Closed valosekj closed 1 month ago
As clarified in this comment, the sci-zurich dataset was NOT used during v2.4 training. Thus, we should remove datasplit_sci-zurich_seed50.yaml from the datasplits folder for the v2.4 tag to avoid future confusions.
sci-zurich
datasplit_sci-zurich_seed50.yaml
Steps to do so suggested by chatGPT: https://chatgpt.com/share/670d2daf-bd6c-8008-8c0d-0e024edad032
Thanks for opening the issue and the chatgpt commands! those were really useful! Fixed now
As clarified in this comment, the
sci-zurich
dataset was NOT used during v2.4 training. Thus, we should removedatasplit_sci-zurich_seed50.yaml
from the datasplits folder for the v2.4 tag to avoid future confusions.Steps to do so suggested by chatGPT: https://chatgpt.com/share/670d2daf-bd6c-8008-8c0d-0e024edad032