{"id":5642,"date":"2024-10-07T18:29:15","date_gmt":"2024-10-07T09:29:15","guid":{"rendered":"https:\/\/blog.since2020.jp\/?p=5642"},"modified":"2024-10-08T12:04:29","modified_gmt":"2024-10-08T03:04:29","slug":"permutation-importance","status":"publish","type":"post","link":"https:\/\/since2020.jp\/media\/permutation-importance\/","title":{"rendered":"Permutation Importance: \u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u3092\u300c\u7279\u5fb4\u91cf\u306e\u91cd\u8981\u5ea6\u300d\u304b\u3089\u89e3\u91c8\u3059\u308b"},"content":{"rendered":"\n

\u30d3\u30b8\u30cd\u30b9\u3067\u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u3092\u6d3b\u7528\u3059\u308b\u5834\u5408\u306b\u306f\u3001\u30d6\u30e9\u30c3\u30af\u30dc\u30c3\u30af\u30b9\u306a\u30e2\u30c7\u30eb\u306e\u6319\u52d5\u3092\u89e3\u91c8\u3059\u308b\u3053\u3068\u304c\u5fc5\u8981\u3067\u3042\u308b\u5834\u5408\u304c\u3042\u308a\u307e\u3059\u3002\u672c\u8a18\u4e8b\u3067\u306f\u3001\u7279\u5fb4\u91cf\u306e\u91cd\u8981\u5ea6\u3068\u3044\u3046\u89b3\u70b9\u304b\u3089\u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u3092\u89e3\u91c8\u3059\u308b\u300cPermutation Importance\u300d\u306b\u3064\u3044\u3066\u89e3\u8aac\u3057\u307e\u3059\u3002<\/p>\n\n\n

\u306f\u3058\u3081\u306b<\/h2>\n

\u8fd1\u5e74\u306e\u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u306f\u3001LightGBM\u3084Deep Neural Network\u306a\u3069\u306e\u9ad8\u5ea6\u306a\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306b\u3088\u3063\u3066\u3001\u9ad8\u3044\u4e88\u6e2c\u6027\u80fd\u304c\u9054\u6210\u3055\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u308c\u3089\u306e\u8907\u96d1\u306a\u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u306f\u300c\u30d6\u30e9\u30c3\u30af\u30dc\u30c3\u30af\u30b9\u300d\u3068\u3082\u547c\u3070\u308c\u3001\u30e2\u30c7\u30eb\u304c\u3069\u306e\u3088\u3046\u306b\u4e88\u6e2c\u5024\u3092\u51fa\u529b\u3057\u3066\u3044\u308b\u306e\u304b\u3092\u4eba\u9593\u304c\u7406\u89e3\u3059\u308b\u306e\u306f\u975e\u5e38\u306b\u56f0\u96e3\u3067\u3059\u3002<\/p>\r\n

\u30d3\u30b8\u30cd\u30b9\u3067\u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u3092\u6d3b\u7528\u3059\u308b\u5834\u5408\u306b\u306f\u3001\u30e2\u30c7\u30eb\u304c\u30d6\u30e9\u30c3\u30af\u30dc\u30c3\u30af\u30b9\u3067\u3042\u308b\u3053\u3068\u306f\u5927\u304d\u306a\u554f\u984c\u3068\u306a\u308b\u3053\u3068\u304c\u3042\u308a\u307e\u3059\u3002\u30e2\u30c7\u30eb\u306e\u6319\u52d5\u3092\u7406\u89e3\u3059\u308b\u3053\u3068\u306f\u3001\u4e88\u6e2c\u5024\u306b\u5bfe\u3059\u308b\u4fe1\u983c\u6027\u3092\u9ad8\u3081\u3001\u30e6\u30fc\u30b6\u30fc\u306b\u7d0d\u5f97\u3057\u3066\u30e2\u30c7\u30eb\u3092\u4f7f\u7528\u3057\u3066\u3082\u3089\u3046\u305f\u3081\u306b\u975e\u5e38\u306b\u91cd\u8981\u3067\u3059\u3002\u307e\u305f\u3001\u533b\u7642\u3084\u81ea\u52d5\u904b\u8ee2\u306a\u3069\u5927\u304d\u306a\u30ea\u30b9\u30af\u3092\u4f34\u3046\u610f\u601d\u6c7a\u5b9a\u306bAI\u3092\u6d3b\u7528\u3059\u308b\u5834\u5408\u3001\u30e2\u30c7\u30eb\u306e\u6319\u52d5\u3092\u3042\u308b\u7a0b\u5ea6\u8aac\u660e\u3067\u304d\u308b\u5fc5\u8981\u304c\u3042\u308b\u3067\u3057\u3087\u3046\u3002<\/p>\r\n

