如何使用caffe预测浮点矢量标签? [英] How to predict float vector labels with caffe?

问题描述

我想知道是否可以使用caffe预测与输入图像相关的1-by-n功能. 在此帖子中，有一种解决方案使caffe预测二进制矢量，例如[1 0 1 0].

I was wondering if it's possible to predict a 1-by-n feature associated to an input image using caffe. In this post there is a solution to make caffe predict a binary vector such as [1 0 1 0].

如果我有一个1-by-n浮点矢量作为标签(例如[0.2、0.1、0.3、0.4]，我想预测这样的矢量，而不是二进制矢量标签)，这种解决方案是否也合适?

Is this solution also suitable if I have a 1-by-n float vector as label (such as [0.2, 0.1, 0.3, 0.4] ? I want to predict such a vector, not a binary vector label.

推荐答案

您还可以考虑以下 MultiTaskData层.它可以解析您在问题中提到的浮点型标签向量.

You can also think about this MultiTaskData Layer. It can parse float typed label vector as you mentioned in your question.

这篇关于如何使用caffe预测浮点矢量标签?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！