训练后如何获取神经网络参数? [英] How to get neural network parameter after training?

问题描述

我正在使用matlab的神经网络进行分类. 我想知道如何存储网络参数，例如:时间，时间，毫秒等.训练后在矩阵中存储

I am using matlab's neural network for classification. I want to know how can I store network parameters such as: epoches, time, mse, etc. In a matrix after training?

非常感谢

推荐答案

调用火车，返回的第二个参数是

When calling train, the second returned argument is the training record, this contains the epoches, time and other info about the training. e.g.

[net,tr] = train(net,data,target);
tr.epoch
tr.time

对于 mse ，给出测试数据data，目标数据target和神经网络net:

For the mse, given test data data, target data target and neural network net:

%run inputs through network 
result = net(data);
%get the error
error = result - target;
%calculate mse
mse(error)

这篇关于训练后如何获取神经网络参数?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！