teddy-gã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ / 2016å¹´3æœˆ1æ—¥ - ã¯ã¦ãªãƒ

éžè² å€¤è¡Œåˆ—å› ååˆ†è§£ï¼ˆNMFï¼‰ã«ã‚ˆã‚‹ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‰ã®ã¡ã‚‡ã£ã¨ã—ãŸä¾‹ - About connecting the dots.

æœ€è¿‘ç·šå½¢ä»£æ•°ã«ã¤ã„ã¦ã„ã‚ã„ã‚èªã¿ãªãŠã—ãŸã‚Šã—ã¦ã‚‹ã®ã§ã™ãŒï¼ˆç·šå½¢ä»£æ•°ãƒãƒ¼ãƒˆã‚·ãƒ¼ãƒˆã‚’å‰ã®è¨˜äº‹ã§ã¾ã¨ã‚ã¦ã‚ã‚Šã¾ã™ï¼‰ï¼Œãã®ä¸€ç’°ã§ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‰ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã«ã¤ã„ã¦ã„ãã¤ã‹è©¦ã—ã¦ã¿ãŸã®ã§ï¼Œãã‚Œã‚’è§£èª¬ã—ã¾ã™ï¼Žé †åºã¨ã—ã¦ã¯ï¼ŒåŸºæœ¬ã®å”èª¿ãƒ•ã‚£ãƒ«ã‚¿ãƒªãƒ³ã‚°ï¼ˆãƒ¦ãƒ¼ã‚¶ãƒ™ãƒ¼ã‚¹ãƒ‰ï¼Œã‚¢ã‚¤ãƒ†ãƒ ãƒ™ãƒ¼ã‚¹ãƒ‰ï¼‰â†’ç‰¹ç•°å€¤åˆ†è§£ï¼ˆSVDï¼‰â†’éžè² å€¤è¡Œåˆ—å› ååˆ†è§£ï¼ˆNMFï¼‰ã«ãªã‚Šã¾ã™ï¼Ž åŸºæœ¬çš„ãªè€ƒãˆæ–¹ ã“ã“ã§å–ã‚Šæ‰±ã†ã®ã¯ï¼Œã™ã¹ã¦ä»¥ä¸‹ã®ã‚ˆã†ãªãƒ¦ãƒ¼ã‚¶Ã—å•†å“ã®ãƒžãƒˆãƒªãƒƒã‚¯ã‚¹ã‚’ãƒ™ãƒ¼ã‚¹ã¨ã—ãŸãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‰ã«ãªã‚Šã¾ã™*1ï¼Žã“ã“ã§ã¯æ˜ ç”»ãƒ¬ãƒ³ã‚¿ãƒ«ã‚µãƒ¼ãƒ“ã‚¹ã‚’ä¾‹ã«ã—ã¦è€ƒãˆã¾ã™ï¼Ž6äººã®ãƒ¦ãƒ¼ã‚¶ãŒï¼Œ4ã¤ã®æ˜ ç”»*2ã®ã†ã¡ãƒ¬ãƒ³ã‚¿ãƒ«è¦–è´ã—ãŸã‚‚ã®ã«ã¤ã„ã¦ã¯ï¼Œ1-5ç‚¹ã®5æ®µéšŽè©•ä¾¡ã‚’è¡Œã„ã¾ã—ãŸï¼Ž0ã«ãªã£ã¦ã„ã‚‹ã‚‚ã®ã¯ã€Œã¿ã¦ã„ãªã„ã€ã¨ã„ã†ã“ã¨ã«ãªã‚Šã¾ã™ï¼Ž ã¾ãšã¯ã–ã£ã¨è©•ä¾¡ã®çŠ¶æ³ã‚’ã¿ã‚‹ã¨ï¼Œã€Œåƒã¨åƒå°‹ã®ç¥žéš ã—ã€ãŒæœ€ã‚‚ã‚ˆãè¦–è´ã•ã‚Œã¦ã„ã¦ï¼Œ6äººä¸4äººãŒã¿ã¦ã„ã¾ã™ï¼Žæ¬¡ã«ã¿ã‚‰ã‚Œã¦ã„ã‚‹ã®ã¯ã€Œã¨ãªã‚Šã®ãƒˆãƒˆãƒã€

