é è¤ç²å¸ @LazyWorkz ãã¡ã®è¦ªããã¥ã¼ã¹è¦ãªãããæè¿ã¯æ®ºäººãå¢ãã¨ããªãâ¦æããããã£ã¨ãããããããã¨ãå¯è¨ã¬ãããã§ãã¡ãã£ã¨å¾ ã¦ããã¢ããããã®æ代ãããããã®æ代ã®ã»ãããã¼ã£ã¨æ¥æ¬ã¯è¯ããªã£ã¨ãããã殺人ãªããï¼å²æ¸ãã§ã¨æ°åãè¦ãã¤ããæã pic.twitter.com/KqCTGeyLr1 2016-09-14 08:08:43 é è¤ç²å¸ @LazyWorkz ãæã¯æ®ºäººã®ãã¥ã¼ã¹ãªããçããã£ãã®ã«ã ãããæã¯æ®ºäººãçãããªãã£ããããããçãããªããã¨ãªãããããããã¥ã¼ã¹ã«ããªãããèãã¦ãå¿ããããªã pic.twitter.com/JjUNhzNLG2 2016-09-14 08:19:35 é è¤ç²å¸ @LazyWorkz ä»æ®ºæ»äº¡è æ°ã°ã©ãããã¾ãã®èªæ®ºè æ°ã®ã»ãã«ããããªã«å¤ãã®ãï¼ãã£ã¦é©ã声ãããã¾ããããããé©ãã»ã©å°ãªããã§ããããã ã
æ¨æ¥ã®è¨äºããã®ã¤ã¥ãï¼ ä»æ¥ã¯æ¬é¡ã®é ä½ç¸é¢ä¿æ°ã«ã¤ãã¦æ¸ãï¼ å ã ã¯2ã¤ã®ãªã¹ãã®é¡ä¼¼åº¦ã測ãææ¨ã¨ãã¦ã®æèã§èãã¦ããããã ããã©ï¼å®ã¯2ã¤ã®å¤æ°ãæ£è¦åå¸ããªãã£ããï¼é¢æ£å¤ãåãæãªã©ã«ç¸é¢ä¿æ°ã測ããã³ãã©ã¡ããªãã¯ãªæ¹æ³ã§ããããã¨ãããã£ã (æ¨æ¥ã®è©±)ï¼ 2ã¤ã®é ä½ç¸é¢ä¿æ°ã«ã¤ãã¦è§£èª¬ããå¾ï¼ã¡ãã£ã¨æ°ã«ãªããã¨ï¼ç°ãªãéåã®é åºãªã¹ãã®ç¸é¢ä¿æ°ã測ãæ¹æ³ãªã©ã«ã¤ãã¦æ¸ãï¼ é ä½ç¸é¢ä¿æ° æ å ±æ¤ç´¢ãæ¨è¦ãªã©ã«ããã¦ï¼2ã¤ã®é åºãªã¹ããä¸ããããå ´åã«ï¼ããããã©ãã»ã© ä¼¼ã¦ãããã¨ããé¡ä¼¼æ§è©ä¾¡ã®ããã«é ä½ç¸é¢ä¿æ°ãç¨ããï¼ ãã ï¼åå説æãããã¢ã½ã³ç¸é¢ä¿æ°ã¯ç¢ºçå¤æ°ãæ£è¦åå¸ãã¦ããä»®å®ãç½®ãã¦ããã®ã§ï¼ç¢ºçå¤æ°ãæããã«æ£è¦åå¸ãã¦ããªãå ´åã®è¨ç®ã«å©ç¨ã§ããï¼ é ä½ç¸é¢ä¿æ°ã¯ã±ã³ãã¼ã«ã®é ä½ç¸é¢ä¿æ°ã¨ï¼ã¹ãã¢ãã³ã®é ä½ç¸é¢ä¿æ°ã¨ãããã®ãããï¼æ³¨æããã¹ãç¹ã¯ï¼
ã½ããã¦ã§ã¢éçºã®åç¹ã¯å¯è½æ§ã®è¿½æ±ã§ãããä¸å¯è½ãå¯è½ã«ãããã¨ã§ããã²ã¨ãã³ ã½ããã¦ã§ã¢ ãéçºãããã¨ãã¨ã³ã¸ãã¢ã¯æ¬¡ã« ç¨åº¦ ã¨ãã課é¡ã«åãåããã¨ã«ãªãã¾ããä¼æ¥åãã®ã½ããã¦ã§ã¢ã§ããã°ããé度ã¯ã©ããããããã¨é »ç¹ã«åããããä¿¡é ¼æ§ã¯ã©ã®ç¨åº¦ããã¨ããç¹ãéè¦ããã¾ãã ã½ããã¦ã§ã¢ã®ããã©ã¼ãã³ã¹ã«é¢ãã質åã«çããããã«ã¯æ£ããå 容ãèªãä¸ã§æ¬ ãããªãã®ãçµ±è¨å¦ã§ãã