Linear evaluation on ModelNet40

[Kracozebr]

Hi, first of all thanks for sharing your code.

I'm trying to reproduce Linear evaluation on ModelNet40

I downloaded the dataset and run the evaluation:

python train_svm.py --encoder=dgcnn_cls --restore_path=./pretrained_models/modelnet40_svm.pt

And get the next result:

Transfer linear SVM accuracy: 92.06%

How to achieve the declared 94.2?

[SimingYan]

Hi,

The result of linear evaluation is 92.06%, which you have already achieved. And 94.2 is the fine-tuned classification result. This is different from linear evaluation.

[Kracozebr]

Ok, thanks for clarification.