teddy-g 2016/03/01

ã‚¢ã‚¤ãƒ†ãƒ ãƒ™ãƒ¼ã‚¹ã€ãƒ¦ãƒ¼ã‚¶ãƒ™ãƒ¼ã‚¹ã®ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‰ã®åŸºæœ¬æ‰‹æ³•ã‹ã‚‰ã€SVD&NMFä½¿ã†æ–¹æ³•ã¾ã§ç°¡æ½”ã«èª¬æ˜Žã€‚ã‚¢ã‚¤ãƒ†ãƒ ã®é¡žä¼¼æ€§Ã—ãƒ¦ãƒ¼ã‚¶ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«ã§ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‰ã™ã‚‹éš›ã«ã€SVDæˆ–ã„ã¯NMFã§ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«ã‚’å†æ§‹ç¯‰ã™ã‚‹æ„Ÿã˜ã€‚

ãƒªãƒ³ã‚¯

Latent Dirichlet Allocation ã‚†ã‚‹ãµã‚å…¥é–€ - ã‚ã‚‰ã³ãæ—¥è¨˜

ã“ã®è¨˜äº‹ã¯ abicky.net ã® Latent Dirichlet Allocation (LDA) ã‚†ã‚‹ãµã‚å…¥é–€ ã«ç§»è¡Œã—ã¾ã—ãŸ

teddy-g 2016/03/01

ãƒˆãƒ”ãƒƒã‚¯ãƒ¢ãƒ‡ãƒ«ã®ã²ã¨ã¤ã€LDAã®è§£èª¬ã€‚æ·±ãè€ƒãˆãªã‘ã‚Œã°ã•ã‚‰ã£ã¨ç†è§£ã§ãã‚‹ã€‚æ·±ãç†è§£ã—ã‚ˆã†ã¨ã™ã‚‹ã¨æ·±ã¿ã«ã¯ã¾ã‚‹ã€‚

ãƒªãƒ³ã‚¯

ç·šå½¢ä»£æ•°ã®ç”¨èªžã¨æ„å‘³ã¾ã¨ã‚ï¼ˆä¸»ã«è‡ªåˆ†ç”¨ï¼‰ - About connecting the dots.

æ¥ãšã‹ã—ãªãŒã‚‰ï¼Œç·šå½¢ä»£æ•°å‘¨ã‚Šã®ç”¨èªžã£ã¦ä¼¼ãŸã‚ˆã†ãªã‚‚ã®ãŒå¤šãã¦ï¼Œã™ãã«ã‚¢ãƒ¬ãŒã©ã‚Œã ã£ã‘ã¨æ··åŒã—ã¦ã—ã¾ã„ãŒã¡ã«ãªã‚Šã¾ã™ï¼Žç·šå½¢ä»£æ•°ã®æ‰‹è¨ˆç®—ã¨ã‹ãŒã‚“ã°ã£ã¦ãŸã®ãªã£ã¦ã‚‚ã†10å¹´ã¨ã‹æ˜”ã®è©±ã ã—ï¼Œãƒãƒ¼ãƒˆã‚·ãƒ¼ãƒˆçš„ã«ã¾ã¨ã‚ãªãŠã—ã¦ãŠã“ã†ã¨æ€ã„ã¾ã—ãŸï¼Žå†…å®¹çš„ã«ã¯ï¼Œä¸»ã«çµ±è¨ˆã‚„æ©Ÿæ¢°å¦ç¿’ã§ä½¿ã†ã‚ˆã†ãªå†…å®¹ãŒä¸å¿ƒã«ãªã£ã¦ã„ã¾ã™ï¼Ž æ¦‚è¦ çµ±è¨ˆãƒ»æ©Ÿæ¢°å¦ç¿’ã§ä½¿ã†ç·šå½¢ä»£æ•°ã¯ï¼ŒåŸºæœ¬çš„ã«ã¯ä»¥ä¸‹ã€Œè¨ˆç®—ã®ç°¡ä¾¿åŒ–ã€ã¨ã€Œãƒ‡ãƒ¼ã‚¿å¤‰æ›ã€ã®2ã¤ãŒãƒ¡ã‚¤ãƒ³ã§ã™ï¼Žã‚‚ã¡ã‚ã‚“æ•°å¦çš„ã«çªã£è¾¼ã‚“ã§ã„ã£ãŸã‚Šï¼Œçµ±è¨ˆãƒ»æ©Ÿæ¢°å¦ç¿’ã§ã‚‚å¿œç”¨çš„ãªæ‰‹æ³•ã‚’ç”¨ã„ã‚‹å ´åˆã¯ãã®é™ã‚Šã§ã¯ãªã„ã§ã™ãŒï¼ŒåŸºæœ¬çš„ã«ã¯ä¸‹ã®2ã¤ãŒå¤§ãã„ã¨æ€ã„ã¾ã™*1ï¼Ž è¨ˆç®—ã®ç°¡ä¾¿åŒ– ï¼ˆä¾‹ãˆã°å›ºæœ‰å€¤ãƒ»å›ºæœ‰ãƒ™ã‚¯ãƒˆãƒ«ã‚’ç”¨ã„ã¦ï¼‰è¡Œåˆ—ã‚’å¯¾è§’åŒ–ã™ã‚‹ã“ã¨ã§ï¼Œè¡Œåˆ—ã®ä¹—ç®—ã‚’é«˜é€Ÿã«å®Ÿæ–½ã™ã‚‹ ï¼ˆLUåˆ†è§£ã‚’ç”¨ã„ã¦ï¼‰æ‰±ã„ã‚„ã™ã„å½¢ã«è¡Œåˆ—ã‚’åˆ†è§£ã™ã‚‹ã“ã¨ã§ï¼Œãã®å¾Œã®è¨ˆç®—ã‚’é«˜é€Ÿã«ã™ã‚‹ ãƒ‡ãƒ¼ã‚¿å¤‰æ› SVDã‚’è¡Œã†ã“ã¨ã§LSIã‚„PCAã¨ã„ã£ãŸãƒ‡ãƒ¼ã‚¿ç¸®

