python opencv3.x中svm的模型儲存與載入__python

廢話不多說直接上代碼:

import numpy as npfrom sklearn import datasetsX,y = datasets.make_classification(n_samples=100,n_features=2,n_redundant=0,n_classes=2,random_state=7816)print(X.shape,y.shape)X = X.astype(np.float32)y = y * 2 - 1'''分離資料'''from sklearn import model_selection as msX_train, X_test, y_train, y_test = ms.train_test_split(    X, y, test_size=0.2, random_state=42)import cv2svm = cv2.ml.SVM_create()svm.setKernel(cv2.ml.SVM_LINEAR)'''開始訓練'''y_train = y_train.reshape(-1, 1)# print(y_train)svm.train(X_train, cv2.ml.ROW_SAMPLE, y_train)svm.save("svmtest.mat")print ("Done\n")svm2 = cv2.ml.SVM_load("svmtest.mat")# svm2.load("svmtest.mat")# print(svm2)'''開始預測'''_, y_pred = svm2.predict(X_test)'''用scikit-learn的metrics模組計算準確率'''from sklearn import metricsprint(metrics.accuracy_score(y_test, y_pred))

關鍵代碼如下:

建立：

import cv2svm = cv2.ml.SVM_create()svm.setKernel(cv2.ml.SVM_LINEAR)

其它的寫法都是以前較老的版本，基本上都不行

載入：

svm2 = cv2.ml.SVM_load("svmtest.mat")