Hello, I am participating in this challenge. i am trying to submit my predictions and evaluate. But I am getting the same error each time: "not all filenames in the archive follow the expected naming format" I am using this naming convention: BraTS-GoAT-case_id-000.nii.gz,BraTS-GoAT-case_id-000-0000.nii.gz,BraTS-GoAT-case_id-000-0001.nii.gz,BraTS-GoAT-case_id-000-0002.nii.gz BraTS-GoAT-case_id-000-0003.nii.gz Can someone help me to identofy the error if I have any in this naming convention? I have another question about the validation set: do we need to use the validation data without GT only for this evaluation? I mean can I use those in my training anyway? Thank you in advance for any help.

Created by Mumu Aktar sam_challenge_t7
Hi@mumuaktar,@vchung I really appreciate your suggestions and help. I will try again and keep you updated. Thank you very much!
@vchung I’m sorry for not being clear earlier. Thank you very much for your help!
Thanks, @mumuaktar! @Liu-Lydia, in addition to their suggestion, your prediction filenames should contain the 5-digit case IDs from the validation dataset, as this is how evaluation determines the corresponding groundtruth files. Using arbitrary numbers (like 00000 from your example) will result in the "Unknown case IDs found" validation error. Hope this helps!
Hello @Liu-Lydia, I did the same mistake before. In the submission portal, you just need to upload the predicted files named as: BraTS-GoAT-caseid.nii.gz , that's it, nothing else which will lead to error. I hope it will help you to run!
Hi @vchung, I submitted files named like BraTS-GoAT-00000.nii.gz, but I got a message saying my submission (ID: 9754008) was invalid due to unknown scan IDs. Could you please clarify what naming format I should follow? I'd really appreciate your help. Thanks!
@mumuaktar , Of course! Happy to hear you were able to submit successfully :-)
Thank you! It worked. By mistake, I was uploading the validation cases also along with the predictions.
Hi @mumuaktar , All files in your submitted zipfile must end with the 5-digit case ID, followed by `.nii.gz`. Looking at your latest submission (ID: 9753745), I see some filenames like `BraTS-GoAT-02489-0000.nii.gz`, which do not follow this required convention. This is why your submissions are not passing validation. > I am using this naming convention: BraTS-GoAT-caseid-0002.nii.gz (for one channel) and BraTS-GoAT-caseid.nii.gz (for prediction). It sounds like you may be including more files in your zipfile than required. It should only contain the predictions. Hope this helps!
@vchung For your note, I tried the name of the cases with both time point 000 added and without it as you recommended for the prediction files. I got the same error. It would be great if you could suggest further. Thanks
Hello, Correcting those I am still getting the same error:below are the invalid reasons: Not all filenames in the archive follow the expected naming format. Please check the Submission Tutorial of the task you're submitting to for more details. I am using this naming convention: BraTS-GoAT-caseid-0002.nii.gz (for one channel) and BraTS-GoAT-caseid.nii.gz (for prediction). Thank you
Hi @mumuaktar, > Can someone help me to identofy the error if I have any in this naming convention? Apologies for the confusion! For the Generalizability task, since the filenames of the BraTS-GoAT dataset only provide the 5-digit case ID, this will be all you need for your prediction files. The dash and 3-digit timepoint are not required in the filenames for this task. Examples of valid prediction filenames for BraTS-GoAT would be: * {case_id}.nii.gz * seg-{case_id}.nii.gz * BraTS-GoAT-{case_id}.nii.gz * BraTS-GoAT-seg-{case_id}.nii.gz

Task 7-Generalizability across Tumors page is loading…