How to use it with multivariate time series????

[12suyash]

Plz tell where in the code should I change the parameters to make it multivariate

[hfawaz]

You should first make sure that all of your Multivariate time series are of equal length.

So if you have an x_train whose shape is equal to [n,l,m] with m being the number of dimensions, then the code should work just fine.

[12suyash]

My series is of equal length and have 39 dimension can you plz elaborate where in code should I change can you tell me the line number

[hfawaz]

The only thing you need to change is to re-define this function and change the name of the archive according to your folder that contains the dataset.

For example if you have your dataset in /archive_name/dataset_name/data.csv you should change this line to UNIVARIATE_ARCHIVE_NAMES = ['archive_name'].
You should also add here the following UNIVARIATE_DATASET_NAMES=['dataset_name'].

Do not forget to re-define the read_all_datasets function according to your data.