{"id":7584,"date":"2022-11-02T17:20:49","date_gmt":"2022-11-02T08:20:49","guid":{"rendered":"https:\/\/blogs.mathworks.com\/japan-community\/?p=7584"},"modified":"2022-11-02T17:20:49","modified_gmt":"2022-11-02T08:20:49","slug":"jp_whats-new-in-interoperability-with-tensorflow-and-pytorch","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/japan-community\/2022\/11\/02\/jp_whats-new-in-interoperability-with-tensorflow-and-pytorch\/","title":{"rendered":"TensorFlow \uff06 PyTorch \u3068 MATLAB"},"content":{"rendered":"<div class=\"rtcContent\">\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u203b\u3053\u306e\u6295\u7a3f\u306f 2022 \u5e74 10 \u6708 4 \u65e5\u306b Deep Learning (<a href=\"https:\/\/www.mathworks.com\/matlabcentral\/profile\/authors\/16845908?s_tid=blg_to_profile\">Sivylla Paraskevopoulou<\/a>) \u306b<a href=\"https:\/\/blogs.mathworks.com\/deep-learning\/2022\/10\/04\/whats-new-in-interoperability-with-tensorflow-and-pytorch\/?from=jp\">\u6295\u7a3f<\/a>\u3055\u308c\u305f\u3082\u306e\u306e\u6284\u8a33\u3067\u3059\u3002<\/div>\n<\/div>\n<div>\n<hr \/>\n<\/div>\n<div><\/div>\n<div class=\"rtcContent\">\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><a href=\"https:\/\/jp.mathworks.com\/company\/user_stories\/case-studies\/mitsui-chemicals-deploys-ai-and-automation-systems-with-tensor-flow-and-matlab.html\">\u4e09\u4e95\u5316\u5b66\u3001TensorFlow\u3068MATLAB\u3092\u4f7f\u7528\u3057\u3066AI\u30fb\u81ea\u52d5\u5316\u6280\u8853\u3092\u73fe\u5834\u5c55\u958b<\/a> \u306a\u3069 TensorFlow\u2122 \u3084 PyTorch<span style=\"font-weight: bold;\">\u00ae<\/span> \u3068 MATLAB \u3092\u9023\u643a\u3059\u308b\u30b1\u30fc\u30b9\u3082\u5897\u3048\u3066\u304d\u307e\u3057\u305f\u3002\u4eca\u56de\u306f\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u306e\u4ed6\u306e\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3068\u306e\u9023\u643a\u306b\u3064\u3044\u3066 R2022b \u3067\u306e\u65b0\u6a5f\u80fd\u3068\u95a2\u9023\u3059\u308b\u30b5\u30f3\u30d7\u30eb\u3092\u7d39\u4ecb\u3057\u307e\u3059\u3002<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">R2022b \u4ee5\u524d\u306e\u30e2\u30c7\u30eb\u5909\u63db\u306e\u30b5\u30dd\u30fc\u30c8\u306f\u3001ONNX\u2122 \u304b\u3089\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\u3068\u30a8\u30af\u30b9\u30dd\u30fc\u30c8\u3001TensorFlow \u304b\u3089\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\u306e\u307f\u3067\u3057\u305f\u304c\u3001R2022b \u304b\u3089\u306f <span style=\"font-weight: bold;\">TensorFlow \u306b Python\u00ae \u30b3\u30fc\u30c9\u3068\u3057\u3066\u30e2\u30c7\u30eb\u3092<\/span><span style=\"font-weight: bold;\">\u30a8\u30af\u30b9\u30dd\u30fc\u30c8<\/span>\u3067\u304d\u308b\u3088\u3046\u306b\u306a\u308a\u3001<span style=\"font-weight: bold;\">PyTorch \u304b\u3089\u30e2\u30c7\u30eb\u3092\u30a4\u30f3\u30dd\u30fc\u30c8<\/span>\u3067\u304d\u308b\u3088\u3046\u306b\u306a\u308a\u307e\u3057\u305f\u3002\uff08\u753b\u50cf\u5206\u985e\u304b\u3089\u30b5\u30dd\u30fc\u30c8\u958b\u59cb\uff09<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><\/div>\n<h2 style=\"margin: 20px 10px 5px 4px; padding: 0px; line-height: 20px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 20px; font-weight: bold; text-align: left;\"><span style=\"font-weight: bold;\">Export to TensorFlow<\/span><\/h2>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u307e\u305a\u306f TensorFlow \u3078\u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u51fa\u529b\u306b\u3064\u3044\u3066\u3002<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><img decoding=\"async\" loading=\"lazy\" class=\"imageNode\" style=\"vertical-align: baseline; width: 420px; height: 143px;\" src=\"https:\/\/blogs.mathworks.com\/japan-community\/files\/2022\/10\/InteroperabilitywithTensorFlowandPyTorch_1.png\" alt=\"\" width=\"420\" height=\"143\" \/><\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><a href=\"https:\/\/www.mathworks.com\/matlabcentral\/fileexchange\/64649-deep-learning-toolbox-converter-for-tensorflow-models\">Deep Learning Toolbox Converter for TensorFlow Models <\/a>\u306e Support Package \u306b TensorFlow \u3078\u51fa\u529b\u3059\u308b\u6a5f\u80fd\u304c\u8ffd\u52a0\u3055\u308c\u307e\u3057\u305f\u3002\u3053\u306e\u65b0\u3057\u3044\u95a2\u6570 <a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/exportnetworktotensorflow.html\"><span style=\"font-family: monospace;\">exportNetworkToTensorFlow<\/span><\/a> \u3092\u304a\u30b9\u30b9\u30e1\u3059\u308b\u7406\u7531\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3067\u3059\u3002<\/div>\n<ol style=\"margin: 10px 0px 20px; padding-left: 0px; font-family: Helvetica, Arial, sans-serif; font-size: 14px;\">\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3084\u30ec\u30a4\u30e4\u30b0\u30e9\u30d5\uff08\u7573\u307f\u8fbc\u307f\u3001LSTM\u306a\u3069\uff09\u3092\u76f4\u63a5 TensorFlow \u306b\u66f8\u304d\u51fa\u3059\u3053\u3068\u304c\u3067\u304d\u308b<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092\u51fa\u529b\u3057\u3066\u3001TensorFlow \u306e\u30e2\u30c7\u30eb\u3068\u3057\u3066\u8aad\u307f\u8fbc\u307f\u4e88\u6e2c\u306b\u4f7f\u7528\uff08\u30b5\u30f3\u30d7\u30eb\uff1a<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/exportnetworktotensorflow.html#mw_b88d248d-a405-4cfa-a1c3-f610dfc41675\">Export Network to TensorFlow and Classify Image<\/a>\uff09<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">\u5b66\u7fd2\u524d\u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092\u51fa\u529b\u3057\u3001\u305d\u308c\u3092 TensorFlow \u30e2\u30c7\u30eb\u3068\u3057\u3066\u8aad\u307f\u8fbc\u307f\u3057\u3066\u5b66\u7fd2\uff08\u30b5\u30f3\u30d7\u30eb\uff1a<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/exportnetworktotensorflow.html#mw_969bbaaf-2bad-47fd-bba4-e62ff2bb79d3\">Export Untrained Layer Graph to TensorFlow<\/a>\uff09<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">\u51fa\u529b\u3055\u308c\u305f\u30e2\u30c7\u30eb\u306f\u3001SavedModel \u3084 HDF5 \u5f62\u5f0f\u306a\u3069\u3001TensorFlow \u306e\u6a19\u6e96\u7684\u306a\u5f62\u5f0f\u3067\u4fdd\u5b58\u53ef\u80fd<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">\u591a\u304f\u306e MATLAB \u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u5c64\u306f TensorFlow \u306b\u5909\u63db\u53ef\u80fd\uff08\u5bfe\u5fdc\u3057\u3066\u3044\u308b\u5c64\u306e\u30ea\u30b9\u30c8\uff1a<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/exportnetworktotensorflow.html#mw_5df2ceb2-14b1-4133-9a27-3542d9a517de\">Layers Supported for Export to TensorFlow<\/a>\uff09<\/li>\n<\/ol>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u3053\u3053\u3067\u306f\u3001\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092 TensorFlow \u306b\u51fa\u529b\u3001TensorFlow \u30e2\u30c7\u30eb\u3068\u3057\u3066\u8aad\u307f\u8fbc\u307f\u3057\u3001SavedModel \u5f62\u5f0f\u3067\u4fdd\u5b58\u3059\u308b\u7c21\u5358\u306a\u30ef\u30fc\u30af\u30d5\u30ed\u30fc\u3092\u7d39\u4ecb\u3057\u307e\u3059\u3002<\/div>\n<h3 style=\"margin: 15px 10px 5px 4px; padding: 0px; line-height: 18px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 17px; font-weight: bold; text-align: