train, validation and test data set

[abramed]

Hello,

I am a thesis student currently working on timeseries data, and I am interested in using your code to assist with my research. However, I have encountered an issue during the preprocessing phase where the missing mask generates several pickle files. Subsequently, when I attempt to run the 1D convolution or any other file, I observe that it is trying to import files with "_train," "_val," and "test" suffixes, as evident in the attached screenshot. It appears that these files haven't been created previously.

To address this, I am unsure whether I should run the preprocessing file with different imports or if I need to split the file into three parts and save them as pickles separately. Could you please provide guidance on the appropriate steps to resolve this issue?

Thank you for your time and assistance.