\u305d\u3053\u3067\u672c\u8a18\u4e8b\u3067\u306f\u3001\u4e88\u6e2c\u30e2\u30c7\u30eb\u3092\u89e3\u91c8\u3059\u308b\u624b\u6cd5\u306e\u4e00\u3064\u3067\u3042\u308b\u3001Permutation Importance<\/strong>\u306b\u3064\u3044\u3066\u89e3\u8aac\u3057\u307e\u3059\u3002<\/p>\n\n

Permutation Importance \u306e\u8a08\u7b97\u65b9\u6cd5<\/h2>\n

Permutation Importance (PI)<\/strong> \u306f\u3001\u65e2\u5b58\u306e\u5b66\u7fd2\u6e08\u307f\u30e2\u30c7\u30eb\u3092\u518d\u5b66\u7fd2\u3059\u308b\u3053\u3068\u306a\u304f\u3001\u30e2\u30c7\u30eb\u304c\u4e88\u6e2c\u306b\u3069\u306e\u7279\u5fb4\u91cf\u3092\u91cd\u8981\u8996\u3057\u3066\u3044\u308b\u304b\u3092\u8a55\u4fa1\u3059\u308b\u65b9\u6cd5\u3067\u3059\u3002\u91cd\u8981\u5ea6\u3092\u77e5\u308a\u305f\u3044\u7279\u5fb4\u91cf\u306e\u5024\u3092\u30b7\u30e3\u30c3\u30d5\u30eb\u3057\u3001\u305d\u306e\u7279\u5fb4\u91cf\u304c\u6301\u3064\u91cd\u8981\u5ea6\u3092\u6570\u5024\u5316\u3057\u307e\u3059\u3002<\/p>\r\n

\u5177\u4f53\u7684\u306b\u306f\u3001\u4ee5\u4e0b\u306e\u624b\u9806\u306b\u3088\u3063\u3066\u7279\u5fb4\u91cf\u306e\u91cd\u8981\u5ea6\u3092\u8a08\u7b97\u3057\u307e\u3059\u3002<\/p>\r\n

1<\/strong> \u30e2\u30c7\u30eb\u306e\u30d9\u30fc\u30b9\u30e9\u30a4\u30f3\u6027\u80fd\uff08\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u5168\u4f53\u306b\u5bfe\u3059\u308b\u4e88\u6e2c\u7cbe\u5ea6\uff09\u3092\u6e2c\u5b9a\u3059\u308b\u3002<\/p>\r\n

2<\/strong> \u4e00\u3064\u306e\u7279\u5fb4\u91cf\u3092\u30b7\u30e3\u30c3\u30d5\u30eb\u3057\u305f\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u4f5c\u6210\u3059\u308b\u3002<\/p>\r\n

3<\/strong>\u00a0\u305d\u306e\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3067\u518d\u5ea6\u4e88\u6e2c\u3092\u884c\u3044\u3001\u65b0\u305f\u306a\u4e88\u6e2c\u7cbe\u5ea6\u3092\u6e2c\u5b9a\u3059\u308b\u3002<\/p>\r\n

4<\/strong>\u00a0\u4e88\u6e2c\u7cbe\u5ea6\u304c\u3069\u308c\u3060\u3051\u4f4e\u4e0b\u3057\u305f\u304b\u3092\u305d\u306e\u7279\u5fb4\u91cf\u306e\u91cd\u8981\u5ea6\u3068\u3059\u308b\u3002<\/p>\r\n

