search

chunmeifeng / DiffTPT

【ICCV 2023】Diverse Data Augmentation with Diffusions for Effective Test-time Prompt Tuning
56 stars 3 forks source link

Question about the test set. #8

Open haoweiz23 opened 3 days ago

haoweiz23 commented 3 days ago

Hi, thank you for your work. I am attempting to reproduce the results from your paper and noticed that you randomly selected 1,000 samples to construct the test set. Since the specific test samples may differ due to random selection, my reproduced results might vary from yours. Could you please provide the list of test samples used in each test set in your paper? This would be very helpful for ensuring consistency in the results.

Thanks!

chunmeifeng commented 3 days ago

Hi haowei, I also find this question when I reproduce and extend my experiments to the journal. Since I haven't saved the sample list before, I am trying to replicate the previous experiment by conducting several random samples.

haoweiz23 @.***> 于2024年11月5日周二 16:41写道：

Hi, thank you for your work. I am attempting to reproduce the results from your paper and noticed that you randomly selected 1,000 samples to construct the test set. Since the specific test samples may differ due to random selection, my reproduced results might vary from yours. Could you please provide the list of test samples used in each test set in your paper? This would be very helpful for ensuring consistency in the results.

Thanks!

— Reply to this email directly, view it on GitHub https://github.com/chunmeifeng/DiffTPT/issues/8, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR75XN5MFETLWLJ4PHSJGHTZ7CAEZAVCNFSM6AAAAABRF76QMGVHI2DSMVQWIX3LMV43ASLTON2WKOZSGYZTINZZHAYDMOA . You are receiving this because you are subscribed to this thread.Message ID: @.***>