競馬予測ã®è«–文紹介 - stockedge.jpã®æŠ€è¡“メモã©ã†ã‚„らã€ç§ã®ãƒ–ãƒã‚°ã‚’見ã¦ç«¶é¦¬ã®äºˆæ¸¬ã‚’始ã‚ãŸã¨ã„ã†äººãŒã¡ã‚‰ã»ã‚‰ã„るよã†ã 。ãã†ã„ã†äººãŸã¡ã®åŠ©ã‘ã«ãªã‚‹ã‹ã‚‚ã—ã‚Œãªã„ã®ã§ã€ç§ãŒèªã‚“ã 競馬予測ã®è«–文（ã“ã‚Œã¾ã§ã®è¨˜äº‹ã§ç´¹ä»‹ã—ãªã‹ã£ãŸã‚‚ã®ï¼‰ã‚’ã„ãã¤ã‹ç´¹ä»‹ã™ã‚‹ã€‚ Comparing the effectiveness of one-and two-step conditional logit models for predicting outcomes in a speculative market 競馬予測ã§ã¯æ¡ä»¶ä»˜ããƒã‚¸ãƒƒãƒˆãƒ¢ãƒ‡ãƒ«ã‚’使ã£ãŸäºˆæ¸¬ãŒæœ‰åŠ¹ãªã®ã ãã†ã 。æ¡ä»¶ä»˜ããƒã‚¸ãƒƒãƒˆãƒ¢ãƒ‡ãƒ«ã¨ã¯ã€å¤šé …ãƒã‚¸ãƒƒãƒˆãƒ¢ãƒ‡ãƒ«ã¨ä¼¼ã¦ã„ã‚‹ãŒã¡ã‚‡ã£ã¨é•ã†*1モデル。ã“ã®è§£èª¬ãŒã‚ã‹ã‚Šã‚„ã™ã„ã‹ã‚‚。 ã“ã®è«–æ–‡ã§ã¯ã€2ステップã®ãƒ¢ãƒ‡ãƒ«ä½œæˆæ–¹æ³•ã‚’æ案ã—ã¦ã„る。ã¾ãšãƒ•ã‚¡ãƒ³ãƒ€ãƒ¡ãƒ³ã‚¿ãƒ«å¤‰æ•°ã ã‘ã‚’ç´ æ€§ã¨ã—ã¦ä½¿ã£ã¦æ¡ä»¶ä»˜ããƒã‚¸ãƒƒãƒˆãƒ¢ãƒ‡ãƒ«ã®ãƒ‘ラメータを推定ã—ã€ã“れをステップ1ã¨ã™ã‚‹ã€‚次ã«ã€ã‚¹
LSTMを使ã£ã¦ã‚ºãƒ³ãƒ‰ã‚³ã‚ヨシをå¦ç¿’ã™ã‚‹ - Qiita
Matrix Factorizationã¨ã¯ - Qiita
15æ³ã§ãƒ‡ãƒ“ューã—ãŸèŠ¸èƒ½ç•Œã€å¤§å¦ã§å§‹ã‚ãŸãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°â€¦å¶ç„¶ãŒä½œã£ãŸè‡ªåˆ†ã ã‘ã®ç”Ÿã方。â€ã‚®ãƒ¼ã‚¯ãªå¥³å„ªâ€ãŒæ©ã‚“ã§ããŸã€ã“ã‚Œã¾ã§ã®é“程 | meeta MAGAZINE　(ミータマガジン)
https://scikit-learn.org/stable/modules/tree.html
iPython-Notebooks / Comparing machine learning classifiers based on their hyperplanes or decision boundaries.ipynb / Jupyter Notebook Viewer
Bayesian Optimizationã«é–¢ã™ã‚‹ãƒ¡ãƒ¢ | å°„æ’ƒã—ã¤ã¤å‰è»¢
ãƒã‚¸ã‚¹ãƒ†ã‚£ãƒƒã‚¯å›žå¸°
第18回　ãƒã‚¸ã‚¹ãƒ†ã‚£ãƒƒã‚¯å›žå¸°ï¼šæ©Ÿæ¢°å¦ç¿’ ã¯ã˜ã‚よã†ï½œgihyo.jp … 技術評論社
æ¼¢ã®zsh (6) æ¼¢ã¯ã ã¾ã£ã¦å…ˆæ–¹äºˆæ¸¬ - ã‚ãªãŸã¯ã“ã®æ©Ÿèƒ½ä½¿ã„ã“ãªã›ã‚‹ã‹ | マイナビニュース
ズンドコã‚ヨシ.jl with Iteration （追記ã‚り） - Qiita
TdX講演会#01「ãƒãƒ¼ãƒ ã¨é–‹ç™ºè€…Kaigiã¥ãり〠(2016/04/22 18:30〜)
http://www.comico.jp/challenge/articleList.nhn?titleNo=14519
論文メモ：NLP2016 - skozawa's blog
AlphaGoã®ã‚µãƒ¼ãƒæ–™é‡‘ã¯æœ€ä½Ž60億円？ - ãµãゆãブãƒã‚°
{{#tags}}- {{label}}
{{/tags}}