关于Matlab的Nerual Network Toolbox中TrainOptions的Loss Function的理解

TrainOptions函数用处如下：

options = trainingOptions(solverName)

options = trainingOptions(solverName,Name,Value)

 

options = trainingOptions('sgdm',...
    'LearnRateSchedule','piecewise',...
    'LearnRateDropFactor',0.2,...
    'LearnRateDropPeriod',5,...
    'MaxEpochs',20,...
    'MiniBatchSize',64,...
    'Plots','training-progress')

　具体可以点击网页

 

而损失函数的用处是和最后一层名字相关 原文说明如下：

Training loss, smoothed training loss, and validation loss — The loss on each mini-batch, its smoothed version, and the loss on the validation set, respectively. If the final layer of your network is a classificationLayer, then the loss function is the cross entropy loss. For more information about loss functions for classification and regression problems, see Output Layers.

所以说 所有网络中最后有一层是classificationLayer的 都是使用cross entropy交叉熵函数作为损失函数的。