Given the training data the algorithm estimates mean vectors and covariation matrices for every class and then it uses them for prediction.

• 当给定了训练数据，算法将会估计每一个类别的向量均值和方差矩阵，然后根据这些进行预测。