Did not find decoder for lvis:1.3.0. Please specify decoders for all datasets in DECODERS

Question

Did not find decoder for lvis:1.3.0. Please specify decoders for all datasets in DECODERS

mayank0621 opened this issue 8 months ago · 6 comments

Hi everyone,
when running,
python -m scenic.projects.owl_vit.main
--alsologtostderr=true
--workdir=/tmp/training
--config=scenic/projects/owl_vit/configs/clip_b32_finetune.py

command as per readme I'm getting below error:
Did not find decoder for lvis:1.3.0. Please specify decoders for all datasets in DECODERS

I'm using:
ubuntu 20.04
jaxlib 0.4.20+cuda12.cudnn89
lvis 0.5.3
python-lvis 1.1.0

logs are in this file: log.txt
all library version of my environment can be found here:
library_version.txt

any help would be great for me.
Thanks

Answer 1 · 2024-05-21T08:13:59.000Z

I also got the same issue but different jaxlib version

Answer 2 · 2024-05-21T10:34:52.000Z

are you able to solve that?

Answer 3 · 2024-05-21T10:39:27.000Z

No, I searched a lot still no idea

Answer 4 · 2024-07-08T07:17:20.000Z

I am getting same error. did you find any solution? Were you able to find any info on what might be the reason for this error? Any kind of information.

Answer 5 · 2024-07-22T07:04:23.000Z

I was finally able to solve the error.

At scenic/scenic/projects/owl_vit/preprocessing/input_pipeline.py line no 65, there is a list of DECODERS. You have to change the version number of livs to 1.3.0. That's it. If you want to use different dataset, you just have to add the name of your dataset and which decoder you want to use. You have to make some changes as shown here.