5<\/strong> \u91cd\u8981\u5ea6\u3092\u77e5\u308a\u305f\u3044\u5168\u3066\u306e\u7279\u5fb4\u91cf\u306b\u3064\u3044\u3066\u30012 ~ 4 \u3092\u7e70\u308a\u8fd4\u3059\u3002<\/p>\r\n

PI\u3067\u306f\u3001\u7279\u5fb4\u91cf\u3092\u30b7\u30e3\u30c3\u30d5\u30eb\u3059\u308b\u3053\u3068\u3067\u305d\u306e\u7279\u5fb4\u91cf\u306e\u60c5\u5831\u304c\u306a\u304f\u306a\u308b\u3053\u3068\u306e\u5f71\u97ff\u3092\u6e2c\u5b9a\u3057<\/strong>\u3001\u4e88\u6e2c\u30e2\u30c7\u30eb\u306e\u51fa\u529b\u304c\u305d\u306e\u7279\u5fb4\u91cf\u306b\u3069\u308c\u3060\u3051\u4f9d\u5b58\u3057\u3066\u3044\u305f\u304b\u3092\u8a55\u4fa1\u3057\u307e\u3059\u3002<\/p>\r\n

\"\"\uff08\u68ee\u4e0b (2021)\u300e\u6a5f\u68b0\u5b66\u7fd2\u3092\u89e3\u91c8\u3059\u308b\u6280\u8853\u300f\u3092\u5143\u306b\u4f5c\u6210\uff09<\/p>\r\n

<\/p>\n\n

Python\u3067\u306e\u5b9f\u884c\u4f8b<\/h2>\n

\u5b9f\u884c\u74b0\u5883<\/strong><\/span><\/p>\r\n

Python 3.11.9<\/p>\r\n


\r\n\u30c7\u30fc\u30bf\u306e\u6e96\u5099<\/strong><\/span><\/p>\r\n

\u4eca\u56de\u306f\u3001scikit-learn\u306e fetch_california_housing<\/strong><\/a> \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u4f7f\u7528\u3057\u307e\u3059\u3002\u3053\u308c\u306f\u30011990\u5e74\u306e\u30ab\u30ea\u30d5\u30a9\u30eb\u30cb\u30a2\u306b\u304a\u3051\u308b\u4f4f\u5b85\u4fa1\u683c\u306e\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3067\u3059\u3002\u30bf\u30fc\u30b2\u30c3\u30c8\u5909\u6570\u3067\u3042\u308b\u4f4f\u5b85\u4fa1\u683c\u3092\u4e88\u6e2c\u3059\u308b\u56de\u5e30\u554f\u984c\u3092\u89e3\u304f\u305f\u3081\u306b\u4f7f\u308f\u308c\u307e\u3059\u3002<\/a><\/p>\r\n

