def GMM():
data_mfcc,data_label = IEMPCAP_dataset.ge_dataset()
print(len(data_mfcc),":",len(data_label))
# GMM模型
K = 4
gmm = GMM(data_mfcc,K)
gmm.GMM_EM()
y_pre = gmm.prediction
print("GMM预测结果:\n",y_pre)
print("GMM正确率为:\n",accuracy_score(data_label,y_pre))
plt.scatter(data_mfcc[:, 0], data_mfcc[:, 1], c=y_pre)
plt.title("GMM结果显示")
plt.show()
data, data_label = IEMPCAP_dataset.ge_dataset()
iris_X = data # 特征
iris_Y = data_label # 标签
X_train, X_test, Y_train, Y_test = train_test_split(iris_X, iris_Y, test_size=0.2,random_state=True) #
knn = KNeighborsClassifier() # K近邻(K-Nearest Neighbor)分类器
knn.fit(X_train, Y_train) # 进行分类
Y_predict = knn.predict(X_test)
print(accuracy_score(Y_predict, Y_test)) # 正确率
传统机器学习实现情感识别分类任务