teddy-g 2016/03/01

å‚™å¿˜ç”¨ã«ã€‚

ãƒªãƒ³ã‚¯

Pythonã‚‰ã—ã„ã‚³ãƒ¼ãƒ‰ã®æ›¸ãæ–¹ - Kesinã®çŸ¥è¦‹ç½®ãå ´

ã¯ã¦ãªãƒ–ãƒã‚°ã«ç§»è¡Œã—ã¦æœ€åˆã®è¨˜äº‹ã¯ã‚„ã¯ã‚ŠPythonãƒã‚¿ã«ã—ã¾ã—ãŸã€‚ ã¯ã¦ãªãƒ–ãƒã‚°ã„ã„ã§ã™ãã€‚ãƒ‡ã‚¶ã‚¤ãƒ³ã®ç·¨é›†ãŒã‚„ã‚Šã‚„ã™ããªã£ãŸã®ã¨ã€Markdownã§æ›¸ã‘ã‚‹ã®ãŒç´ æ™´ã‚‰ã—ã„ã§ã™ã€‚ PyCon 2013ã®å‹•ç”»ã‚’è¦‹ã¦ã„ãŸã‚‰ã€ç´ æ™´ã‚‰ã—ã„"Transf orming Code into Beautiful, Idiomatic Python"ã¨ã„ã†ç™ºè¡¨ã‚’è¦‹ã¤ã‘ãŸã®ã§ãã®ã¾ã¨ã‚ã§ã™ã€‚ ä»Šã©ãã®Pythonã‚³ãƒ¼ãƒ‰ã®ãƒ™ã‚¿ãƒ¼ãªæ›¸ãæ–¹ã‚’ç´¹ä»‹ã—ã¦ã„ã¾ã™ã€‚ Transf orming Code into Beautiful, Idiomatic Python ... ã‚¹ãƒ©ã‚¤ãƒ‰ã¯ã“ã¡ã‚‰ã«ã‚ã‚Šã¾ã—ãŸ çµæ§‹é•·ããªã£ã¦ã—ã¾ã£ãŸã®ã§ã€çŸ¥ã£ã¦ã‚‹é …ç›®ã¯èªã¿é£›ã°ã—ã¦ã‚‚ã‚‰ãˆã°ã¨æ€ã„ã¾ã™ã€‚ ãƒ«ãƒ¼ãƒ—ã®åŸºæœ¬ æ•´æ•°ã®ãƒ«ãƒ¼ãƒ— ã¾ãšã¯åŸºæœ¬ã®ãƒ«ãƒ¼ãƒ—ã€‚ Cã®for int i=0; i<6; i++ã‚’Pythonã§å˜ç´”ã«æ›¸ãã¨ã“ã†ãªã‚Šã¾ã™ã€‚ for