left;\">MATLAB\u5074\u306e\u4f5c\u696d<\/h3>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u4e8b\u524d\u5b66\u7fd2\u6e08\u307f\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092\u8aad\u307f\u8fbc\u3080\u30b5\u30f3\u30d7\u30eb\uff08<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ug\/pretrained-convolutional-neural-networks.html\">Pretrained Deep Neural Networks<\/a>\uff09\u3067\u306f\u3001\u4e8b\u524d\u5b66\u7fd2\u3055\u308c\u305f\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092\u8aad\u307f\u8fbc\u3080\u65b9\u6cd5\u304c\u7d39\u4ecb\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/div>\n<div style=\"background-color: #f5f5f5; margin: 10px 0 10px 0;\">\n<div class=\"inlineWrapper\">\n<div style=\"border-radius: 4px; padding: 6px 45px 4px 13px; line-height: 18.004px; min-height: 0px; white-space: nowrap; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 14px; border: 1px solid #bfbfbf;\"><span style=\"white-space: pre;\">net = darknet19;<\/span><\/div>\n<\/div>\n<\/div>\n<div style=\"margin: 10px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u30cd\u30c3\u30c8\u30ef\u30fc\u30af <span style=\"font-family: monospace;\">net<\/span> \u3092 TensorFlow \u306b\u51fa\u529b\u3057\u307e\u3059\u3002<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/exportnetworktotensorflow.html\"><span style=\"font-family: monospace;\">exportNetworkToTensorFlow<\/span><\/a> \u95a2\u6570\u306f\u3001TensorFlow \u30e2\u30c7\u30eb\u3092 Python \u30d1\u30c3\u30b1\u30fc\u30b8 DarkNet19 \u3068\u3057\u3066\u4fdd\u5b58\u3059\u308b\u95a2\u6570\u3067\u3059\u3002<\/div>\n<div style=\"background-color: #f5f5f5; margin: 10px 0 10px 0;\">\n<div class=\"inlineWrapper\">\n<div style=\"border-radius: 4px; padding: 6px 45px 4px 13px; line-height: 18.004px; min-height: 0px; white-space: nowrap; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 14px; border: 1px solid #bfbfbf;\"><span style=\"white-space: pre;\">exportNetworkToTensorFlow(net, <span style=\"color: #a709f5;\">&#8220;DarkNet19&#8221;<\/span>)<\/span><\/div>\n<\/div>\n<\/div>\n<div style=\"margin: 10px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">DarkNet19 \u30d1\u30c3\u30b1\u30fc\u30b8\u306b\u306f4\u3064\u306e\u30d5\u30a1\u30a4\u30eb\u304c\u542b\u307e\u308c\u3066\u3044\u307e\u3059\u3002<\/div>\n<ul style=\"margin: 10px 0px 20px; padding-left: 0px; font-family: Helvetica, Arial, sans-serif; font-size: 14px;\">\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">_init_.py\uff1aDarkNet19 \u30d5\u30a9\u30eb\u30c0\u3092 Python \u30d1\u30c3\u30b1\u30fc\u30b8\u3068\u3057\u3066\u5b9a\u7fa9\u3059\u308b\u3082\u306e<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">model.py\u30d5\u30a1\u30a4\u30eb\uff1a\u5b66\u7fd2\u3057\u3066\u3044\u306a\u3044 TensorFlow-Keras \u30e2\u30c7\u30eb\u3092\u5b9a\u7fa9\u3059\u308b\u30b3\u30fc\u30c9\u304c\u542b\u307e\u308c\u3066\u3044\u307e\u3059<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">TensorFlow \u30e2\u30c7\u30eb\u3092\u8aad\u307f\u8fbc\u307f\u3001HDF5 \u307e\u305f\u306f SavedModel \u5f62\u5f0f\u3067\u4fdd\u5b58\u3059\u308b\u65b9\u6cd5\u3092\u8aac\u660e\u3059\u308bREADME.txt\u30d5\u30a1\u30a4\u30eb<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\">\u30e2\u30c7\u30eb\u306e\u91cd\u307f\u3092 HDF5 \u5f62\u5f0f\u3067\u683c\u7d0d\u3057\u305fweights.h5 \u30d5\u30a1\u30a4\u30eb<\/li>\n<\/ul>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><img decoding=\"async\" loading=\"lazy\" class=\"imageNode\" style=\"vertical-align: baseline; width: 727px; height: 452px;\" src=\"https:\/\/blogs.mathworks.com\/japan-community\/files\/2022\/10\/InteroperabilitywithTensorFlowandPyTorch_2.png\" alt=\"package_diagram3.png\" width=\"727\" height=\"452\" \/><\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u56f3: \u51fa\u529b\u3055\u308c\u305f TensorFlow \u30e2\u30c7\u30eb\u306f\u3001Python \u30d1\u30c3\u30b1\u30fc\u30b8\u3067\u3042\u308b DarkNet19 \u306b\u4fdd\u5b58\u3055\u308c\u308b<\/div>\n<h3 style=\"margin: 3px 10px 5px 4px; padding: 0px; line-height: 18px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 17px; font-weight: bold; text-align: left;\">Python \u5074\u306e\u4f5c\u696d<\/h3>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">DarkNet19 \u30d1\u30c3\u30b1\u30fc\u30b8\u304b\u3089\u3001\u51fa\u529b\u3055\u308c\u305f TensorFlow \u30e2\u30c7\u30eb\u3092\u8aad\u307f\u8fbc\u307f\u307e\u3059\u3002<\/div>\n<div class=\"preformatted-plain\" style=\"margin: 10px 3px 10px 55px; padding: 10px 10px 10px 5px;\">\n<div style=\"border-radius: 0px; padding: 0px; line-height: 15.6px; min-height: 16px; white-space: pre; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 12px; border: 0px none #212121;\"><span style=\"white-space: pre;\">model = DarkNet19.load_model()<\/span><\/div>\n<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u30a8\u30af\u30b9\u30dd\u30fc\u30c8\u3055\u308c\u305f\u30e2\u30c7\u30eb\u3092 SavedModel \u5f62\u5f0f\u3067\u4fdd\u5b58\u3057\u307e\u3059\u3002<\/div>\n<div class=\"preformatted-plain\" style=\"margin: 10px 3px 10px 55px; padding: 10px 10px 10px 5px;\">\n<div style=\"border-radius: 0px; padding: 0px; line-height: 15.6px; min-height: 16px; white-space: pre; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 12px; border: 0px none #212121;\"><span style=\"white-space: pre;\">model.save(&#8220;DarkNet19_savedmodel&#8221;)<\/span><\/div>\n<\/div>\n<h2 style=\"margin: 3px 10px 5px 4px; padding: 0px; line-height: 20px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 20px; font-weight: bold; text-align: left;\"><span style=\"font-weight: bold;\">Import from PyTorch<\/span><\/h2>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u6b21\u306b PyTorch \u304b\u3089\u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u8aad\u307f\u8fbc\u307f\u306b\u3064\u3044\u3066\u3002<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><img decoding=\"async\" loading=\"lazy\" class=\"imageNode\" style=\"vertical-align: baseline; width: 423px; height: 228px;\" src=\"https:\/\/blogs.mathworks.com\/japan-community\/files\/2022\/10\/InteroperabilitywithTensorFlowandPyTorch_3.png\" alt=\"\" width=\"423\" height=\"228\" \/><\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">R2022b \u3067 <a href=\"https:\/\/www.mathworks.com\/matlabcentral\/fileexchange\/111925-deep-learning-toolbox-converter-for-pytorch-models\">Deep Learning Toolbox Converter for PyTorch Models<\/a> \u306e Support Package \u304c\u5c0e\u5165\u3055\u308c\u307e\u3057\u305f\u3002\u3053\u306e\u6700\u521d\u306e\u30ea\u30ea\u30fc\u30b9\u3067\u306f\u3001\u753b\u50cf\u5206\u985e\u30e2\u30c7\u30eb\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\u3092\u30b5\u30dd\u30fc\u30c8\u3057\u3066\u3044\u307e\u3059\u3002\u4ed6\u306e\u30e2\u30c7\u30eb\u30bf\u30a4\u30d7\u306e\u30b5\u30dd\u30fc\u30c8\u306f\u3001\u5c06\u6765\u306e\u30ea\u30ea\u30fc\u30b9\u3067\u8ffd\u52a0\u3055\u308c\u308b\u4e88\u5b9a\u3067\u3059\u3002PyTorch \u30e2\u30c7\u30eb\u3092\u30a4\u30f3\u30dd\u30fc\u30c8\u3059\u308b\u306b\u306f\u3001<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/importnetworkfrompytorch.html\"><span style=\"font-family: monospace;\">importNetworkFromPyTorch<\/span><\/a> \u95a2\u6570\u3092\u4f7f\u7528\u3057\u307e\u3059\u3002\u30a4\u30f3\u30dd\u30fc\u30c8\u3059\u308b PyTorch \u30e2\u30c7\u30eb\u306f\u3001\u4e8b\u524d\u306b\u5b66\u7fd2\u30fb\u30c8\u30ec\u30fc\u30b9\u3055\u308c\u3066\u3044\u308b\u3053\u3068\u3092\u78ba\u8a8d\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u3053\u3053\u3067\u306f\u3001PyTorch \u304b\u3089\u753b\u50cf\u5206\u985e\u30e2\u30c7\u30eb\u3092\u30a4\u30f3\u30dd\u30fc\u30c8\u3057\u521d\u671f\u5316\u3059\u308b\u65b9\u6cd5\u3092\u7d39\u4ecb\u3057\u307e\u3059\u3002<\/div>\n<h3 style=\"margin: 3px 10px 5px 4px; padding: 0px; line-height: 18px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 17px; font-weight: bold; text-align: left;\">Python \u5074\u306e\u4f5c\u696d<\/h3>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><a href=\"https:\/\/pytorch.org\/vision\/0.8\/models.html\" class=\"external\" rel=\"nofollow\" target=\"_blank\">TorchVision library<\/a> \u304b\u3089\u4e8b\u524d\u5b66\u7fd2\u3055\u308c\u305f\u753b\u50cf\u5206\u985e\u30e2\u30c7\u30eb\u3092\u30ed\u30fc\u30c9\u3057\u307e\u3059\u3002<\/div>\n<div class=\"preformatted-plain\" style=\"margin: 10px 3px 10px 55px; padding: 10px 10px 10px 5px;\">\n<div style=\"border-radius: 0px; padding: 0px; line-height: 15.6px; min-height: 16px; white-space: pre; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 12px; border: 0px none #212121;\"><span style=\"white-space: pre;\">Import torch<\/span><\/div>\n<div style=\"border-radius: 0px; padding: 0px; line-height: 15.6px; min-height: 16px; white-space: pre; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 12px; border: 0px none #212121;\"><span style=\"white-space: pre;\">from torchvision import models<\/span><\/div>\n<div style=\"border-radius: 0px; padding: 0px; line-height: 15.6px; min-height: 16px; white-space: pre; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 12px; border: 0px none #212121;\"><span style=\"white-space: pre;\">model = models.mnasnet1_0(pretrained=True)<\/span><\/div>\n<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">PyTorch \u306e\u30e2\u30c7\u30eb\u3092\u30c8\u30ec\u30fc\u30b9\u3057\u307e\u3059\u3002\u30c8\u30ec\u30fc\u30b9\u3059\u308b\u65b9\u6cd5\u306b\u3064\u3044\u3066\u306f <a href=\"https:\/\/pytorch.org\/docs\/stable\/generated\/torch.jit.trace.html\" class=\"external\" rel=\"nofollow\" target=\"_blank\">Torch documentation: Tracing a function<\/a> \u3092\u53c2\u7167\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/div>\n<div class=\"preformatted-plain\" style=\"margin: 10px 3px 10px 55px; padding: 10px 10px 10px 5px;\">\n<div style=\"border-radius: 0px; padding: 0px; line-height: 15.6px; min-height: 16px; white-space: pre; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 12px; border: 0px none #212121;\"><span style=\"white-space: pre;\">X = torch.rand(1,3,224,224)<\/span><\/div>\n<div style=\"border-radius: 0px; padding: 0px; line-height: 15.6px; min-height: 16px; white-space: pre; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 12px; border: 0px none #212121;\"><span style=\"white-space: pre;\">traced_model = torch.jit.trace(model.forward,X)<\/span><\/div>\n<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">PyTorch\u306e\u30e2\u30c7\u30eb\u3092\u4fdd\u5b58\u3057\u307e\u3059\u3002<\/div>\n<div class=\"preformatted-plain\" style=\"margin: 10px 3px 10px 55px; padding: 10px 10px 10px 5px;\">\n<div style=\"border-radius: 0px; padding: 0px; line-height: 15.6px; min-height: 16px; white-space: pre; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 12px; border: 0px none #212121;\"><span style=\"white-space: pre;\">traced_model.save(&#8220;traced_mnasnet1_0.pt&#8221;)<\/span><\/div>\n<\/div>\n<h3 style=\"margin: 3px 10px 5px 4px; padding: 0px; line-height: 18px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 17px; font-weight: bold; text-align: left;\">MATLAB \u5074\u306e\u4f5c\u696d<\/h3>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/importnetworkfrompytorch.html\"><span style=\"font-family: monospace;\">importNetworkFromPyTorch<\/span><\/a> \u95a2\u6570\u3092\u4f7f\u7528\u3057\u3066\u3001PyTorch \u30e2\u30c7\u30eb\u3092 MATLAB \u306b\u30a4\u30f3\u30dd\u30fc\u30c8\u3057\u307e\u3059\u3002\u3053\u306e\u95a2\u6570\u306f\u3001\u5165\u529b\u5c64\u3092\u6301\u305f\u306a\u3044\u521d\u671f\u5316\u3055\u308c\u3066\u3044\u306a\u3044 <span style=\"font-family: monospace;\">dlnetwork<\/span> \u30aa\u30d6\u30b8\u30a7\u30af\u30c8\u3068\u3057\u3066\u30e2\u30c7\u30eb\u3092\u30a4\u30f3\u30dd\u30fc\u30c8\u3057\u307e\u3059\u3002<\/div>\n<div style=\"background-color: #f5f5f5; margin: 10px 0 10px 0;\">\n<div class=\"inlineWrapper\">\n<div style=\"border-radius: 4px; padding: 6px 45px 4px 13px; line-height: 18.004px; min-height: 0px; white-space: nowrap; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 14px; border: 1px solid #bfbfbf;\"><span style=\"white-space: pre;\">net = importNetworkFromPyTorch(<span style=\"color: #a709f5;\">&#8220;traced_mnasnet1_0.pt&#8221;<\/span>);<\/span><\/div>\n<\/div>\n<\/div>\n<div style=\"margin: 10px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u30a4\u30f3\u30dd\u30fc\u30c8\u3057\u305f\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u5165\u529b\u30b5\u30a4\u30ba\u3092\u6307\u5b9a\u3057\u3001\u753b\u50cf\u5165\u529b\u30ec\u30a4\u30e4\u30fc\u3092\u4f5c\u6210\u3057\u307e\u3059\u3002\u6b21\u306b\u3001\u30a4\u30f3\u30dd\u30fc\u30c8\u3057\u305f\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306b\u753b\u50cf\u5165\u529b\u5c64\u3092\u8ffd\u52a0\u3057\u3001<a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/dlnetwork.addinputlayer.html\"><span style=\"font-family: monospace;\">addInputLayer<\/span><\/a> \u95a2\u6570\uff08\u3053\u308c\u3082 R2022b \u306e\u65b0\u6a5f\u80fd\uff09\u3092\u4f7f\u3063\u3066\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3092\u521d\u671f\u5316\u3057\u307e\u3059\u3002<\/div>\n<div style=\"background-color: #f5f5f5; margin: 10px 0 10px 0;\">\n<div class=\"inlineWrapper\">\n<div style=\"border-left: 1px solid #bfbfbf; border-right: 1px solid #bfbfbf; border-top: 1px solid #bfbfbf; border-bottom: 0px none #212121; border-radius: 4px 4px 0px 0px; padding: 6px 45px 0px 13px; line-height: 18.004px; min-height: 0px; white-space: nowrap; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 14px;\"><span style=\"white-space: pre;\">InputSize = [224 224 3];<\/span><\/div>\n<\/div>\n<div class=\"inlineWrapper\">\n<div style=\"border-left: 1px solid #bfbfbf; border-right: 1px solid #bfbfbf; border-top: 0px none #212121; border-bottom: 0px none #212121; border-radius: 0px; padding: 0px 45px 0px 13px; line-height: 18.004px; min-height: 0px; white-space: nowrap; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 14px;\"><span style=\"white-space: pre;\">InputLayer = imageInputLayer(InputSize,Normalization=<span style=\"color: #a709f5;\">&#8220;none&#8221;<\/span>);<\/span><\/div>\n<\/div>\n<div class=\"inlineWrapper\">\n<div style=\"border-left: 1px solid #bfbfbf; border-right: 1px solid #bfbfbf; border-top: 0px none #212121; border-bottom: 1px solid #bfbfbf; border-radius: 0px 0px 4px 4px; padding: 0px 45px 4px 13px; line-height: 18.004px; min-height: 0px; white-space: nowrap; color: #212121; font-family: Menlo, Monaco, Consolas, 'Courier New', monospace; font-size: 14px;\"><span style=\"white-space: pre;\">net = addInputLayer(net,InputLayer,Initialize=true);<\/span><\/div>\n<\/div>\n<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u4e0a\u8a18\u306e\u30b3\u30fc\u30c9\u3067\u3001PyTorch \u3068 MATLAB \u306e\u5165\u529b\u6b21\u5143\u306e\u9806\u5e8f\u304c\u7570\u306a\u308b\u3053\u3068\u306b\u304a\u6c17\u3065\u304d\u304b\u3082\u3057\u308c\u307e\u305b\u3093\u3002\u5165\u529b\u306e\u6b21\u5143\u306b\u3064\u3044\u3066\u306e\u8a73\u7d30\u306f <a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ug\/tips-on-importing-models-from-tensorflow-pytorch-and-onnx.html#mw_ca5d4cba-9c12-4f01-8fe1-6329730c92b2\">Input Dimension Ordering for Deep Learning Platforms<\/a> \u3092\u3054\u89a7\u304f\u3060\u3055\u3044\u3002PyTorch \u30e2\u30c7\u30eb\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\u65b9\u6cd5\u3001\u30a4\u30f3\u30dd\u30fc\u30c8\u3057\u305f\u30e2\u30c7\u30eb\u306e\u521d\u671f\u5316\u65b9\u6cd5\u3001\u30a4\u30f3\u30dd\u30fc\u30c8\u3057\u305f\u30e2\u30c7\u30eb\u3067\u4e88\u6e2c\u3084\u5b66\u7fd2\u306a\u3069\u3092\u5b9f\u884c\u3059\u308b\u65b9\u6cd5\u306b\u3064\u3044\u3066\u306f\u3001importNetworkTFromPyTorch \u306e\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\u30da\u30fc\u30b8\u306e <a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ref\/importnetworkfrompytorch.html#mw_d51e64bc-e9ad-409d-ba4a-d70c47f40449\">Examples<\/a> \u306b\u8a73\u7d30\u304c\u3042\u308a\u307e\u3059\u3002<\/div>\n<h2 style=\"margin: 3px 10px 5px 4px; padding: 0px; line-height: 20px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 20px; font-weight: bold; text-align: left;\">\u76f8\u4e92\u904b\u7528\u306b\u95a2\u3059\u308b\u6a5f\u80fd\u307e\u3068\u3081<\/h2>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\"><img decoding=\"async\" loading=\"lazy\" class=\"imageNode\" style=\"vertical-align: baseline; width: 463px; height: 286px;\" src=\"https:\/\/blogs.mathworks.com\/japan-community\/files\/2022\/10\/InteroperabilitywithTensorFlowandPyTorch_4.png\" alt=\"\" width=\"463\" height=\"286\" \/><\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u4eca\u56de\u7d39\u4ecb\u3057\u305f Support Package \u306b\u3088\u308a\u3001MATLAB \u306e Deep Learning Toolbox \u3068 TensorFlow\u3001Pytorch\u3001ONNX \u9593\u3067\u306e\u76f8\u4e92\u904b\u7528\u304c\u30b9\u30e0\u30fc\u30ba\u306b\u306a\u308a\u3001\u30aa\u30fc\u30d7\u30f3\u30bd\u30fc\u30b9\u3067\u5229\u7528\u53ef\u80fd\u306a\u30e2\u30c7\u30eb\u3092\u6d3b\u7528\u3059\u308b\u3053\u3068\u306f\u3082\u3061\u308d\u3093\u3001\u69d8\u3005\u306a\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u306e\u826f\u3044\u3068\u3053\u308d\u53d6\u308a\u3092\u3057\u305f\u958b\u767a\u304c\u53ef\u80fd\u306b\u306a\u308a\u307e\u3059\u3002<\/div>\n<div style=\"margin: 2px 10px 9px 4px; padding: 0px; line-height: 21px; min-height: 0px; white-space: pre-wrap; color: #212121; font-family: Helvetica, Arial, sans-serif; font-style: normal; font-size: 14px; font-weight: 400; text-align: left;\">\u8a73\u7d30\u306b\u3064\u3044\u3066\u306f\u4ee5\u4e0b\u306e\u30da\u30fc\u30b8\u3082\u78ba\u8a8d\u3057\u3066\u307f\u3066\u304f\u3060\u3055\u3044\u3002<\/div>\n<ul style=\"margin: 10px 0px 20px; padding-left: 0px; font-family: Helvetica, Arial, sans-serif; font-size: 14px;\">\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\"><a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/deep-learning-import-and-export.html\">Deep Learning Import and Export<\/a>\uff1a\u5229\u7528\u53ef\u80fd\u306a\u3059\u3079\u3066\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\u304a\u3088\u3073\u30a8\u30af\u30b9\u30dd\u30fc\u30c8\u95a2\u6570\uff08\u304a\u3088\u3073\u305d\u306e\u30c9\u30ad\u30e5\u30e1\u30f3\u30c8\uff09<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\"><a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ug\/interoperability-between-deep-learning-toolbox-tensorflow-pytorch-and-onnx.html\">Interoperability