MATLAB LSTM chickenpox example is the only working case
9 ビュー (過去 30 日間)
古いコメントを表示
Hi,
I'm studying LSTM networks forecasting, looking at the chickenpox example on MATLAB website, I've learned how to perform a multistep forecasting of time series.
I applied those concepts to a simple sin function time history and it worked, then I used another time history, similar to the chickenpox cases one, give by MATLAB and the forecasting didn't worked.
Even modifying the lstm layer properties or the training options. I tried also to modify the architecture. It simply doesn't work.
The "predicting" is working properly but "forecasting" is only working on that specific time series given for the example.
I attach the test time history I'm trying to forecast, the result is a line or at maximum a sinusoidal shape, not reproducing all the peaks of the original time history.
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/279439/image.png)
1 件のコメント
diana haron
2020 年 9 月 27 日
Hi, how does your data looks like? Is it similar as the chicken pox example where it is just one row with multiple columns ?
回答 (0 件)
参考
カテゴリ
Help Center および File Exchange で Sequence and Numeric Feature Data Workflows についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!