Dear all,
I have a question regarding fine-tuning vision models. When I upload a .jpg file to a server for model training, and after a day, I wish to continue training by adding more data, including additional .jpg files—what happens if the new data contains .jpg files with the same name but different results in the associated JSON files? Specifically, does OpenAI fetch the updated .jpg file from the server, or does it retrieve the cached version of the original file?
I am concerned about potential caching issues and would like to ensure that the model uses the latest .jpg file for continued training.
Thank you for your help.
Best regards, Morteza