ã«ãªããªãç¯å²ã§æ訳ãã¦ããã®å 容ãåå³ãã¦ã¿ããã¨æãã¾ãï¼ç´è¨³ãã¦ãæå³ãåãã¥ããç®æãå¤ãã£ãããããªãã®é¨åãæ訳ãªããæ訳ãã¦ãã¾ãï¼ãããã訳ã®æ¹ãè¯ããªã©ã®ã³ã¡ã³ãããã°æ¯éãå¯ããã ããï¼ãè¨ãããããªã§ãããããããæµ·å¤è¨äºç´¹ä»ãããæã¯ãã¿åãã¨ãããã¨ã§ããæªããããããã å è¨äºã®å 容 1. Not fully understand objective functionï¼ç®çé¢æ°ãä½ã
ãªãªã¼ã¹ãé害æ å ±ãªã©ã®ãµã¼ãã¹ã®ãç¥ãã
ææ°ã®äººæ°ã¨ã³ããªã¼ã®é ä¿¡
å¦çãå®è¡ä¸ã§ã
j次ã®ããã¯ãã¼ã¯
kåã®ããã¯ãã¼ã¯
lãã¨ã§èªã
eã³ã¡ã³ãä¸è¦§ãéã
oãã¼ã¸ãéã
{{#tags}}- {{label}}
{{/tags}}