{"id":360,"date":"2014-09-20T20:49:58","date_gmt":"2014-09-20T11:49:58","guid":{"rendered":"https:\/\/www.dogrow.net\/octave\/?p=360"},"modified":"2019-10-26T20:41:09","modified_gmt":"2019-10-26T11:41:09","slug":"blog34","status":"publish","type":"post","link":"https:\/\/www.dogrow.net\/octave\/blog34\/","title":{"rendered":"(34) \u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u3067sin\u95a2\u6570\u3092\u8fd1\u4f3c(\u305d\u306e2)"},"content":{"rendered":"

(33) \u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u3067sin\u95a2\u6570\u3092\u8fd1\u4f3c<\/a> \u3067\u306f\u30013\u5c64(1-3-1)\u306e\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306bsin\u95a2\u6570\u3092\u5b66\u7fd2\u3055\u305b\u3066\u307f\u305f\u3002<\/p>\n

\u4eca\u56de\u306f\u4e0b\u8868\u306e\uff13\u7a2e\u985e\u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u69cb\u6210\u3067\u6bd4\u8f03\u3057\u3066\u307f\u305f\u3002<\/p>\n\n\n\n\n\n
1<\/td>\n3\u5c64<\/td>\n[I 1] – [H 3] – [O 1]<\/td>\n\u9752<\/td>\n<\/tr>\n
2<\/td>\n3\u5c64<\/td>\n[I 1] – [H 5] – [O 1]<\/td>\n\u7dd1<\/td>\n<\/tr>\n
3<\/td>\n4\u5c64<\/td>\n[I 1] – [H 5] – [H 3] – [O 1]<\/td>\n\u6c34\u8272<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

\"nnet_sin_2\"
\n\u96a0\u308c\u5c64\u306e\u30e6\u30cb\u30c3\u30c8\u6570\u304c\u591a\u3044\u65b9\u304c\u3001\u5b66\u7fd2\u958b\u59cb\u76f4\u5f8c\u306e\u5b66\u7fd2\u9032\u6357\u304c\u901f\u3044\u3088\u3046\u3060\u3002
\n\u554f\u984c\u304c\u7c21\u5358\u306a\u305f\u3081\u304b\u300150\u30a8\u30dd\u30c3\u30af\u8fba\u308a\u304b\u3089\u306f\u5dee\u304c\u308f\u304b\u3089\u306a\u3044\u3002<\/p>\n

\r\npkg load nnet<\/span>\r\nrand('state', 0)\r\nrandn('state', 0)\r\n\r\n% \u5b66\u7fd2\u30c7\u30fc\u30bf\u3092\u4f5c\u6210<\/span>\r\ndata_IN  = [ -pi*2 : 0.1 : pi*2 ];\r\ndata_OUT = sin(data_IN);\r\nRx2      = [min(data_IN), max(data_IN)];\r\n\r\n% 3\u5c64\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af(#1)\u3092\u4f5c\u6210<\/span>\r\nS01 = [3 1];\r\nNNet01 = newff<\/span>( Rx2, S01 );\r\nNNet01.trainParam.epochs = 60;\r\n\r\n% 3\u5c64\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af(#2)\u3092\u4f5c\u6210<\/span>\r\nS02 = [5 1];\r\nNNet02 = newff<\/span>( Rx2, S02 );\r\nNNet02.trainParam.epochs = 60;\r\n\r\n% 4\u5c64\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092\u4f5c\u6210<\/span>\r\nS03 = [5 3 1];\r\nNNet03 = newff<\/span>( Rx2, S03 );\r\nNNet03.trainParam.epochs = 60;\r\n\r\n% \u5b66\u7fd2\u3092\u5b9f\u884c<\/span>\r\nVV.P = data_IN;\r\nVV.T = data_OUT;\r\n\r\n% net1<\/span>\r\nNNet01 = train<\/span>( NNet01, data_IN, data_OUT, [], [], VV );\r\nres01  = sim<\/span>( NNet01, data_IN );\r\n\r\n% net2<\/span>\r\nNNet02 = train<\/span>( NNet02, data_IN, data_OUT, [], [], VV );\r\nres02  = sim<\/span>( NNet02, data_IN );\r\n\r\n% net3<\/span>\r\nNNet03 = train<\/span>( NNet03, data_IN, data_OUT, [], [], VV );\r\nres03  = sim<\/span>( NNet03, data_IN );\r\n\r\nplot(data_IN, res01,    'b','LineWidth',2',    ...\r\n     data_IN, res02,    'g','LineWidth',2',    ...\r\n     data_IN, res03,    'c','LineWidth',2,\r\n     data_IN, data_OUT, 'r','LineWidth',2' );\r\n<\/pre>\n
\u4e0d\u660e\u70b9<\/span>
\n\u4ed5\u69d8\u4e0a\u306f train<\/span> \u306e \u623b\u308a\u5024\u306f 4\u500b<\/span>\u3042\u308b\u306f\u305a\u30023\u756a\u76ee\u306eout \u306f\u304a\u305d\u3089\u304f\u51fa\u529b\u5024\u306e\u3053\u3068\u3002
\n\u3082\u3057\u3053\u308c\u304c\u53d6\u5f97\u3067\u304d\u308c\u3070\u3001\u4e0a\u8a18\u30d7\u30ed\u30b0\u30e9\u30e0\u306e sim <\/span>\u306f\u4e0d\u8981\u306b\u306a\u308b\u306f\u305a\u3060\u3002
\n\u3057\u304b\u3057\u3001\u73fe\u72b6\u306e train\u306f \u623b\u308a\u5024\u3092 1\u500b<\/span> \u3057\u304b\u8fd4\u3055\u306a\u3044… \u306a\u305c\u3060\uff1f<\/p>\n
\r\n[net,tr,out,E] = train<\/span>(MLPnet,mInputN,mOutput,[],[],VV);\r\n<\/pre>\n
\u3053\u306e\u8fba\u304c Unmaintained packages<\/span> \u3068\u3044\u3046\u3053\u3068\u306a\u306e\u3060\u308d\u3046\u304b…<\/p>\n
\n","protected":false},"excerpt":{"rendered":"
(33) \u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u3067sin\u95a2\u6570\u3092\u8fd1\u4f3c \u3067\u306f\u30013\u5c64(1-3-1)\u306e\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306bsin\u95a2\u6570\u3092\u5b66\u7fd2\u3055\u305b\u3066\u307f\u305f\u3002 \u4eca\u56de\u306f\u4e0b\u8868\u306e\uff13\u7a2e\u985e\u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u69cb\u6210\u3067\u6bd4\u8f03\u3057\u3066\u307f\u305f\u3002 1 3\u5c64 [I 1] – [\u2026 \u7d9a\u304d\u3092\u8aad\u3080 »<\/a><\/span><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[24],"tags":[],"class_list":["post-360","post","type-post","status-publish","format-standard","hentry","category-24"],"views":3020,"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/posts\/360","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/comments?post=360"}],"version-history":[{"count":25,"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/posts\/360\/revisions"}],"predecessor-version":[{"id":1007,"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/posts\/360\/revisions\/1007"}],"wp:attachment":[{"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/media?parent=360"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/categories?post=360"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.dogrow.net\/octave\/wp-json\/wp\/v2\/tags?post=360"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}