trainNetwork error- categorical responses

3 次查看（过去 30 天）
Marianna Kocturova
Marianna Kocturova 2020-10-21
回答： Srivardhan Gadila 2020-10-28
Hi, I am not able to resolve the problem with my Neural network
I tried my network on example data set: Breast Cancer
inputs: cancerInputs 9x699
targets: cancerTargets 2x699
I change targets in categorical, but there is still the same error:
>> c = categorical(cancerTargets);
>> net = trainNetwork(cancerInputs,c,layers_4,options)
Error using trainNetwork (line 170)
Invalid training data. Responses must be a vector of categorical responses, or a cell array of categorical response sequences.

请先登录，再回答此问题。

回答（1 个）

Srivardhan Gadila
Srivardhan Gadila 2020-10-28
You can refer to the docoumenatation of trainNetwork.
As per your syntax of trainNetwork: net = trainNetwork(X,Y,layers,options), Y i.e., c in your question should be a N-by-1 categorical vector of labels, where N is the number of observations(=699 in your question). Make sure that the data format of your training data cancerInputs & c is according to the data format mentioned in the docoumenatation of trainNetwork.

类别

在 Help Center 和 File Exchange 中查找有关 Gaussian Process Regression 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