teddy-g 2016/03/01

Pythonã‚‰ã—ã„ã‚³ãƒ¼ãƒ‰ã®æ›¸ãæ–¹ã¨ã„ã†ã“ã¨ã§TipsãŒã²ã¨ã¾ã¨ã‚ã«ãªã£ã¦ã„ã‚‹ã€‚å…ƒãƒã‚¿ã¯ãƒ“ãƒ‡ã‚ªã ãŒæ–‡å—ã«èµ·ã“ã—ã¦ã‚ã‚‹æ–¹ãŒå‚ç…§ã—ã‚„ã™ã„ã€‚

ãƒªãƒ³ã‚¯

1.3.5.16. Kolmogorov-Smirnov Goodness-of-Fit Test

The Kolmogorov-Smirnov test (Chakravart, Laha, and Roy, 1967) is used to decide if a sample comes from a population with a specific distribution. The Kolmogorov-Smirnov (K-S) test is based on the empirical distribution function (ECDF). Given N ordered data points Y1, Y2, ..., YN, the ECDF is defined as \[ E_{N} = n(i)/N \] where n(i) is the number of points less than Yi and the Yi are ordered from

teddy-g 2016/03/01

K-Sæ¤œå®šã®ä¸å¯§ãªèª¬æ˜Žã€‚äºŒã¤ã®ç¢ºçŽ‡åˆ†å¸ƒã®æœ€å¤§è·é›¢ã‚’çµ±è¨ˆé‡Dã¨ã—ã¦å¾—ã‚‹ã€‚på€¤ãŒä½Žã„ã»ã©äºŒã¤ã®åˆ†å¸ƒã¯ç•°ãªã£ã¦ã„ã‚‹ã€‚

ãƒªãƒ³ã‚¯

ç›¸é–¢ä¿‚æ•°ã®æœ‰æ„æ€§ã‚’ç¢ºã‹ã‚ã‚‹æ–¹æ³•ã«ã¤ã„ã¦ -ç›¸é–¢ä¿‚æ•°ã«ã¤ã„ã¦1æ©è¸ã¿è¾¼ã‚€- - Data Science by R and Python