ã¨ã¯ãããçµ±è¨å¦ã«ã¤ãã¦å¤ããèªããéçºè ã¯ããã¯ãã¾ãããã¾ãã«æ°å¦ã¨åãã§ãä¸è¬çãªããã¸ã§ã¯ãã§çµ±è¨å¦ã話é¡ã«ä¸ããã¨ãªã©ãªãã®ã§ããã§ã¯ãæ°è¦ã«ã³ã¼ãã£ã³ã°ãããããå¤ãã³ã¼ãã®ã¡ã³ããã³ã¹ããããããåéã«ãæã空ãã®ã¯èª°ã§ããããï¼ ã¨ã³ã¸ãã¢ã®æ¹ã¯ããã²æéãä½ã£ã¦ãã ãããè¿é ã¯ã15åã§ãè²´éãªæéã¨è¨ããã§ãããããã ãã¡ãã®è¨äºãããã¯ãã¼ã¯ã«è¿½å ãã¦ããã¦ãããã§ããããã¨ã«
åºéç 究室ã¯ï¼è¦è¦ã®æ å ±å¦çãåºç¤ã¨ããç»åå¦çï¼æ©æ¢°å¦ç¿ï¼ç¥çµåè·¯ã¢ãã«ï¼æ·±å±¤å¦ç¿ï¼ã¨ãã£ãåéãä¸å¿ã«ç 究ãè¡ã£ã¦ãã¾ãï¼ ç¥çµåè·¯ã¢ãã«ãæ©æ¢°å¦ç¿ããã¼ã¹ã«ï¼è¨æ¸¬ç»åãªã©ã«å¯¾ããæ å ±å¦çæè¡ãæ¥å¤éçºãã¦ãã¾ãï¼èå³ãããæ¹ã¯ãé£çµ¡ãã ããï¼ äººéãåç©ã®è³ãã©ã®ãããªæ å ±å¦çã ãããªã£ã¦ãããã¯æªã 人é¡ã«ã¨ã£ã¦æªç¥ã®é åã¨ãªã£ã¦ãã¾ãï¼ããã¯æ å ±å¦çã®æ©æ§ãç°ãªã£ã¦ãããã¨ã«èµ·å ãã¦ããã¨èãããã¾ãï¼è³ã¯æ¯è¼çåç´ãªé¨åã§ãããã¥ã¼ãã³ã¨å¼ã°ããé¨åãããªããã£ã¦ããï¼ã²ã¨ã¤ã²ã¨ã¤ã® ãã¥ã¼ãã³ã¯é«ãè½åããã£ã¦ããããã§ã精度ã®é«ãåä½ããã¦ãããããããã¾ããï¼ããã©ãããçä½ã®å é¨ã®ãã¾ãã¾ãªéé³ã«ãã£ã¦é常ã«ä¸ç¢ºããªæ å ±å¦çãããããããªãç¶æ ã«ããã¾ãï¼ ã¨ãããï¼ãã®ãã¥ã¼ãã³ãè«å¤§ãªåæ°ï¼äººéã§ç´ï¼ï¼ï¼ååï¼éã¾ãã¨äººéã®ããã«ç©ãè¦ã¦ï¼è¨èãæãï¼æèãå·¡ããã¨
ãç¸é¢ãã£ã¦ä½ï¼ çµ±è¨å¦ãå¦ãã ãã¨ãããã°ã誰ããä¸åº¦ã¯èãããã¨ããããç¸é¢ãã¨ããè¨èãæè¿ã§ã¯ãé«æ ¡ã§ãæ±ãããæ¦å¿µã«ãªã£ãããã§ããããç¸é¢ãã£ã¦ãã£ããä½ã§ããããï¼æç§æ¸ï¼ã¨ããåã®Wikipediaï¼ã«ã¯ãããã風ã«æ¸ããã¦ãã¾ãã ç¸é¢ä¿æ°ï¼ãããããããããè±: correlation coefficientï¼ã¨ã¯ã2 ã¤ã®ç¢ºçå¤æ°ã®éã®ç¸é¢ï¼é¡ä¼¼æ§ã®åº¦åãï¼ã示ãçµ±è¨å¦çææ¨ã§ãããååãåä½ã¯ç¡ããâ1 ãã 1 ã®éã®å®æ°å¤ãã¨ãã1 ã«è¿ãã¨ãã¯2 ã¤ã®ç¢ºçå¤æ°ã«ã¯æ£ã®ç¸é¢ãããã¨ãããâ1 ã«è¿ããã°è² ã®ç¸é¢ãããã¨ããã0 ã«è¿ãã¨ãã¯ãã¨ã®ç¢ºçå¤æ°ã®ç¸é¢ã¯å¼±ããå ã¿ã« 1 ããã㯠â1 ã¨ãªãå ´å㯠2 ã¤ã®ç¢ºçå¤æ°ã¯ç·å½¢å¾å±ã®é¢ä¿ã«ããã ããã§ã¯ãç¸é¢ã§ã¯ãªããç¸é¢ä¿æ°ã¨ããè¨èãç¨ãã¦ãã¾ããããç¸é¢ãã¨ããã®ã¯ã2ã¤ä»¥ä¸ã®ãã¨ãããã¨ãã«ãããã