\r\n
# \u5fc5\u8981\u306a\u30e9\u30a4\u30d6\u30e9\u30ea\u3092\u30a4\u30f3\u30dd\u30fc\u30c8\r\nimport numpy as np\r\nimport pandas as pd\r\nimport matplotlib.pyplot as plt\r\nimport japanize_matplotlib\r\nfrom sklearn.datasets import fetch_california_housing\r\nfrom sklearn.ensemble import GradientBoostingRegressor\r\nfrom sklearn.model_selection import train_test_split\r\nfrom sklearn.inspection import permutation_importance\r\n\r\n# \u30ab\u30ea\u30d5\u30a9\u30eb\u30cb\u30a2\u4f4f\u5b85\u4fa1\u683c\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u8aad\u307f\u8fbc\u3080\r\ncalifornia = fetch_california_housing()\r\n\r\n# \u30c7\u30fc\u30bf\u3092DataFrame\u306b\u5909\u63db\r\ndf = pd.DataFrame(california.data, columns=california.feature_names)\r\n\r\n# \u30ab\u30e9\u30e0\u540d\u3092\u65e5\u672c\u8a9e\u306b\u5909\u66f4\r\njapanese_columns = {\r\n\u00a0 \u00a0 'MedInc': '\u4e16\u5e2f\u306e\u4e2d\u592e\u5024\u53ce\u5165',\r\n\u00a0 \u00a0 'HouseAge': '\u4f4f\u5b85\u306e\u7bc9\u5e74\u6570',\r\n\u00a0 \u00a0 'AveRooms': '\u5e73\u5747\u90e8\u5c4b\u6570',\r\n\u00a0 \u00a0 'AveBedrms': '\u5e73\u5747\u5bdd\u5ba4\u6570',\r\n\u00a0 \u00a0 'Population': '\u4eba\u53e3',\r\n\u00a0 \u00a0 'AveOccup': '\u4e16\u5e2f\u306e\u5e73\u5747\u4eba\u6570',\r\n\u00a0 \u00a0 'Latitude': '\u7def\u5ea6',\r\n\u00a0 \u00a0 'Longitude': '\u7d4c\u5ea6',\r\n}\r\ndf.rename(columns=japanese_columns, inplace=True)\r\ndf['\u4f4f\u5b85\u4fa1\u683c'] = california.target\r\n\r\n# \u7279\u5fb4\u91cf\u3068\u30bf\u30fc\u30b2\u30c3\u30c8\u306b\u5206\u3051\u308b\r\nX = df.drop('\u4f4f\u5b85\u4fa1\u683c', axis=1)\r\ny = df['\u4f4f\u5b85\u4fa1\u683c']\r\n\r\n# \u30c7\u30fc\u30bf\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30bb\u30c3\u30c8\u3068\u30c6\u30b9\u30c8\u30bb\u30c3\u30c8\u306b\u5206\u5272\uff088:2\uff09\r\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)<\/code><\/pre>\r\n<\/div>\r\n
\u5b66\u7fd2\u6e08\u307f\u30e2\u30c7\u30eb\u306e\u6e96\u5099<\/strong><\/span><\/p>\r\n
\u4eca\u56de\u306f\u3001scikit-learn\u306e GradientBoostingRegressor<\/strong><\/a>\u00a0\u3092\u7528\u3044\u3066\u4e88\u6e2c\u30e2\u30c7\u30eb\u3092\u5b66\u7fd2\u3057\u307e\u3059\u3002<\/p>\r\n
PI\u306f\u5b66\u7fd2\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306b\u4f9d\u5b58\u305b\u305a\u306b\u7279\u5fb4\u91cd\u8981\u5ea6\u3092\u8a08\u7b97\u3067\u304d\u308b\u305f\u3081\u3001\u89e3\u304d\u305f\u3044\u554f\u984c\u8a2d\u5b9a\u306b\u5fdc\u3058\u3066\u3001\u4ed6\u306e\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3092\u7528\u3044\u308b\u3053\u3068\u3082\u3067\u304d\u307e\u3059\u3002<\/p>\r\n
\r\n
# \u30e2\u30c7\u30eb\u306e\u69cb\u7bc9\r\ngbr = GradientBoostingRegressor()\r\n\r\n# \u30e2\u30c7\u30eb\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\r\ngbr.fit(X_train, y_train)<\/code><\/pre>\r\n<\/div>\r\n
PI\u306e\u8a08\u7b97<\/strong><\/span><\/p>\r\n
Python\u3067\u306f\u3001scikit-learn\u306e permutation_importance<\/strong>\u00a0\u3092\u7528\u3044\u3066\u7c21\u5358\u306bPI\u3092\u8a08\u7b97\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002<\/p>\r\n
\r\n
# \u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u3092\u7528\u3044\u3066\u7279\u5fb4\u91cf\u91cd\u8981\u5ea6\u3092\u8a08\u7b97\r\nresult = permutation_importance(gbr, X_test, y_test, n_repeats=10, \r\n                                random_state=42, n_jobs=-1)<\/code><\/pre>\r\n<\/div>\r\n
\u3053\u3053\u3067\u306f\u3001\u5404\u7279\u5fb4\u91cf\u306e\u30b7\u30e3\u30c3\u30d5\u30eb\u309210\u56de\u7e70\u308a\u8fd4\u3057\u3066\u3001\u4e88\u6e2c\u7cbe\u5ea6\u306e\u5dee\u306e\u5e73\u5747\u3092\u7279\u5fb4\u91cd\u8981\u5ea6\u3068\u3057\u3066\u3044\u307e\u3059\u3002<\/p>\r\n

