Time series forecasting using deep learning with 2 numFeatures

Question

israt fatema 2020-9-15

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/593737-time-series-forecasting-using-deep-learning-with-2-numfeatures

评论： israt fatema 2021-2-12

How to use two input features to get 1 output using this example?

https://au.mathworks.com/help/deeplearning/ug/time-series-forecasting-using-deep-learning.html

This example use 1 numFeatures, for example if i have time series data with 2 features, wind speed and temperature, how can i use this example to forecast weather?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Anshika Chaurasia 2020-9-30

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/593737-time-series-forecasting-using-deep-learning-with-2-numfeatures#answer_502753

在 MATLAB Online 中打开

Hi Israt,

It is my understanding that you have two features as input and want to train the network with this input.

You could change numFeatures = 2 while defining sequenceInputLayer function as shown:

numFeatures = 2; %take 2 features as input
numResponses = 1; %give 1 output
layers = [ ...
    sequenceInputLayer(numFeatures)
    lstmLayer(numHiddenUnits)
    fullyConnectedLayer(numResponses)
    regressionLayer];

2 个评论
显示无隐藏无

israt fatema 2020-10-29

Thank you for the response. In this case how can i read and load the time series data with these two features such as wind speed and temperature?

The example contains a single time series, the output is a cell array, where each element is a single time step and then reshaped the data to be a row vector.

israt fatema 2021-2-12

Hi, the Matlab example https://au.mathworks.com/help/deeplearning/ug/time-series-forecasting-using-deep-learning.html read 1 (one feature). I couldn't read more than 1 feature from the dataset (timeseries) if i want to use numFeatures = 2

%data = chickenpox_dataset;

%data = [data{:}];

How to read more than 1 (for example numFeatures = 2 and output = 1) feature from the dataset to get 1 output?

Thanks in advance.

请先登录，再进行评论。