Creating dataloader for multivariate time series dataset

Dear All,

I am trying to classify the multivariate time series (MTS) data. Let’s say the data set contains 10000 data samples. Out of the 10000 data samples, training data contains 6000 data samples, 2000 data samples for validation data, and the remaining 2000 data samples in the test dataset.
I want to write PyTorch based script that uses CNN to classify the MTS. For this purpose, I need to create a data loader to load the data from the CSV files that contain MTS data. I am very new to PyTorch. Can you please help me with the steps to create a data loader from the MTS.

Thanks
Achyut