Pointnet++ classification practice

PointNet++ Classification practice

 Insert a code chip here 
eval mean loss: 0.482859
eval accuracy: 0.902755
eval avg class acc: 0.880052
airplane:     1.000
bathtub:     0.940
bed:     0.970
bench:     0.800
bookshelf:     0.940
bottle:     0.960
bowl:     0.950
car:     0.990
chair:     0.970
cone:     1.000
cup:     0.850
curtain:     0.900
desk:     0.919
door:     0.850
dresser:     0.744
flower_pot:     0.300
glass_box:     0.960
guitar:     1.000
keyboard:     1.000
lamp:     0.900
laptop:     1.000
mantel:     0.970
monitor:     0.980
night_stand:    0.709
person:     0.900
piano:     0.960
plant:     0.750
radio:     0.800
range_hood:     0.970
sink:     0.800
sofa:     0.970
stairs:     0.950
stool:     0.700
table:     0.730
tent:     0.950
toilet:     1.000
tv_stand:     0.850
vase:     0.770
wardrobe:     0.750
xbox:     0.750