search

SimingYan / IAE

[ICCV 2023] "Implicit Autoencoder for Point-Cloud Self-Supervised Representation Learning"
99 stars 17 forks source link

trained ShapeNet decoder network for pretraining phase #18

Closed JunseongAHN closed 1 year ago

JunseongAHN commented 1 year ago

Hello, thank you for your great job!

I am wondering if you can also release a decoder model that you trained with the ShapeNet dataset. I could find an encoder model $f$ for the pre-training stage, but a decoder model $g$.

By the way, I realized that you released a pre-trained model with a decoder under downstream/segmentation, which I failed to load with dgcnn repo. Could you double-check?

Thank you!

SimingYan commented 1 year ago

Hi,

Thanks for your interest! I believe all the released pretrained models here should include the weight of decoder.

Could you please provide more information on the error messages for the failure cases?

JunseongAHN commented 1 year ago

Hello! Thank you for your fast reply!

You are right. I checked the shared weights again. It included the weight of the decoder, and I successfully loaded the released model for training!

Additionally, it was also my bad, I didn't use copy_parameters(model, pretrained, verbose=True) when loading the model haha...

Thank you so much for your help!:)