Open revaeb opened 2 years ago
Considering the fairness of the experimental comparison, the data list of all experiments is not fixed. The experimental results under each label-ratios are averaging the mAP from three randomly split data lists.
Hope that helps!
Considering the fairness of the experimental comparison, the data list of all experiments is not fixed. The experimental results under each label-ratios are averaging the mAP from three randomly split data lists.
Hope that helps!
Thx for your reply! So you just split the whole training set into different labeled ratio like 1/8 labeled images and 7/8 unlabeled images and test the model on the whole validation set?
yeap!
revaeb @.***> 于2022年3月21日周一 15:52写道:
Considering the fairness of the experimental comparison, the data list of all experiments is not fixed. The experimental results under each label-ratios are averaging the mAP from three randomly split data lists.
Hope that helps!
Thx for your reply! So you just split the whole training set into different labeled ratio like 1/8 labeled images and 7/8 unlabeled images and test the model on the whole validation set?
— Reply to this email directly, view it on GitHub https://github.com/yh-pengtu/GuidedMix-Net/issues/2#issuecomment-1073567621, or unsubscribe https://github.com/notifications/unsubscribe-auth/AQUEE3GEWKKVDMIMF6FQ4D3VBATFFANCNFSM5PAX72WQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
You are receiving this because you commented.Message ID: @.***>
Congratulations for your excellent work! I want to commit semi-supervised experiments on Pascal Context dataset and compare the results with your work. So could you plz offer me the data-split files used in the experiments on Pascal Context? Thank you so much! :)