[B! Machine Learning] yyamanoã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

yyamano id:yyamano

Machine Learningã«é–¢ã™ã‚‹yyamanoã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (4)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

ãƒ©ãƒ³ãƒ€ãƒ ãƒ•ã‚©ãƒ¬ã‚¹ãƒˆã®ç†è«–ã¨é‡è¦ãªç‰¹å¾´é‡ã®é¸å®š - drilldripperâ€™s blog
ãƒ©ãƒ³ãƒ€ãƒ ãƒ•ã‚©ãƒ¬ã‚¹ãƒˆã¨æ±ºå®šæœ¨å¦ç¿’ ãƒ©ãƒ³ãƒ€ãƒ ãƒ•ã‚©ãƒ¬ã‚¹ãƒˆã‚’ç†è§£ã™ã‚‹ãŸã‚ã«ã¯ã€æ±ºå®šæœ¨å¦ç¿’ã®æ‰‹æ³•ã«ã¤ã„ã¦ç†è§£ã™ã‚‹å¿…è¦ãŒã‚ã‚Šã¾ã™ã€‚ã¾ãšæœ€åˆã«æ±ºå®šæœ¨å¦ç¿’ã®ç†è«–ã«ã¤ã„ã¦èª¬æ˜Žã—ã¾ã™ã€‚ æ±ºå®šæœ¨å¦ç¿’ æ±ºå®šæœ¨ã¯è¦ªã‹ã‚‰é †ã«æ¡ä»¶åˆ†å²ã‚’è¾¿ã£ã¦ã„ãã“ã¨ã§ã€çµæžœã‚’å¾—ã‚‹æ‰‹æ³•ã§ã™ã€‚ä¸‹ã¯æ±ºå®šæœ¨ã®ã‚¤ãƒ¡ãƒ¼ã‚¸ã§ã™ã€‚ æ±ºå®šæœ¨å¦ç¿’ã¨ã¯ãƒ‡ãƒ¼ã‚¿ã®å¿œã˜ã¦ä¸Šã®å›³ã®ã‚ˆã†ãªæ±ºå®šæœ¨ã‚’æ§‹æˆã—ã€åˆ†é¡žã‚’è¡Œã†æ©Ÿæ¢°å¦ç¿’ã®æ‰‹æ³•ã®ã“ã¨ã‚’æŒ‡ã—ã¾ã™ã€‚ æ±ºå®šæœ¨å¦ç¿’ã¯ã€ãƒ‡ãƒ¼ã‚¿ã®ç¨®é¡žã«å¿œã˜ã¦æ±ºå®šæœ¨ã‚’æˆé•·ã•ã›ã¦ã„ãã¾ã™ã€‚ æ±ºå®šæœ¨ã®åˆ†é¡žæ¡ä»¶ã¯ã€ãƒ‡ãƒ¼ã‚¿ã‚’åˆ†é¡žã—ãŸã¨ãã®æƒ…å ±åˆ©å¾—IG(Infomation Gain)ãŒæœ€å¤§ã«ãªã‚‹ã‚ˆã†ã«ã™ã‚‹ã“ã¨ã§ã™ã€‚æƒ…å ±åˆ©å¾—ã¯å¼(1)ã§è¡¨ã•ã‚Œã¾ã™ã€‚ ã¯è¦ªã®ãƒ‡ãƒ¼ã‚¿ã€ã¯ãƒŽãƒ¼ãƒ‰ã€ã¯æ³¨ç›®ã—ã¦ã„ã‚‹ãƒ‡ãƒ¼ã‚¿ã‚’è¡¨ã—ã¾ã™ã€‚ ã¯æœ¨ã‚’åˆ†å‰²ã™ã‚‹ãƒŽãƒ¼ãƒ‰æ•°ã§ã™ã€‚ä¸€èˆ¬çš„ã«æ±ºå®šæœ¨ã¯äºŒåˆ†æœ¨ã¨ã—ã¦å®Ÿè£…ã•ã‚Œã‚‹ã®ã§ã€ã»ã¨ã‚“ã©ã®å ´åˆã¯ã¨ãªã‚Šã¾ã™ã€‚ ã¯ä¸ç´”åº¦ã¨ã„ã†æŒ‡æ¨™ã§ã€å«ã¾ã‚Œã‚‹ãƒ‡ãƒ¼ã‚¿ã«åã‚ŠãŒã‚ã‚‹
yyamano 2018/11/15
Machine Learning
ãƒªãƒ³ã‚¯
SVMã¨ãƒ©ãƒ³ãƒ€ãƒ ãƒ•ã‚©ãƒ¬ã‚¹ãƒˆã®ã©ã¡ã‚‰ã®æ‰‹æ³•ã‚’ä½¿ãˆã°ã‚ˆã„ã‹ï¼Ÿ - drilldripperâ€™s blog
åˆ†é¡žã‚„å›žå¸°ã®å•é¡Œã‚’æ‰±ã†å ´åˆã€é¸æŠžã™ã‚‹æ‰‹æ³•ã¨ã—ã¦ã‚µãƒãƒ¼ãƒˆãƒ™ã‚¯ã‚¿ãƒ¼ãƒžã‚·ãƒ³(SVM)ã¨ãƒ©ãƒ³ãƒ€ãƒ ãƒ•ã‚©ãƒ¬ã‚¹ãƒˆãŒå€™è£œã«ä¸ŠãŒã£ã¦ãã‚‹ã¨æ€ã„ã¾ã™ã€‚ ã—ã‹ã—ã€ã©ã¡ã‚‰ã®æ‰‹æ³•ã‚’ä½¿ã†ã¹ããªã®ã§ã—ã‚‡ã†ã‹ï¼Ÿã©ã®ã‚ˆã†ãªå•é¡Œã«å¯¾ã—ã¦ã‚‚ã€ä¸€æ–¹ã®æ‰‹æ³•ã‚’ä½¿ã„ç¶šã‘ã‚Œã°è‰¯ã„ã®ã§ã—ã‚‡ã†ã‹ï¼Ÿãã‚Œã¨ã‚‚å•é¡Œã«ã‚ˆã£ã¦ä½¿ã„åˆ†ã‘ã‚‹å¿…è¦ãŒã‚ã‚‹ã®ã§ã—ã‚‡ã†ã‹ï¼Ÿ æ‰‹æ³•ã®æ‰‹è»½ã• Python æ©Ÿæ¢°å¦ç¿’ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ã®è‘—è€…ã§ã‚ã‚‹Sebastian Raschkaæ°ã¯æ¬¡ã®ã‚ˆã†ã«è¿°ã¹ã¦ã„ã¾ã™ã€‚ I would say that random forests are probably THE â€œworry-freeâ€ approach - if such a thing exists in ML: There are no real hyperparameters to tune (maybe except for the number of trees; typ
yyamano 2018/11/15
Machine Learning
ãƒªãƒ³ã‚¯
Deep Learning Tutorials â€” DeepLearning 0.1 documentation
Deep Learning TutorialsÂ¶ Deep Learning is a new area of Machine Learning research, which has been introduced with the objective of moving Machine Learning closer to one of its original goals: Artificial Intelligence. See these course notes for a brief introduction to Machine Learning for AI and an introduction to Deep Learning algorithms. Deep Learning is about learning multiple levels of represen
yyamano 2012/11/26
Machine Learning
ãƒªãƒ³ã‚¯
ç¬¬1å›žã€€æ©Ÿæ¢°å¦ç¿’ ã“ã¨ã¯ã˜ã‚ | gihyo.jp
æ¬¡ã®ã‚µãƒ¼ãƒ“ã‚¹ã‚„è£½å“ã¯ã©ã‚Œã‚‚èº«è¿‘ã«ã‚ã‚Šã¾ã™ãŒã€ã“ã‚Œã‚‰ã«å…±é€šã—ã¦ã„ã‚‹ã“ã¨ã¯ãªã‚“ã§ã—ã‚‡ã†ã€‚ Amazonã®ã€Œã“ã®å•†å“ã‚’è²·ã£ãŸäººã¯ã“ã‚“ãªå•†å“ã‚‚è²·ã£ã¦ã„ã¾ã™ã€ ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ã®ã€Œé–¢é€£ã‚¨ãƒ³ãƒˆãƒªãƒ¼ã€ Google ç¿»è¨³ Google æ—¥æœ¬èªžå…¥åŠ› ãƒ¡ãƒ¼ãƒ«ã‚¯ãƒ©ã‚¤ã‚¢ãƒ³ãƒˆã®ã‚¹ãƒ‘ãƒ ãƒ•ã‚£ãƒ«ã‚¿ ãƒ‡ã‚¸ã‚¿ãƒ«ã‚«ãƒ¡ãƒ©ã®è‡ªå‹•é¡”èªè˜ ãƒ‹ãƒ³ãƒ†ãƒ³ãƒ‰ãƒ¼DSã®æ‰‹æ›¸ãæ–‡å—èªè˜ è²·ã„ç‰©å±¥æ´ã€ãƒ¦ãƒ¼ã‚¶ãŒæ›¸ã„ãŸã‚³ãƒ¡ãƒ³ãƒˆã‚„ã‚¿ã‚°ã€Webã«ç„¡æ•°ã«ã‚ã‚‹ãƒšãƒ¼ã‚¸ã€ãƒ¡ãƒ¼ãƒ«ã€ç”»åƒã‚„å‹•ç”»ã¨å¯¾è±¡ã¯ãã‚Œãžã‚Œç•°ãªã£ã¦ã„ã¾ã™ãŒã€ã©ã‚Œã‚‚ã€Œãƒ‡ãƒ¼ã‚¿ã‹ã‚‰æœ‰ç›Šãªæƒ…å ±ã‚’å–ã‚Šå‡ºã™ã€ã¨ã„ã†ã“ã¨ã‚’è¡Œã£ã¦ã„ã¾ã™ã€‚ ã“ã‚Œã‚‰ã¯ã€Œæ©Ÿæ¢°å¦ç¿’ã€ã¨ã„ã†æŠ€è¡“ã‚’ä½¿ã£ã¦å®Ÿç¾ã•ã‚Œã¦ã„ã‚‹ã®ã§ã™ã€‚ æ©Ÿæ¢°å¦ç¿’ã®å¿œç”¨ç¯„å›² æ©Ÿæ¢°å¦ç¿’ã¯å†’é ã§æŒ™ã’ãŸä»¥å¤–ã«ã‚‚ã€æ§˜ã€…ãªåˆ†é‡Žã§ä½¿ã‚ã‚Œã¦ã„ã¾ã™ã€‚ ä¾‹ãˆã°ã€ãƒŽã‚¤ã‚ºé™¤åŽ»ã‚„ç‰¹å¾´ã®æŠ½å‡ºã‚’ç›®çš„ã¨ã—ãŸåˆ©ç”¨ãƒ‘ã‚¿ãƒ¼ãƒ³ãŒã‚ã‚Šã¾ã™ã€‚éŸ³å£°èªè˜ã‚„ç”»åƒèªè˜ã€æ–‡å—èªè˜ï¼ˆOCRï¼‰ãªã©ã¯ãã®ä»£è¡¨æ ¼ã§ã™ã€‚ãã‚Œã‚‰ã‚‚
yyamano 2010/06/25
Machine Learning
ãƒªãƒ³ã‚¯
1