{"blog_url":"https://www.mof-mof.co.jp/tech-blog/","blog_title":"\u3082\u3075\u3082\u3075\u6280\u8853\u90e8","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fwww.mof-mof.co.jp%2Ftech-blog%2Fmachine-learning-gradient-descent\" title=\"\u6700\u6025\u964d\u4e0b\u6cd5\u3092\u5b9f\u88c5\u3057\u3066\u7dda\u5f62\u56de\u5e30\u306e\u03b8\u306e\u5024\u3092\u63a2\u7d22\u3057\u3066\u307f\u305f\u3051\u3069\u3046\u307e\u304f\u3044\u304b\u306a\u304b\u3063\u305f - \u3082\u3075\u3082\u3075\u6280\u8853\u90e8\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","url":"https://www.mof-mof.co.jp/tech-blog/machine-learning-gradient-descent","image_url":"https://images.ctfassets.net/683ogktz4b18/2UDT6TcFP49lB8GUFVRqsg/949464233f9a7e99349a071aed1424f7/graph.png","categories":["machine learning","Coursera","Octave","Gradient Descent"],"title":"\u6700\u6025\u964d\u4e0b\u6cd5\u3092\u5b9f\u88c5\u3057\u3066\u7dda\u5f62\u56de\u5e30\u306e\u03b8\u306e\u5024\u3092\u63a2\u7d22\u3057\u3066\u307f\u305f\u3051\u3069\u3046\u307e\u304f\u3044\u304b\u306a\u304b\u3063\u305f","author_name":"redhornet96","published":"2016-01-09 00:00:00","provider_url":"https://hatena.blog","description":"\u524d\u56de\u306eOctave\u3067\u6563\u5e03\u56f3\u3092\u30d7\u30ed\u30c3\u30c8\u3057\u3066\u307f\u308b\u306b\u7d9a\u3044\u3066\u3001\u4eca\u5f8c\u306fOctave\u3067\u6700\u6025\u964d\u4e0b\u6cd5\u3092\u5b9f\u88c5\u3057\u3066\u3001\u03b8\u306e\u5024\u3092\u63a2\u7d22\u3057\u3066\u307f\u307e\u3059\u3002 \u7d50\u8ad6\u304b\u3089\u8a00\u3046\u3068\u5931\u6557\u3057\u305f\u3002\u3069\u3053\u304bfeature scaling\u3067\u9593\u9055\u3063\u3066\u3044\u308b\u3063\u307d\u3044\u3093\u3060\u3051\u3069\u3001\u3069\u3046\u9593\u9055\u3063\u3066\u3044\u308b\u306e\u304b\u304c\u308f\u304b\u3089\u305a\u3002\u6539\u3081\u3066\u30ea\u30d9\u30f3\u30b8\u3057\u305f\u3044\u3002\u3072\u3068\u307e\u305a\u306f\u8a18\u9332\u3092\u6b8b\u3057\u3066\u304a\u304f\u3002 \u524d\u306b\u5b9f\u88c5\u3057\u305f\u76ee\u7684\u95a2\u6570(costFunctionJ.m)\u3002 function J = costFunctionJ(X, y, theta) m = size(X,1); predictions = X*theta; sqrErrors = (predictions-y).^2; J = 1 / (2*m)\u2026","provider_name":"Hatena Blog","type":"rich","author_url":"https://blog.hatena.ne.jp/redhornet96/","width":"100%","version":"1.0","height":"190"}