Skip to content

ZonglinL/EquivPerformer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

45 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Command this

python3 -W ignore pccls_run.py --num_points 256 --batch_size 64 --head 8 --num_channels 8 --lr 3e-3 --num_layers 5

This is for reload, make sure to move your favorite model to directory trained

--restore ./trained/pc3d_dgl.pt

Linear selfint

--siend 1x1  
--simid 1x1  

Attentive selfint

--siend att 
--simid att 

use performer adjust max number of random features and antithetic or not

--kernel --num_random 20 --antithetic

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages