ANN的输出值不稳定并提高了准确性 [英] Unstable output values from ANN and improving accuracy

问题描述

我正在尝试使用PyBrain开发一个人工神经网络来对生物学数据进行建模.我的ANN可以编译并运行，但是其准确性值非常低，从未超过〜62％.从编码角度来看，如何提高ANN的准确性?我注意到的是，即使测试数据集没有变化，每次ANN的输出也不相同，这是ANN动作不稳定的原因吗，我该如何改善呢?

I am trying to develop an Artificial Neural Network using PyBrain to model biological data. My ANN compiles and runs, but its accuracy value is very low, never surpassing ~62%. From a coding perspective, how can I improve the ANN's accuracy? Something I noticed was that each time, the outputs of the ANN are not the same, either, even though the test data set doesn't change--is there a reason the ANN is acting to unstably, and how can I improve this?

谢谢！ :)

推荐答案

如果每次运行脚本都创建新网络，则输出不同是正常的.

If you creating new network each time you run your script then it is normal that outputs are different.

每次创建ANN pybrain都会使用随机值(范围为0到1)初始化连接权重.

Each time you create ANN pybrain initialize weights of connections with random values (range 0 to 1).

您可以使用NetworkWriter保存ANN并使用pybrain.tools.customxml中的NetworkReader读取它(请参阅代码文档以获取参考，pybrain API缺少一些内容).

You can save your ANN with NetworkWriter and read it with NetworkReader in pybrain.tools.customxml (see code documentation for reference, pybrain API is missing few things).

您可以通过学习速度和动力来调整培训过程.另外，您可以将更多的培训纪元应用到您的网络中.

You can adjust training process with learning rate and momentum. Also you could apply more training epoch to your network.

如果您提供代码，我可以说更多.

If you provide your code I could say more.

这篇关于ANN的输出值不稳定并提高了准确性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！