\r\n\u7d50\u679c\u306e\u53ef\u8996\u5316<\/strong><\/span><\/p>\r\n
\r\n
# \u7d50\u679c\u3092\u53ef\u8996\u5316\r\nsorted_idx = result.importances_mean.argsort()\r\n\r\nplt.figure(figsize=(10, 6))\r\nplt.barh(\r\n\u00a0 \u00a0 X_test.columns[sorted_idx], result.importances_mean[sorted_idx],\r\n\u00a0 \u00a0 color='skyblue', edgecolor='black',\r\n\u00a0 \u00a0 )\r\nplt.xlabel('\u30d1\u30fc\u30df\u30e5\u30c6\u30fc\u30b7\u30e7\u30f3\u306b\u3088\u308b\u91cd\u8981\u5ea6')\r\nplt.title('PI\u306b\u3088\u308b\u7279\u5fb4\u91cf\u91cd\u8981\u5ea6')\r\nplt.show()<\/code><\/pre>\r\n<\/div>\r\n
\u7d50\u679c\u3068\u3057\u3066\u3001\u4ee5\u4e0b\u306e\u3088\u3046\u306a\u6a2a\u68d2\u30b0\u30e9\u30d5\u304c\u63cf\u753b\u3055\u308c\u307e\u3059\u3002
\r\n\u7e26\u8ef8\u304c\u7279\u5fb4\u91cf\u3001\u6a2a\u8ef8\u304c\u91cd\u8981\u5ea6\u3092\u8868\u3057\u307e\u3059\u3002<\/p>\r\n
\"\"<\/p>\n\n
\u30e1\u30ea\u30c3\u30c8\u3068\u9650\u754c<\/h2>\n
PI\u306b\u306f\u3001\u4e3b\u306b\u6b21\u306e\u3088\u3046\u306a\u30e1\u30ea\u30c3\u30c8\u304c\u3042\u308a\u307e\u3059\u3002<\/p>\r\n
\u76f4\u611f\u7684\u306b\u7406\u89e3\u3067\u304d\u308b<\/span>
\r\n<\/strong>\u8a08\u7b97\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u304c\u30b7\u30f3\u30d7\u30eb\u3067\u3042\u308b\u305f\u3081\u3001\u30e2\u30c7\u30eb\u306e\u6319\u52d5\u3092\u628a\u63e1\u3057\u3084\u3059\u3044\u3002<\/p>\r\n
\u6a5f\u68b0\u5b66\u7fd2\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306b\u4f9d\u5b58\u3057\u306a\u3044<\/span>
\r\n<\/strong>\u30e2\u30c7\u30eb\u306e\u5b66\u7fd2\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306b\u4f9d\u5b58\u305b\u305a\u306b\u4f7f\u7528\u3067\u304d\u308b\u305f\u3081\u3001\u30e2\u30c7\u30eb\u9593\u306e\u6bd4\u8f03\u3082\u53ef\u80fd\u3002<\/p>\r\n
\u30e2\u30c7\u30eb\u306e\u518d\u5b66\u7fd2\u306e\u5fc5\u8981\u6027\u304c\u306a\u3044<\/span>
\r\n<\/strong>PI\u3067\u306f\u5b66\u7fd2\u6e08\u307f\u30e2\u30c7\u30eb\u3092\u5229\u7528\u3057\u3066\u91cd\u8981\u5ea6\u3092\u8a55\u4fa1\u3067\u304d\u308b\u305f\u3081\u3001\u65b0\u305f\u306b\u30e2\u30c7\u30eb\u3092\u518d\u5b66\u7fd2\u3059\u308b\u5fc5\u8981\u304c\u3042\u308a\u307e\u305b\u3093\u3002
\r\n\u91cd\u8981\u5ea6\u3092\u77e5\u308a\u305f\u3044\u7279\u5fb4\u91cf\u3092\u30b7\u30e3\u30c3\u30d5\u30eb\u3059\u308b\u4ee3\u308f\u308a\u306b\u3001\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u304b\u3089\u53d6\u308a\u9664\u3044\u3066\u4e88\u6e2c\u7cbe\u5ea6\u306e\u4f4e\u4e0b\u3092\u6e2c\u308b\u3053\u3068\u3067\u91cd\u8981\u5ea6\u3092\u8a08\u7b97\u3059\u308b\u3053\u3068\u3082\u3067\u304d\u307e\u3059\u304c\u3001\u305d\u306e\u5834\u5408\u306f\u30e2\u30c7\u30eb\u306e\u518d\u5b66\u7fd2\u304c\u5fc5\u8981\u306b\u306a\u308a\u307e\u3059\u3002<\/p>\r\n
 <\/p>\r\n
