试用软件

Error using trainNetwork: asking for same sequence length.

1 次查看（过去 30 天）

显示更早的评论

Andrew 2022-8-29

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1789595-error-using-trainnetwork-asking-for-same-sequence-length

回答： Ranjeet ，2023-6-8

For some reason, trainNetwork funciton is not working while pulling out this error:

Error using trainNetwork
Invalid training data. Sequence responses must have the same sequence
length as the corresponding predictors.

This is my current code for the Training:

%% Network
layers = [
    sequenceInputLayer(1,"Name","input")
    lstmLayer(128,"Name","lstm")
    dropoutLayer(0.5,"Name","drop")
    fullyConnectedLayer(2,"Name","fc")
    softmaxLayer("Name","softmax")
    classificationLayer("Name","classification")];
miniBatchSize = 27;
options = trainingOptions('adam', ...
    'ExecutionEnvironment','cpu', ...
    'MaxEpochs',250, ...
    'MiniBatchSize',miniBatchSize, ...
    'ValidationData',{XValidation,YValidation}, ...
    'GradientThreshold',2, ...
    'Shuffle','every-epoch', ...
    'Verbose',false, ...
    'Plots','training-progress');
net = trainNetwork(XTrain,YTrain,layers,options);

and the Values inside the XTrain, YTrain, XValidation, and YValidation is,

XTrain:

YTrain:

I tried converting YTrain into the non-cell categorical variables, but it pulled out the error, so I had to use this format.

XValidation and YValidation is also in same format like above. What is the problem?

I don't see any problem with the length of sequence.

The total summary of the values are like this,

0 个评论
显示 -1更早的评论隐藏 -1更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Ranjeet 2023-6-8

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1789595-error-using-trainnetwork-asking-for-same-sequence-length#answer_1252309

Hi Andrew,

It seems you are providing wrong input sequence layer length while initializing the following layer

sequenceInputLayer(1, "Name", "input");

XTrain input is a sequence of length 150000 as shown in the provided figure. Try changing the sequence input layer initialization with

sequenceInputLayer(150000, "Name", "input");

Also, refer to the following resource dealing with same issue:

Invalid training data, incompatible sequence length

0 个评论
显示 -1更早的评论隐藏 -1更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

类别

AI, Data Science, and Statistics Deep Learning Toolbox Image Data Workflows

在 Help Center 和 File Exchange 中查找有关 Image Data Workflows 的更多信息

标签

产品

MATLAB

版本

R2022a

Community Treasure Hunt

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

Translated by

试用软件