ç›¸é–¢ä¿‚æ•°ã¨ã¯ï¼Ÿ ç›¸é–¢ä¿‚æ•°ã«ã¤ã„ã¦ã¯ã€ï¼˜æœˆã®ã‚¨ãƒ³ãƒˆãƒªãƒ¼ã‚’å‚ç…§ã—ã¦ã‚‚ã‚‰ã†ã¨è‰¯ã„ã‹ã¨æ€ã„ã¾ã™ã€‚ ã€Œç›¸é–¢ä¿‚æ•°ã€ã£ã¦ãªã‚“ã§ã™ã‹ï¼Ÿ -æ„å‘³ã¨åˆ©ç‚¹ã¨æ¬ ç‚¹ã‚’ã‚ã‹ã‚Šã‚„ã™ã- - Data Science by R and Python ç°¡å˜ã«èª¬æ˜Žã—ã¦ãŠãã¨ã€ç›¸é–¢ã¨ã„ã†ã®ã¯ã€Œ2ã¤ã®å¤‰æ•°ã®é–“ã«å˜åœ¨ã™ã‚‹ã€ç›´ç·šçš„é–¢ä¿‚ã€ã‚’æŒ‡ã—ã¦ã„ã¾ã™ã€‚ç›¸é–¢ä¿‚æ•°ã¨ã„ã†ã®ã¯ã€ãã®ç›´ç·šåº¦åˆã„ã‚’0~1ã®æ•°å—ã§è¡¨ã—ãŸã‚‚ã®ã¨è¨€ãˆã¾ã™ã€‚ãŸã ã—ã€å€¤ã«ã¯å¤§å°é–¢ä¿‚ã—ã‹ã‚ã‚Šã¾ã›ã‚“ï¼ˆç›¸é–¢ä¿‚æ•°0.8ã¯ã€ç›¸é–¢ä¿‚æ•°0.4ã®ã¨ãã‚ˆã‚Šã‚‚2å€ç›¸é–¢ãŒå¼·ã„ã¨ã„ã†æ„å‘³ã§ã¯ãªã„ã®ã§æ³¨æ„ã—ã¦ãã ã•ã„ï¼‰ã€‚ ã•ã¦ã€ãƒ‡ãƒ¼ã‚¿ã®åˆ†æžãƒ»è§£æžã«é–¢ã‚ã‚‰ãšã€ã„ãã¤ã‹ã®å¤‰æ•°ãŒã‚ã‚‹ãƒ‡ãƒ¼ã‚¿ã§ã¯ã€ç›¸é–¢ä¿‚æ•°ã‚’è¨ˆç®—ã—ã¦ã€å¤‰æ•°ã®é–“ã«é–¢ä¿‚ãŒã‚ã‚‹ã‹ã‚’ãƒã‚§ãƒƒã‚¯ã™ã‚‹ã®ãŒä¸€èˆ¬çš„ã§ã™ã€‚å‰ã®ã‚¨ãƒ³ãƒˆãƒªãƒ¼ã«ã‚‚ã‚ã‚‹ã‚ˆã†ã«ã“ã‚Œã¯ã€Œç›´ç·šçš„ãªé–¢ä¿‚ã€ã—ã‹èª¿ã¹ã‚‰ã‚Œãªã„ã§ã™ã—ã€è‰²ã€…ã¨åˆ¶ç´„ã‚‚ã‚ã‚Šã€ã„ã¡ã„ã¡ç¢ºèªã™ã‚‹ã®ã¯é¢å€’ã§ã™ãŒã€ãã‚Œã§ã‚‚ä½¿ã„å‹

teddy-g 2016/03/01

ç›¸é–¢ä¿‚æ•°ãŒæœ‰æ„ã‹ã©ã†ã‹ã‚’èª¿ã¹ã‚‹ãŸã‚ã«K-Sæ¤œå®šã‚’ä½¿ã†ã€‚âˆµæœ‰æ„ãªã‚‰æ£è¦åˆ†å¸ƒã«æ²¿ã£ã¦ã‚‹ã‹ã‚‰ã€‚ä»–ã«ãƒ’ã‚¹ãƒˆã‚°ãƒ©ãƒ ä½œã£ã¦ãƒ“ã‚¸ãƒ¥ã‚¢ãƒ«ã«ç¢ºèªã™ã‚‹ã€Q-Qãƒ—ãƒãƒƒãƒˆã‚’æãã¨ã„ã†æ‰‹æ³•ã‚‚ã€‚

ãƒªãƒ³ã‚¯

ã€çµ±è¨ˆå¦ã€‘Q-Qãƒ—ãƒãƒƒãƒˆã®ä»•çµ„ã¿ã‚’ã‚¢ãƒ‹ãƒ¡ãƒ¼ã‚·ãƒ§ãƒ³ã§ç†è§£ã™ã‚‹ã€‚ - Qiita

<tr> <th>185</th> <td>11</td> <td>B</td> <td>8600</td> <td>1K</td> <td>20.79</td> <td>åŒ—æ±</td> <td>0</td> </tr> <tr> <th>186</th> <td>8</td> <td>B</td> <td>7100</td> <td>1K</td> <td>22</td> <td>è¥¿</td> <td>17</td> </tr> <tr> <th>187</th> <td>9</td> <td>B</td> <td>18400</td> <td>1LDK</td> <td>54.68</td> <td>è¥¿</td> <td>10</td> </tr> ã“ã®ãƒ‡ãƒ¼ã‚¿ã®ãƒ€ã‚¦ãƒ³ãƒãƒ¼ãƒ‰ã¯ã“ã¡ã‚‰ã®çµ±è¨ˆå¦æ¤œå®š2ç´šã®å…¬å¼æ•™ç§‘æ›¸ã®ãƒšãƒ¼ã‚¸ä¸ç¨‹ã«ã‚ã‚‹ã€Œãƒ€ã‚¦ãƒ³ãƒãƒ¼ãƒ‰ç”¨ãƒ‡ãƒ¼ã‚¿ã€ã®ãƒªãƒ³ã‚¯ã‚ˆã‚Šè¡Œãˆã¾ã™ã€‚ãƒ€ã‚¦ãƒ³ãƒãƒ¼ãƒ‰ã—

