'CTranModel' object has no attribute 'module'. Did you mean: 'modules'?

[ans92]

Hi,
I want to run the model by freezing the backbone I am getting the following error in main.py line 56:
'CTranModel' object has no attribute 'module'. Did you mean: 'modules'?

When I change module to modules then I get following error on same line 56 in main.py:
AttributeError: 'function' object has no attribute 'backbone'

Could you please help me regarding this?

[jacklanchantin]

Can you share the command line you are using? Are you using GPUs?

[ans92]

Thank you for your reply
Yes I am using GPU. I am using following command line:

python3 "/home/ans/CZSL/C-Tran-main/main.py"  --batch_size 7  --lr 0.00001 --optim 'adam' --layers 3  --dataset 'mit_states' --grad_ac_step 2 --dataroot data/ --epochs 100 --freeze_backbone

[jacklanchantin]

And were you able to run the commands in the README?

[ans92]

Yes I were able to download and run the model on voc20 dataset. Even I am able to run the model on new dataset when I also update the image extractor. However, if I keep the image extractor fixed through --freeze_backbone parameter then this error occurs.

[jacklanchantin]

Can you try running without a GPU?

…

On Wed, Feb 15, 2023 at 3:30 AM ans92 ***@***.***> wrote: Yes I were able to download and run the model on voc20 dataset. Even I am able to run the model on new dataset when I also update the image extractor. However, if I keep the image extractor fixed through --freeze_backbone parameter then this error occurs. — Reply to this email directly, view it on GitHub <#20 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABJFX7GCIUVFG6VCOIHTVQ3WXSHZPANCNFSM6AAAAAAURD35DI> . You are receiving this because you commented.Message ID: ***@***.***>

[ans92]

Yes I have run the model with cpu but error still exists.
My torch version is different than listed in requirements.txt file
I have 1.13.0+cu117 (torch) and 0.14.0+cu117 (torchvision)

Do you think it is causing this problem?

[jacklanchantin]

It could be. I can't replicate the error