1. åé¡ é¢æ£ã®å ´å é£ç¶ã®å ´å HSIC å®é¨ ã¾ã¨ã Bayes Independence Test - HSIC ã¨æ§è½ãæ¯è¼ãã - é´æ¨ è² å¤§éªå¤§å¦ 人工ç¥è½å¦ä¼ åºæ¬åé¡ç ç©¶ä¼ (å®®å¤å³¶) 2014 å¹´ 3 æ 3 æ¥ Bayes Independence Test - HSIC ã¨æ§è½ãæ¯è¼ãã 1 / 19 3. åé¡ é¢æ£ã®å ´å é£ç¶ã®å ´å å®é¨ HSIC ã¾ã¨ã åé¡: (x1 , y1 ), · · · , (xn , yn ) ãããX ⥠Y ãå¦ããæ¤å® ⥠ç¸äºæ å ±é: I (X , Y ) := ââ x PXY (x, y ) log y PXY (x, y ) PX (x)PY (y ) I (X , Y ) = 0 ââ X ⥠Y ⥠Hilbert Schmidt independent criterion: ç¸é¢ä¿æ°ã®éç·åå ç¸é¢ä¿æ° (X
ç¸å¤ããããã¡ã®ãã¼ã ã§ã¯è«æ輪èªä¼ããã£ã¦ã¾ãã¦ãå æ¥åãæ å½ããã®ãã21ä¸ç´ã®ç¸é¢ã®æ¬å½ãHSIC (Hilbert-Schmidt Independence Criteria)ã®åè«æããããã¤âã§ããã Measuring Statistical Dependence with Hilbert-Schmidt Norms (Gretton et al., Algorithmic Learning Theory, 2005) HSICã¨è¨ãã°@motivic_æ°ã®æã«ããä¸æ¨å¹´å¤ã®TokyoRã§ã®çºè¡¨ãä¸çªãã¤ã¹ãªè§£èª¬ã ã¨æãã¾ãã 21ä¸ç´ã®ææ³å¯¾æ±º (MIC vs HSIC) è¿½è¨ HSICé¢é£ã ã¨é´æ¨å çã®ããããå§ã(æç´¹ä»ãããã Bayes Independence Test - HSIC ã¨æ§è½ãæ¯è¼ãã- http://t.co/IMWUpB5qEuâ motivi
1. . ã¯ããã« . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 ä¸ç´ã®ææ³å¯¾æ±º . . åè . ...... 21 ä¸ç´ã®ææ³å¯¾æ±º @motivic 第 33 å R åå¼·ä¼@æ±äº¬ 2013 å¹´ 8 æ 31 æ¥ @motivic 21 ä¸ç´ã®ææ³å¯¾æ±º 2. . ã¯ããã« . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 ä¸ç´ã®ææ³å¯¾æ±º . . åè ç®æ¬¡ ...1 ã¯ããã« èªå·±ç´¹ä» ...2 21 ä¸ç´ã®ææ³å¯¾æ±º 20 ä¸ç´ã®ç¬ç«æ§ãè¦ã¤ããæ¹æ³ã®éç 21 ä¸ç´ã®ç¸é¢ï¼MIC 21 ä¸ç´ã®æ¤å®ï¼HSIC test ããåè² ï¼ ...3 åè åèæç® åè HP @motivic 21 ä¸ç´ã®ææ³å¯¾æ±º 3. . ã¯ããã« . . .