\u4e00\u65b9\u3067PI\u306b\u306f\u6b21\u306e\u3088\u3046\u306a\u9650\u754c\u304c\u3042\u308b\u305f\u3081\u3001\u4f7f\u7528\u3059\u308b\u969b\u306b\u306f\u6ce8\u610f\u304c\u5fc5\u8981\u3067\u3059\u3002<\/p>\r\n
\u56e0\u679c\u95a2\u4fc2\u3068\u3057\u3066\u89e3\u91c8\u3067\u304d\u306a\u3044<\/span>
\r\n<\/strong>PI\u3067\u8a08\u7b97\u3055\u308c\u308b\u306e\u306f\u3042\u304f\u307e\u3067\u4e88\u6e2c\u30e2\u30c7\u30eb\u306b\u304a\u3051\u308b\u7279\u5fb4\u91cf\u3068\u30bf\u30fc\u30b2\u30c3\u30c8\u5909\u6570\u306e\u76f8\u95a2\u95a2\u4fc2\u3067\u3042\u308b\u3002\u91cd\u8981\u5ea6\u304c\u9ad8\u3044\u3068\u8a08\u7b97\u3055\u308c\u305f\u7279\u5fb4\u91cf\u3068\u30bf\u30fc\u30b2\u30c3\u30c8\u5909\u6570\u3068\u306e\u9593\u306b\u56e0\u679c\u95a2\u4fc2\u304c\u3042\u308b\u3068\u306f\u9650\u3089\u306a\u3044\u70b9\u306b\u6ce8\u610f\u3059\u308b\u5fc5\u8981\u304c\u3042\u308a\u307e\u3059\u3002<\/p>\r\n
\u7279\u5fb4\u91cf\u9593\u306e\u76f8\u95a2\u306e\u5f71\u97ff<\/span>
\r\n<\/strong>\u76f8\u95a2\u304c\u5f37\u3044\u7279\u5fb4\u91cf\u306e\u30da\u30a2\u304c\u3042\u308b\u5834\u5408\u3001PI\u306e\u5024\u304c\u6b6a\u3080\u53ef\u80fd\u6027\u304c\u3042\u308a\u307e\u3059\u3002 \u305d\u306e\u5834\u5408\u306f\u5225\u306e\u624b\u6cd5\u3092\u7528\u3044\u308b\u3053\u3068\u304c\u63a8\u5968\u3055\u308c\u307e\u3059\u3002<\/p>\r\n
<\/p>\n\n
\u307e\u3068\u3081<\/h2>\n
Permutation Importance\u306f\u3001\u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u306b\u30b0\u30ed\u30fc\u30d0\u30eb\u306a\u89e3\u91c8\u3092\u4e0e\u3048\u308b\u6709\u52b9\u306a\u624b\u6cd5\u3067\u3059\u3002\u3042\u308b\u7279\u5fb4\u91cf\u306e\u5024\u3092\u30b7\u30e3\u30c3\u30d5\u30eb\u3059\u308b\u3053\u3068\u306b\u3088\u308b\u4e88\u6e2c\u7cbe\u5ea6\u306e\u4f4e\u4e0b\u3092\u6e2c\u308b\u3068\u3044\u3046\u30b7\u30f3\u30d7\u30eb\u306a\u624b\u6cd5\u3067\u3042\u308a\u3001\u76f4\u611f\u7684\u306b\u89e3\u91c8\u3059\u308b\u3053\u3068\u304c\u53ef\u80fd\u3067\u3059\u3002\u4e00\u65b9\u3067\u3001\u56e0\u679c\u95a2\u4fc2\u306e\u8aac\u660e\u306b\u306f\u4f7f\u7528\u3067\u304d\u305a\u3001\u7279\u5fb4\u91cf\u9593\u306e\u76f8\u95a2\u306b\u5f31\u3044\u306a\u3069\u3001\u3044\u304f\u3064\u304b\u306e\u9650\u754c\u3082\u5b58\u5728\u3057\u307e\u3059\u3002\u3053\u308c\u3089\u306e\u9650\u754c\u3092\u7406\u89e3\u3059\u308b\u3053\u3068\u304c\u3001PI\u3092\u7528\u3044\u3066\u30e2\u30c7\u30eb\u3092\u6b63\u3057\u304f\u89e3\u91c8\u3059\u308b\u305f\u3081\u306b\u91cd\u8981\u3067\u3059\u3002<\/p>\n\n
\u53c2\u8003\u8cc7\u6599<\/h2>\n
\u68ee\u4e0b\uff082021\uff09\u300e\u6a5f\u68b0\u5b66\u7fd2\u3092\u89e3\u91c8\u3059\u308b\u6280\u8853\u300f\u6280\u8853\u8a55\u8ad6\u793e<\/p>\r\n
[blogcard url=”https:\/\/gihyo.jp\/book\/2021\/978-4-297-12226-3″]<\/p>\r\n
<\/p>\r\n
Christoph Molnar (2024) Interpretable Machine Learning.
\r\n<\/span><\/p>\r\n
[blogcard url=”https:\/\/christophm.github.io\/interpretable-ml-book\/”]<\/p>\r\n

