简体   繁体   English

weka 逻辑回归分类器不工作

[英]weka logistic regression classifier not wroking

I have a text dataset of size 15MB and it has close to 27k of rows, J48 TREE and NAÏVE BAYES classifiers work just fine, but when using logistic classifier i keep getting this error enter image description here I have tried reszing the heap size but it still not working, same thing goes to deep learning and random forest.我有一个大小为 15MB 的文本数据集,它有接近 27k 的行,J48 TREE 和 NAÏVE BAYES 分类器工作得很好,但是当使用逻辑分类器时,我一直收到这个错误在这里输入图像描述我已经尝试调整堆大小但是它仍然没有用,同样的事情也适用于深度学习和随机森林。 what can i do?我能做什么?

Does your data have nominal attributes with many values?您的数据是否具有具有许多值的标称属性? If so, 2GB may not be enough.如果是这样,2GB 可能还不够。

Why?为什么? Logistic applies the NominalToBinary filter, which generates binary attributes from labels (absence or presence). Logistic应用NominalToBinary过滤器,它根据标签(不存在或存在)生成二进制属性。 That can generate a lot of additional attributes if you have many multi-valued nominal atttibutes.如果您有许多多值标称属性,那会生成很多附加属性。

Solution?解决方案? Work with a sub-sample of your data or increase the heap size.使用数据的子样本或增加堆大小。

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM