cshizhe / HM3DAutoVLN

Official implementation of Learning from Unlabeled 3D Environments for Vision-and-Language Navigation (ECCV'22).
34 stars 1 forks source link

cannot get score in step03 #3

Open hieu1999210 opened 1 year ago

hieu1999210 commented 1 year ago

Hi Shizhe,

I got an issue while running the step03 in the data generation pipeline. it seems like mask2former does not return scores for each instance but your pipeline requires them. Do you have to modify mask2former code to get the score? if so could you share how do you get a score for each instance?