Between Deep Learning Toolbox, TensorFlow, PyTorch, and ONNX<\/a>\uff1a\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\u3068\u30a8\u30af\u30b9\u30dd\u30fc\u30c8\u306e\u65b9\u6cd5<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\"><a href=\"https:\/\/www.mathworks.com\/help\/deeplearning\/ug\/tips-on-importing-models-from-tensorflow-pytorch-and-onnx.html\">Tips on Importing Models from TensorFlow, PyTorch, and ONNX<\/a>\uff1a\u30e2\u30c7\u30eb\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\u306b\u95a2\u3059\u308b\u3088\u304f\u3042\u308b\u8cea\u554f\u3068\u56de\u7b54<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\"><a href=\"https:\/\/github.com\/matlab-deep-learning\/MATLAB-Deep-Learning-Model-Hub\" class=\"external\" rel=\"nofollow\" target=\"_blank\">MATLAB Deep Learning Model Hub<\/a>\uff1aMATLAB \u306e\u30e2\u30c7\u30eb\u30ea\u30dd\u30b8\u30c8\u30ea\uff08\u9069\u5207\u306a\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u304c\u898b\u3064\u304b\u308b\u304b\u3082\u3057\u308c\u307e\u305b\u3093\uff09<\/li>\n<li style=\"margin-left: 56px; line-height: 21px; min-height: 0px; text-align: left; white-space: pre-wrap;\"><a href=\"https:\/\/blogs.mathworks.com\/deep-learning\/2022\/03\/18\/importing-models-from-tensorflow-pytorch-and-onnx\/?from=jp\">Importing Models from TensorFlow, PyTorch, and ONNX<\/a>\uff1a\u30a4\u30f3\u30dd\u30fc\u30c8\u306b\u95a2\u3059\u308b\u30d2\u30f3\u30c8\u3084\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3067\u304d\u308b\u30b5\u30f3\u30d7\u30eb\uff08TensorFlow \u304b\u3089\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\u306b\u30d5\u30a9\u30fc\u30ab\u30b9\uff09<\/li>\n<\/ul>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<div class=\"overview-image\"><img decoding=\"async\"  class=\"img-responsive\" src=\"https:\/\/blogs.mathworks.com\/japan-community\/files\/2022\/10\/InteroperabilitywithTensorFlowandPyTorch_1.png\" onError=\"this.style.display ='none';\" \/><\/div>\n<p>\n\u203b\u3053\u306e\u6295\u7a3f\u306f 2022 \u5e74 10 \u6708 4 \u65e5\u306b Deep Learning (Sivylla Paraskevopoulou) \u306b\u6295\u7a3f\u3055\u308c\u305f\u3082\u306e\u306e\u6284\u8a33\u3067\u3059\u3002<\/p>\n<p>\u4e09\u4e95\u5316\u5b66\u3001TensorFlow\u3068MATLAB\u3092\u4f7f\u7528\u3057\u3066AI\u30fb\u81ea\u52d5\u5316\u6280\u8853\u3092\u73fe\u5834\u5c55\u958b \u306a\u3069 TensorFlow\u2122 \u3084 PyTorch\u00ae \u3068 MATLAB&#8230; <a class=\"read-more\" href=\"https:\/\/blogs.mathworks.com\/japan-community\/2022\/11\/02\/jp_whats-new-in-interoperability-with-tensorflow-and-pytorch\/\">read more >><\/a><\/p>\n","protected":false},"author":159,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[61],"tags":[],"_links":{"self":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/7584"}],"collection":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/users\/159"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/comments?post=7584"}],"version-history":[{"count":3,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/7584\/revisions"}],"predecessor-version":[{"id":7593,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/posts\/7584\/revisions\/7593"}],"wp:attachment":[{"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/media?parent=7584"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/categories?post=7584"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.mathworks.com\/japan-community\/wp-json\/wp\/v2\/tags?post=7584"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}