search

qizekun / ReCon

[ICML 2023] Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
https://arxiv.org/abs/2302.02318
MIT License
120 stars 13 forks source link

Zero-shot classification of ModelNet40 #2

Closed kaimingkuang closed 1 year ago

kaimingkuang commented 1 year ago

Thanks for your amazing work! I was trying to run zero-shot classification of ModelNet40 using the pipeline you provided. However, I only got 61.2% accuracy (66.8% reported in the paper) using the following script:

python main.py \
--config=cfgs/zeroshot/modelnet40.yaml \
--zeroshot \
--exp_name=zeroshot_modelnet \
--ckpts=ckpts/zeroshot_66_78.pth

Am I missing something here? Thanks!

qizekun commented 1 year ago

Have you made any modifications to the config file or model? We retested and achieved 66.7751% zero-shot performance. We upload the log file to https://drive.google.com/file/d/1GgKasd2f-4S_FCuIXj2MtNtsnV8dQu5-/view?usp=share_link