{"id":12352,"date":"2024-10-20T16:35:03","date_gmt":"2024-10-20T07:35:03","guid":{"rendered":"http:\/\/pineplanter.moo.jp\/non-it-salaryman\/?p=12352"},"modified":"2024-10-20T21:11:32","modified_gmt":"2024-10-20T12:11:32","slug":"sklearn-svc","status":"publish","type":"post","link":"http:\/\/pineplanter.moo.jp\/non-it-salaryman\/2024\/10\/20\/sklearn-svc\/","title":{"rendered":"\u3010Python\/scikit-learn\u3011\u30b5\u30dd\u30fc\u30c8\u30d9\u30af\u30bf\u30de\u30b7\u30f3\uff08SVM\uff09\u3092\u4f7f\u3063\u305f\u30b7\u30f3\u30d7\u30eb\u306a\u5206\u985e\u30e2\u30c7\u30eb\u306e\u4f5c\u6210\u3068\u53ef\u8996\u5316"},"content":{"rendered":"

\u30b5\u30dd\u30fc\u30c8\u30d9\u30af\u30bf\u30de\u30b7\u30f3\uff08SVM\uff09\u306f\u3001\u5206\u985e\u3084\u56de\u5e30\u3001\u7570\u5e38\u691c\u77e5\u306a\u3069\u306b\u4f7f\u7528\u3055\u308c\u308b\u5f37\u529b\u306a\u6a5f\u68b0\u5b66\u7fd2\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3067\u3059\u3002SVM\u306f\u3001\u30c7\u30fc\u30bf\u3092\u7570\u306a\u308b\u30af\u30e9\u30b9\u306b\u5206\u985e\u3059\u308b\u305f\u3081\u306b\u3001\u30c7\u30fc\u30bf\u9593\u306e\u300c\u30de\u30fc\u30b8\u30f3\u300d\u3092\u6700\u5927\u5316\u3059\u308b\u5883\u754c\u7dda\uff08\u30cf\u30a4\u30d1\u30fc\u30d7\u30ec\u30fc\u30f3\uff09\u3092\u898b\u3064\u3051\u307e\u3059\u3002\u3053\u306e\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306f\u3001\u975e\u7dda\u5f62\u30c7\u30fc\u30bf\u306b\u5bfe\u3057\u3066\u3082\u30ab\u30fc\u30cd\u30eb\u30c8\u30ea\u30c3\u30af\u3068\u547c\u3070\u308c\u308b\u624b\u6cd5\u3092\u4f7f\u3063\u3066\u30c7\u30fc\u30bf\u3092\u9ad8\u6b21\u5143\u306b\u30de\u30c3\u30d4\u30f3\u30b0\u3059\u308b\u3053\u3068\u3067\u52b9\u679c\u7684\u306b\u52d5\u4f5c\u3057\u307e\u3059\u3002<\/p>\n

Scikit-learn\u3067\u306fSVM\u3092\u7c21\u5358\u306b\u5b9f\u88c5\u3067\u304d\u308b\u30e9\u30a4\u30d6\u30e9\u30ea\u3092\u63d0\u4f9b\u3057\u3066\u3044\u307e\u3059\u3002\u4ee5\u4e0b\u306fScikit-learn\u3067SVM\u3092\u4f7f\u7528\u3057\u30662\u30af\u30e9\u30b9\u5206\u985e\u3092\u884c\u3046\u4f8b\u3067\u3059\u3002Scikit-learn\u306b\u5185\u8535\u3055\u308c\u3066\u3044\u308b “`make_blobs“` \u95a2\u6570\u3092\u4f7f\u3063\u3066\u30c7\u30fc\u30bf\u3092\u751f\u6210\u3057\u5206\u985e\u3057\u307e\u3059\u3002<\/p>\n

\u30bd\u30fc\u30b9\u30b3\u30fc\u30c9<\/h2>\n
\r\n# \u5fc5\u8981\u306a\u30e9\u30a4\u30d6\u30e9\u30ea\u306e\u30a4\u30f3\u30dd\u30fc\u30c8\r\nimport numpy as np\r\nimport matplotlib.pyplot as plt\r\nfrom sklearn import datasets\r\nfrom sklearn.model_selection import train_test_split\r\nfrom sklearn.svm import SVC\r\nfrom sklearn.metrics import accuracy_score\r\n\r\n# 2\u30af\u30e9\u30b9\u306e\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u751f\u6210\r\nX, y = datasets.make_blobs(n_samples=100, centers=2, random_state=6)\r\n\r\n# \u8a13\u7df4\u30c7\u30fc\u30bf\u3068\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u306b\u5206\u5272\r\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)\r\n\r\n# SVM\u30e2\u30c7\u30eb\u306e\u4f5c\u6210\uff08\u7dda\u5f62\u30ab\u30fc\u30cd\u30eb\uff09\r\nsvm_model = SVC(kernel='linear')\r\n\r\n# \u30e2\u30c7\u30eb\u306e\u8a13\u7df4\r\nsvm_model.fit(X_train, y_train)\r\n\r\n# \u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u3092\u4f7f\u3063\u3066\u4e88\u6e2c\r\ny_pred = svm_model.predict(X_test)\r\n\r\n# \u30e2\u30c7\u30eb\u306e\u7cbe\u5ea6\u3092\u8a08\u7b97\r\naccuracy = accuracy_score(y_test, y_pred)\r\nprint(f"Accuracy: {accuracy * 100:.2f}%")\r\n\r\n# \u5883\u754c\u7dda\u3092\u63cf\u753b\r\ndef plot_svm_decision_boundary(model, X, y):\r\n    # \u8ef8\u306e\u7bc4\u56f2\u3092\u53d6\u5f97\r\n    x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1\r\n    y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1\r\n    xx, yy = np.meshgrid(np.arange(x_min, x_max, 0.01),\r\n                         np.arange(y_min, y_max, 0.01))\r\n\r\n    # \u5404\u70b9\u306e\u30af\u30e9\u30b9\u3092\u4e88\u6e2c\r\n    Z = model.predict(np.c_[xx.ravel(), yy.ravel()])\r\n    Z = Z.reshape(xx.shape)\r\n\r\n    # \u5883\u754c\u7dda\u306e\u63cf\u753b\r\n    plt.contourf(xx, yy, Z, alpha=0.8)\r\n    plt.scatter(X[:, 0], X[:, 1], c=y, edgecolors='k', marker='o')\r\n    plt.show()\r\n\r\n# \u8a13\u7df4\u30c7\u30fc\u30bf\u3068SVM\u306b\u3088\u308b\u5883\u754c\u7dda\u3092\u63cf\u753b\r\nplot_svm_decision_boundary(svm_model, X_test, y_test)\r\n \r\n<\/pre>\n
 <\/p>\n
\u30bd\u30fc\u30b9\u30b3\u30fc\u30c9\u306e\u89e3\u8aac<\/h2>\n