teddy-g 2016/03/01

æ£è¦ç´¯ç©åˆ†å¸ƒé–¢æ•°ã¨è¨ˆæ¸¬ãƒ‡ãƒ¼ã‚¿ã‚’æ¯”è¼ƒã—ã¦ãƒ—ãƒãƒƒãƒˆã—ãŸã®ãŒQ-Qãƒ—ãƒãƒƒãƒˆã€‚ä¸€ç·’ã ã£ãŸã‚‰ãƒªãƒ‹ã‚¢ã«ãªã‚‹ã€‚

ãƒªãƒ³ã‚¯

ã‚¢ãƒ³ãƒ€ãƒ¼ã‚½ãƒ³â€“ãƒ€ãƒ¼ãƒªãƒ³ã‚°æ¤œå®š - Wikipedia

ã‚¢ãƒ³ãƒ€ãƒ¼ã‚½ãƒ³â€“ãƒ€ãƒ¼ãƒªãƒ³ã‚°æ¤œå®šï¼ˆã‚¢ãƒ³ãƒ€ãƒ¼ã‚½ãƒ³â€“ãƒ€ãƒ¼ãƒªãƒ³ã‚°ã‘ã‚“ã¦ã„ã€è‹±: Andersonâ€“Darling testï¼‰ã¯çµ±è¨ˆå¦ã«ãŠã‘ã‚‹ä»®èª¬æ¤œå®šã®ä¸€ç¨®ã§ã‚ã‚‹ [1]ã€‚æœ‰é™å€‹ã®æ¨™æœ¬ãŒå¸°ç„¡ä»®èª¬ã§æç¤ºã•ã‚ŒãŸåˆ†å¸ƒã¨ç•°ãªã£ã¦ã„ã‚‹ã‹ã©ã†ã‹ã‚’èª¿ã¹ã‚‹ãŸã‚ã«ç”¨ã„ã‚‰ã‚Œã‚‹ã€‚åŒæ§˜ã®æ¤œå®šã¨ã—ã¦ã‚³ãƒ«ãƒ¢ã‚´ãƒãƒ•â€“ã‚¹ãƒŸãƒ«ãƒŽãƒ•æ¤œå®šï¼ˆKSæ¤œå®šï¼‰ãŒã‚ã‚‹ãŒã€ã‚¢ãƒ³ãƒ€ãƒ¼ã‚½ãƒ³â€“ãƒ€ãƒ¼ãƒªãƒ³ã‚°æ¤œå®šã§ã¯ã€åˆ†å¸ƒã®è£¾ã§ã®ä¸€è‡´æ€§ï¼ˆãƒ†ãƒ¼ãƒ«åˆ†å¸ƒã®ä¸€è‡´æ€§ï¼‰ãŒã‚ˆã‚Šå¼·ãåæ˜ ã•ã‚Œã‚‹ãŸã‚ã€é‡‘èžåˆ†é‡Žãªã©ã®ãƒ†ãƒ¼ãƒ«ãƒªã‚¹ã‚¯ãŒé‡è¦ãªãƒ¢ãƒ‡ãƒ«ã®æ¤œå®šã«ä½¿ã‚ã‚Œã‚‹ã€‚ä»–æ–¹ã€KSæ¤œå®šã¨ç•°ãªã‚Šã€å¸°ç„¡ä»®èª¬ã®åˆ†å¸ƒã«ã‚ˆã£ã¦çµ±è¨ˆé‡ã®åŸºæº–å€¤ãŒå¤‰ã‚ã‚‹ã“ã¨ã«æ³¨æ„ãŒå¿…è¦ã€‚

teddy-g 2016/03/01

K-Sæ¤œå®šã®æ”¹å–„ç‰ˆã€‚K-Sæ¤œå®šã ã¨çœŸã‚“ä¸ã‚ãŸã‚Šã®é•ã„ã«ã‚»ãƒ³ã‚·ãƒ†ã‚£ãƒ–ã§ã€Anderson-Darlingã ã¨è£¾ã®æ–¹ã®é•ã„ã«ã‚»ãƒ³ã‚·ãƒ†ã‚£ãƒ–ã«ãªã‚‹ã€‚

