search

treder / MVPA-Light

Matlab toolbox for classification and regression of multi-dimensional data
MIT License
70 stars 35 forks source link

Duplicates in k-fold #46

Open SueJaH opened 6 months ago

SueJaH commented 6 months ago

I'm curious about why duplicates are appearing here(more participants, greater duplicates). Could this issue potentially impact my results?

4c8369365c0b286fa2259aee5b64d83
treder commented 6 months ago

Hi @SueJaH , can you give some more details on the problem?

You are oversampling the data, so some classes will be oversampled by design. Is this what you mean?

SueJaH commented 6 months ago

Hi, Treder！

Thank you for your reply! I have solved this problem. Actually, that is caused by a small error in the for loop. I'm so sorry about that.🤯

In fact, I have another question. I wonder whether the multiple structs(in cell of result) I generated in level 1 have been analyzed together in the function that executes cluster permutation (mv_statistic)? Do I need to cluster permutation for each struct loop？

treder commented 6 months ago

Hi again - you might have to combine them, but it's hard to know without more details about your analysis. Can you please review understanding_statistics starting from example 7, it provides some examples on how to conduct group-level stats. Let me know if you have any more questions.