[hieutq]

Dear all,
I would like to seek clarification regarding the required format for the submission file. In the provided submission.csv template, there appears to be a single row for each test file from test_0.csv to test_51.csv, formatted as follows:

file,Cycles_to_WW,Cycles_to_HPC_SV,Cycles_to_HPT_SV
test_0,0,0,0
test_1,0,0,0
…
My question is: should each row in the submission correspond to the final prediction (i.e., the last time step) from each test file, or are we expected to generate predictions for every data point within each test set and include multiple rows per file in the submission? Thank you for your guidance.

[klopetx]

Hi Hieutq,

Did you figure out the answer to this question?

It is not very clear to me what we should provide given the explanation and the example file.

The example submission file only includes a row per file, but my guts tell me it should be a row per observation…

And I am afraid I haven’t been able to provide a valid submission file yet.

Regards and thank you in advance

[snowroom]

I’m not the organizer, but I think that each row in the submission correspond to the final prediction (i.e., the last time step) from each test file. You can submit the provided submission.csv template as is.
Sorry if there are any mistakes.

[Author]

Posts