\r\n<\/span><\/span><\/p>","protected":false},"excerpt":{"rendered":"
\u30d3\u30b8\u30cd\u30b9\u3067\u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u3092\u6d3b\u7528\u3059\u308b\u5834\u5408\u306b\u306f\u3001\u30d6\u30e9\u30c3\u30af\u30dc\u30c3\u30af\u30b9\u306a\u30e2\u30c7\u30eb\u306e\u6319\u52d5\u3092\u89e3\u91c8\u3059\u308b\u3053\u3068\u304c\u5fc5\u8981\u3067\u3042\u308b\u5834\u5408\u304c\u3042\u308a\u307e\u3059\u3002\u672c\u8a18\u4e8b\u3067\u306f\u3001\u7279\u5fb4\u91cf\u306e\u91cd\u8981\u5ea6\u3068\u3044\u3046\u89b3\u70b9\u304b\u3089\u6a5f\u68b0\u5b66\u7fd2\u30e2\u30c7\u30eb\u3092\u89e3\u91c8\u3059\u308b\u300cPermutation Importa […]<\/p>\n","protected":false},"author":46,"featured_media":5651,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"content-type":"","swell_btn_cv_data":"","footnotes":"","_wp_rev_ctl_limit":""},"categories":[1249],"tags":[96,331,748,484,39,749],"class_list":["post-5642","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-knowledge","tag-ai","tag-python","tag-xai","tag-484","tag-39"],"_links":{"self":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts\/5642","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/users\/46"}],"replies":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/comments?post=5642"}],"version-history":[{"count":1,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts\/5642\/revisions"}],"predecessor-version":[{"id":5657,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts\/5642\/revisions\/5657"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/media\/5651"}],"wp:attachment":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/media?parent=5642"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/categories?post=5642"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/tags?post=5642"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}