ãƒªãƒ³ã‚¯

ã‚«ã‚¤ï¼’ä¹—åˆ†å¸ƒï¼Œã‚«ã‚¤ï¼’ä¹—æ¤œå®š

â– ã‚«ã‚¤ï¼’ä¹—åˆ†å¸ƒï¼ˆÏ‡2åˆ†å¸ƒï¼‰[chi-square distribution]ã€€ï½¥ï½¥ï½¥ã€€æ¯”çŽ‡ã®æ¤œå®š â€»ã€€ã“ã®ãƒšãƒ¼ã‚¸ã¯æŽ¨å®šï½¥æ¤œå®šã®å†…å®¹ã‚’å‰æã¨ã—ã¦ã„ã¾ã™ï¼Ž â—‡ç°¡å˜ãªä¾‹ã§ã‚¤ãƒ¡ãƒ¼ã‚¸ä½œã‚Š(1)â—‡ ä¾‹ï¼‘ æ—¥æœ¬äººã®ABOå¼è¡€æ¶²åž‹ã®åˆ†å¸ƒã¯ãŠã‚ˆãï¼¡åž‹40%ï¼ŒBåž‹20%ï¼ŒABåž‹10%ï¼ŒOåž‹30%ã ã¨ã„ã‚ã‚Œã¦ã„ã‚‹ï¼Žã‚ã‚‹æ‘ã§çŒ®è¡€ã«å¿œã˜ãŸè€…ã®ã†ã¡å…ˆç€100äººã®è¡€æ¶²åž‹ã¯æ¬¡ã®è¡¨ã®ã¨ãŠã‚Šã§ã‚ã£ãŸï¼Ž(ãŸã ã—ãƒ‡ãƒ¼ã‚¿ã¯æž¶ç©ºã®ã‚‚ã®ï¼‰

teddy-g 2016/03/01

ã‚«ã‚¤2ä¹—æ¤œå®šã®èª¬æ˜Žã€‚åº¦æ•°ã®å·®åˆ†ã®2ä¹—ã‚’æœŸå¾…åº¦æ•°ã§å‰²ã£ãŸã‚‚ã®ã‚’åˆè¨ˆã™ã‚‹ã€‚è‡ªç”±åº¦ã¯n-1ï¼ˆn>1ï¼‰ã€‚é¡žä¼¼æ€§ã‚’åˆ¤å®šã™ã‚‹ã‹ã‚‰åŸºæœ¬ã¯è·é›¢ãªã‚“ã ãªã‚ã€‚

ãƒªãƒ³ã‚¯

ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

ã‚¿ã‚°

2016å¹´3æœˆ1æ—¥ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (9ä»¶)

éžè² å€¤è¡Œåˆ—å› ååˆ†è§£ï¼ˆNMFï¼‰ã«ã‚ˆã‚‹ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‰ã®ã¡ã‚‡ã£ã¨ã—ãŸä¾‹ - About connecting the dots.

Latent Dirichlet Allocation ã‚†ã‚‹ãµã‚å…¥é–€ - ã‚ã‚‰ã³ãæ—¥è¨˜

ç·šå½¢ä»£æ•°ã®ç”¨èªžã¨æ„å‘³ã¾ã¨ã‚ï¼ˆä¸»ã«è‡ªåˆ†ç”¨ï¼‰ - About connecting the dots.

Pythonã‚‰ã—ã„ã‚³ãƒ¼ãƒ‰ã®æ›¸ãæ–¹ - Kesinã®çŸ¥è¦‹ç½®ãå ´

1.3.5.16. Kolmogorov-Smirnov Goodness-of-Fit Test

ç›¸é–¢ä¿‚æ•°ã®æœ‰æ„æ€§ã‚’ç¢ºã‹ã‚ã‚‹æ–¹æ³•ã«ã¤ã„ã¦ -ç›¸é–¢ä¿‚æ•°ã«ã¤ã„ã¦1æ©è¸ã¿è¾¼ã‚€- - Data Science by R and Python

