Resolved: splitting the data
in contrast to what you mentioned, in regression lectures we did split the data after standarizing the inputs.
so which procedures are more accurate.
1 answers ( 1 marked as helpful)
Hey Doaa,
Thank you for reaching out!
In this current lecture—The YouTube Dataset: Preprocessing—our primary focus is on the issue of data leakage. In contrast, standardization before splitting the data hasn't been the central topic in other lectures and likely hasn't impacted the results.
For a more detailed explanation, I’ve addressed a similar concern in another Q&A thread, which you might find helpful:
https://365datascience.com/q/6670a91021
Kind regards,
365 Hristina