search

TRI-ML / packnet-sfm

TRI-ML Monocular Depth Estimation Repository
https://tri-ml.github.io/packnet-sfm/
MIT License
1.23k stars 242 forks source link

ERROR：Unknown class PackNetSlimEnc01 #175

Closed Cc-Hy closed 2 years ago

Cc-Hy commented 3 years ago

I'm trying to do inference on a KITTI 3D detection image with command: scripts/infer.py --checkpoint ckpt/PackNetSAN01_HR_sup_K.ckpt --input ../data/kitti/training/image_2/000002.png --output output/000002.png with ckpt 'PackNetSAN, Supervised (densified GT), 352x1216, K' And in line model_wrapper= ModelWrapper(config, load_datasets=False) I got error Unknown class PackNetSlimEnc01 image Can you help me with it? Thank you.

alphaxvzf commented 3 years ago

This error only appears with https://tri-ml-public.s3.amazonaws.com/github/packnet-sfm/models/PackNet01_MR_selfsup_D.ckpt checkpoint. Probably, it cannot find the corresponding DenseNet. There is no problem with, for example, with https://tri-ml-public.s3.amazonaws.com/github/packnet-sfm/models/PackNet01_MR_semisup_CStoK.ckpt checkpoint .

VitorGuizilini-TRI commented 2 years ago

Hi, sorry for the delay in replying. The configuration file in that checkpoint uses an older name for the packnet-san model. If you replace it with the proper name "PackNetSAN01" it should work. I hope that helps!

alphaxvzf commented 2 years ago

Hello, Thank you for your reply. Actually, I fixed it. I only wanted to note that the old error still existed in the repository. Best regards, Pavel Redkin.

 

 

  Hi, sorry for the delay in replying. The configuration file in that checkpoint uses an older name for the packnet-san model. If you replace it with the proper name "PackNetSAN01" it should work. I hope that helps! — You are receiving this because you commented. Reply to this email directly, view it on GitHub , or unsubscribe . Triage notifications on the go with GitHub Mobile for iOS or Android .
    С уважением, Павел Редькин @.***  

VitorGuizilini-TRI commented 2 years ago

Thank you, I will push a fix for that soon!