ã€çµ±è¨ˆå¦ã€‘Q-Qãƒ—ãƒãƒƒãƒˆã®ä»•çµ„ã¿ã‚’ã‚¢ãƒ‹ãƒ¡ãƒ¼ã‚·ãƒ§ãƒ³ã§ç†è§£ã™ã‚‹ã€‚ - Qiita

ã‚¢ãƒ³ãƒ€ãƒ¼ã‚½ãƒ³â€“ãƒ€ãƒ¼ãƒªãƒ³ã‚°æ¤œå®š - Wikipedia

ã‚«ã‚¤ï¼’ä¹—åˆ†å¸ƒï¼Œã‚«ã‚¤ï¼’ä¹—æ¤œå®š

ãŠçŸ¥ã‚‰ã›

ä»Šé€±ã®ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯æ•°ãƒ©ãƒ³ã‚ãƒ³ã‚°ï¼ˆ2024å¹´12æœˆç¬¬4é€±ï¼‰

ã€Œã‚ã¨ã§èªã‚€ã€ã‚¿ã‚°ã§æŒ¯ã‚Šè¿”ã‚‹2024å¹´ ã€œä»Šå¹´ã®ã€Œã‚ã¨ã§èªã‚€ã€ã€ä»Šå¹´ã®ã†ã¡ã«ã€œ

ä»Šé€±ã®ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯æ•°ãƒ©ãƒ³ã‚ãƒ³ã‚°ï¼ˆ2024å¹´12æœˆç¬¬3é€±ï¼‰

å…¬å¼Twitter

ã‚ãƒ¼ãƒœãƒ¼ãƒ‰ã‚·ãƒ§ãƒ¼ãƒˆã‚«ãƒƒãƒˆä¸€è¦§

ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

å…¬å¼Twitter

ã¯ã¦ãªã®ã‚µãƒ¼ãƒ“ã‚¹

ã‚¿ã‚°

2016å¹´3æœˆ1æ—¥ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (9ä»¶)

ãŠçŸ¥ã‚‰ã›

ä»Šé€±ã®ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯æ•°ãƒ©ãƒ³ã‚­ãƒ³ã‚°ï¼ˆ2024å¹´12æœˆç¬¬4é€±ï¼‰

ã€Œã‚ã¨ã§èª­ã‚€ã€ã‚¿ã‚°ã§æŒ¯ã‚Šè¿”ã‚‹2024å¹´ ã€œä»Šå¹´ã®ã€Œã‚ã¨ã§èª­ã‚€ã€ã€ä»Šå¹´ã®ã†ã¡ã«ã€œ

ä»Šé€±ã®ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯æ•°ãƒ©ãƒ³ã‚­ãƒ³ã‚°ï¼ˆ2024å¹´12æœˆç¬¬3é€±ï¼‰

å…¬å¼Twitter

ã‚­ãƒ¼ãƒœãƒ¼ãƒ‰ã‚·ãƒ§ãƒ¼ãƒˆã‚«ãƒƒãƒˆä¸€è¦§

å…¬å¼Twitter

ã¯ã¦ãªã®ã‚µãƒ¼ãƒ“ã‚¹

2016å¹´3æœˆ1æ—¥ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (9ä»¶)

ãŠçŸ¥ã‚‰ã›

ä»Šé€±ã®ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯æ•°ãƒ©ãƒ³ã‚ãƒ³ã‚°ï¼ˆ2024å¹´12æœˆç¬¬4é€±ï¼‰

ã€Œã‚ã¨ã§èªã‚€ã€ã‚¿ã‚°ã§æŒ¯ã‚Šè¿”ã‚‹2024å¹´ ã€œä»Šå¹´ã®ã€Œã‚ã¨ã§èªã‚€ã€ã€ä»Šå¹´ã®ã†ã¡ã«ã€œ

ä»Šé€±ã®ã¯ã¦ãªãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯æ•°ãƒ©ãƒ³ã‚ãƒ³ã‚°ï¼ˆ2024å¹´12æœˆç¬¬3é€±ï¼‰

å…¬å¼Twitter

ã‚ãƒ¼ãƒœãƒ¼ãƒ‰ã‚·ãƒ§ãƒ¼ãƒˆã‚«ãƒƒãƒˆä¸€è¦§

å…¬å¼Twitter

ã¯ã¦ãªã®ã‚µãƒ¼ãƒ“ã‚¹