ä»å¹´ã®1æã«ãããªè©±é¡ãåãä¸ããããã§ããã ãã®è¨äºã®æå¾ã«ã¡ããã£ã¨æ¸ããéããå®éã«ã¯ãããã"too good to be true"å³ã¡ããã®ã¢ãã«ã®ç²¾åº¦ãããä½ã§ãé«éããããããªãã®ï¼ãâãå®ã¯æ±åæ§è½è¦ã¦ã¾ããã§ãããã¿ãããªã±ã¼ã¹ã£ã¦ãæ³åãããé¥ãã«å¤ããã®ä¸ã®ä¸åå¨ããã¿ãããªãã§ãããã¨ãããã¨ã§ããããããã¯ããã¿ãã®2ç« ã¨ãPRMLã®æåã®æ¹ã«åºã¦ããåæ©ä¸ã®åæ©ãªãã§ããããã®è¾ºã®è©±ãæ¹ãã¦ã ãã ãæ¸ãã¦ã¿ããã¨æãã¾ãã ããããã精度100%ãã¨ããç¸é¢ä¿æ°0.9以ä¸ãã¨ãè¦ãã身æ§ããã¹ã åé ã«æããä¾ã¯ãããããã精度100%ãªãã¦ãããããã¨ãã声ããã¡ãã¡ããæãã£ããã¨ã§è©±é¡ã«ãªããèãéãã¦ã¿ããleakageã¯ãããè¨ç·´èª¤å·®ã§ããè©ä¾¡ãã¦ãªããã§æ£ã ã ã£ãããã§ãã ä¸è¬ã«ãå®ä¸çã®ãã¼ã¿ã»ããã§çµ±è¨ã¢ããªã³ã°ã«ããæ©æ¢°å¦ç¿ã«ããã¢ããªã³
Hadley Wickham, Lisa Stryjewski. 40 years of boxplots. Download: pre-print The boxplot plot has been around for over 40 years. This paper summarises the improvements, extensions and variations since Tukey first introduced his 'schematic plot in 1970. We focus particularly on richer displays of density and extensions to 2d. @TechReport{boxplots, author = {Hadley Wickham and Lisa Stryjewski}, instit
æ£è¦åå¸ (normal distribution) ã¯ã¬ã¦ã¹åå¸ (Gaussian distribution) ã¨ãå¼ã°ããé£ç¶ç¢ºçåå¸ã®ã²ã¨ã¤ã§ãããååã«å ããã¦ããã¬ã¦ã¹ã¨ã¯ã19ä¸ç´ã®ãã¤ãã®æ°å¦è Carolus Fridericus Gauss ã«ç±æ¥ãããçµ±è¨å¦ã«ããã¦ã¯çè«ä¸ã»å®ç¨ä¸ãåããæãéè¦ãªç¢ºçåå¸ã¨ããããå¿ç¨ã®ç¯å²ã¯éããªãåºããèªç¶ç§å¦ãã人æç§å¦ã«è³ãã¾ã§ã®å¤ãã®ç¾è±¡ã®ã¢ãã«åã«ç¨ãããããä¾ãã°ãã¹ãã¥ã¼ãã³ãã®tæ¤å®çã«ä»£è¡¨ããããã©ã¡ããªãã¯ãªçµ±è¨æ¤å®ã§ã¯å¤ãã®å ´åãæ¨æ¬å¤æ°ãæ£è¦åå¸ã«å¾ããã¨ãä»®å®ããã¦ãããã¾ãã身è¿ãªã¨ããã§ã¯ããã®æ§å¥æ¯ã®èº«é·ã®åå¸ã¯æ£è¦åå¸ã«å¾ããã¨ããããã¦ããããã ããä½éã¯æ£è¦åå¸ã§ã¯ãªãã¬ã³ãåå¸ã«å¾ãããã©ã¡ã¼ã¿ã¼ã¯æå¾ å¤ Î¼ ã¨åæ£ Ï2 ã§ãããæ£è¦åå¸ã¯ N(μ, Ï2) ã«ã¦ç¥è¨ãããã確çå¤æ°ãXã¨
