how to define target for my neural network problem

2 次查看（过去 30 天）
muzafar
muzafar 2013-2-25
i have 48 positive and 48 negative images in separate folders,i have extracted 3328 features from each image.now i have two mat files consisting of 48 rows and 3328 columns i.e. 48x3328 where each row belongs to one image and the columns belong to their respective features.after extracting features i am combining(concatinating) then together i.e. both positive and negative feature in a single mat file in such a way that first 3328 columns belong to positive case or class one and next 3328 columns or features belong to the negative case or class 2nd.i have combined them so that i can give it to the neural network.now my question is how to define the target for my above concatenated mat file so that the i can train my network and the network will classify them correctly

请先登录，再回答此问题。

采纳的回答

Shashank Prasanna
Shashank Prasanna 2013-2-25
Your targets are positives or negatives. Which means you can create a vector of 1s or 0s to represent positives or negatives.

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by