Bokeh documentation# Bokeh is a Python library for creating interactive visualizations for modern web browsers. It helps you build beautiful graphics, ranging from simple plots to complex dashboards with streaming datasets. With Bokeh, you can create JavaScript-powered visualizations without writing any JavaScript yourself. Finding the right documentation resources# Bokehâs documentation consists
å æ¥è¡ããã第9åããã¼ã¿è§£æã®ããã®çµ±è¨ã¢ããªã³ã°å ¥éãèªæ¸ä¼ã«ã¦ã ãå¯è¦åã§ç解ãããã«ã³ãé£éã¢ã³ãã«ã«ãæ³ãã¨ããã¿ã¤ãã«ã§çºè¡¨ããã¦é ãã¾ããã çºè¡¨ã¹ã©ã¤ãã¯ä»¥ä¸ã§ãã å¯è¦åã§ç解ãããã«ã³ãé£éã¢ã³ãã«ã«ãæ³ from hoxo_m ãã®çºè¡¨ã¯ãã¿ã©ãã¼ãã«ç»å ´ããããã«ã³ãé£éã¢ã³ãã«ã«ãæ³(MCMC)ã®ã¢ã«ã´ãªãºã ã§ãããã¡ããããªã¹æ³ãã¨ãã®ãã¹ã»ãµã³ãã©ã¼ãã«ã¤ãã¦ãå¯è¦åãã¦ç解ãããã¨ããã話ã§ãã ããã«ã³ãé£éã¢ã³ãã«ã«ãæ³ãã¨ããã®ã¯ãåé¢ã ãè¦ãã¨é£ãããã§ããããã®çºè¡¨ã§ç解ãã¹ããã¤ã³ãã¯ã次ã®ã¹ã©ã¤ã 1æã«å縮ããã¦ãã¾ãã ãã®ãã¨ã念é ã«ç½®ãã¦ãããããã®ææ³ãè¦ã¦ããã¾ãããã ã¾ããã¡ããããªã¹æ³ã§ãããããã¯ã åã®ç¶æ ã®è¿ãã®ç¹ã次ã®é·ç§»å åè£ã¨ãã¦é¸ã¶ï¼ãã«ã³ãé£éï¼ ãã®ã¨ãã®ç¢ºçæ¯ r < 1 ãªãã°ç¢ºç r ã§æ£å´ããããã
ããã°ã©ãã¼ã®ããã®ç¢ºçããã°ã©ãã³ã°ã¨ãã¤ãºæ¨å®Â¶Pythonã¨PyMCã®ä½¿ãæ¹Â¶ãã¤ãºæ¨å®(Bayesian method)ã¯ï¼ç¢ºçæ¨è«ã®ããã®ãã£ã¨ãé©åãªã¢ããã¼ãã§ããã«ããããããï¼æ¸ç±ãèªãã¨ãã¼ã¸æ°ãæ°å¼ãå¤ãã®ã§ï¼ãã¾ãç©æ¥µçã«èªããã¨ããèªè ã¯å°ãªãã®ãç¾ç¶ã§ããï¼å ¸åçãªãã¤ãºæ¨å®ã®æç§æ¸ã§ã¯ï¼æåã®3ç« ã使ã£ã¦ç¢ºçã®çè«ã説æãï¼ãããããã¤ãºæ¨è«ã¨ã¯ä½ãã説æããï¼æ®å¿µãªããå¤ãã®ãã¤ãºã¢ãã«ã¯è§£æçã«è§£ããã¨ãå°é£ã§ããããï¼èªè ãç®ã«ããã®ã¯ç°¡åã§äººå·¥çãªä¾é¡ã°ããã«ãªã£ã¦ãã¾ãï¼ãã®ããï¼ãã¤ã¹æ¨è«ã¨èãã¦ããã ããä½ï¼ãã¨æã£ã¦ãã¾ãã®ã§ããï¼å®éï¼èè ã®ç§ãããæã£ã¦ããã®ã ããï¼ æè¿ã®æ©æ¢°å¦ç¿ã®ã³ã³ãã¹ãã§è¯ãæ績ãåãããã¨ãã§ããã®ã§ï¼ç§ã¯ãã®ãããã¯ã復ç¿ãããã¨æãç«ã£ãï¼ ç§ã¯æ°å¦ã«ã¯å¼·ãæ¹ã§ããï¼ãããããã§ãï¼ä¾é¡ã説æãèªãã§é ã®ä¸ã§
ã©ã³ãã³ã°
ã©ã³ãã³ã°
ã©ã³ãã³ã°
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}}